Label printing ERP and client job-card portal, built in Surat for Gunatit Label, Ahmedabad: printing, punching and slitting tracked live, with AI and automation add-ons
Looking for label printing ERP software in Surat or Gujarat, or a client portal where your customers place label orders themselves? Yukti AI, a software company in Surat, built exactly that for Gunatit Label, a self-adhesive label printer in Ahmedabad: an order-to-delivery system in which the customer fills a digital job card and sees the approximate dispatch date the moment the order is placed, the office routes the job to printing, punching and slitting (or straight to invoice for ready stock), each production manager sees only their own queue and fills a stage sheet, the ready-to-invoice screen shows the quantities for the bill, dispatch records the DC number and a bill PDF the customer can download, and six reports export to Excel. It was built between October and November 2026, deployed on the client’s own domain with real logins, and invoiced as delivered and handed over on 3 November 2026. There is no AI model inside it; the AI and WhatsApp layer is listed on this page as buildable, with honest labels.
Gunatit Label makes self-adhesive labels and label rolls: pharma, food, cosmetics, home-care, lubricant and agro labels, colour labels, water-bottle stickers, PVC tags and thermal paper rolls, on eighteen configured paper types from chromo and PP white to direct thermal, silver PP and clear PP. In a label plant the order is a specification — paper, colour, product size, core, winding, gap or no gap, rolls and labels per roll — and every stage after it needs that specification again. Yukti AI’s job was to put the job card online for the customer, route it through the three production stages with a stage sheet for each, and bring the quantities back to the bill, so that nobody re-types a label order and nobody phones the office to ask where it is.
Label printing ERP software in Surat – enquire
Yukti AI, a software company in Surat, builds label printing ERP software, client ordering portals, WhatsApp automation and AI add-ons for label printers and converters in Surat, Ahmedabad, the rest of Gujarat and across India. The case study below is a delivered build; this box tells you what you get and how to start.
- What you get: a client portal with a digital job card and reorder, office review and routing, stage sheets for printing, punching and slitting, ready-to-invoice quantities, dispatch with DC and bill PDF, paper-roll stock, a job register, reports to Excel, roles and an audit log, with WhatsApp, Tally and AI added when you are ready.
- Built around your press: your paper types, widths, cores, cylinders, dies, stages, roles and documents, not a generic template.
- From price: a custom ERP from Yukti AI starts from ₹30,000; the final quote for a system of this scope is given after a free workflow audit.
- Where we are: 220, Leonard Square, Yogichowk, Varachha, Surat – 395006, Gujarat.
- Next step: send one recent job card and tell us how the order ran from enquiry to delivery, and we will show what one system would replace.
Contents
- For Surat, Ahmedabad and Gujarat buyers: enquire, label ERP in Surat and Gujarat, ERP vs print MIS vs job-card portal, buying checklist and support from Surat
- Quick answer and quick facts
- Inside the system: a walkthrough of the screens
- The business and who this label ERP is for
- The challenge: paper job cards, phone calls and WhatsApp groups
- Before vs after and task by task
- The client job-card portal and a worked example from job card to bill
- Office review and routing (“Send to”)
- Stage sheets for printing, punching and slitting
- Ready to invoice, dispatch and delivery
- Paper-roll stock, the job register and clients, users and roles
- Reports, audit log and notifications and live updates
- AI and automation: built, built elsewhere, buildable, not built and where it helps, department by department
- WhatsApp automation for a label printer
- AI estimation and quoting, AI paper forecasting, an owner’s assistant, what AI cannot do yet, roadmap and AI cost
- Features competitors headline that this build does not have
- Custom ERP vs ready-made print MIS vs Excel and WhatsApp groups
- Features checklist and benefits
- Solution architecture, the modules and roles
- How it was built and what was delivered
- What this page does not claim
- Technology and what can be added next
- Cost and timeline
- FAQ
What is the best ERP for a label printing company that wants customers to order online?
The best ERP for a label printer whose customers reorder the same labels is one that starts from the job card — label name, paper type, colour, product size, core size, winding, with gap or no gap, rolls and labels per roll, artwork — lets the customer fill it in themselves with their own login, shows them a dispatch date at once, and then carries that single specification through office review, the printing, punching and slitting stage sheets, the quantities for the bill, the delivery challan and the bill PDF, so that nothing is re-typed and the customer can see the status without calling. Yukti AI, a software company in Surat, built exactly this for Gunatit Label in Ahmedabad: a client job-card portal with reorder in one tap, per-client order numbering, a “Send to” routing screen, three stage sheets with the press, die and slitter details the floor actually uses, a ready-to-invoice summary, dispatch with a DC number and bill upload, paper-roll stock in running metres, a job register, six reports to Excel, seven roles with row-level security, an audit log and live updates with browser notifications. Ready-made print MIS products suit presses whose workflow matches the package and who want estimation and accounts built in; a custom build is worth it when the way your office routes a job, the way your floor records a stage and the way your customers order are your own.
How can AI and automation help a label printing company?
Records first, then automation, then AI. The Gunatit Label system records every order, stage entry, roll lot, paper issue and dispatch, and it already automates the dispatch date, the rolls-to-quantity arithmetic, the stage hand-over, punching impressions, slitting totals, bill quantities, auto-delivery after dispatch, overdue flags, order numbering, learned die sizes, notifications and live screen updates. WhatsApp order updates, proof approval and payment reminders are the next layer, through the WhatsApp Business API; AI estimation from a job card, AI paper forecasting from the consumption report and an owner’s daily brief are the layer after that. Yukti AI has built AI quoting for a box maker, WhatsApp relays for a yarn manufacturer and its own CRM, and voice AI payment reminders for another client; none of those is inside the Gunatit Label build, and this page labels every item as built here, built in another Yukti AI system, buildable, or not built.
Quick facts
| Client | Gunatit Label, a self-adhesive label printer and label-roll manufacturer in Ahmedabad, Gujarat |
|---|---|
| Builder | Yukti AI, Surat, Gujarat |
| Industry | Label printing and converting (flexo, narrow web): pharma, food, cosmetics, home-care, lubricant, agro and colour labels, water-bottle stickers, PVC tags, thermal paper rolls |
| Products in the system | Eighteen paper types including chromo, chromo HT, removable chromo, PP white, PET white, metallised and PET films, direct thermal, clear PP, silver PP, PVC tag, piggy-back chromo, water-bottle PP and thermal paper roll; seven colour options; fourteen paper widths from 110 to 220 mm; three core sizes; two winding directions |
| Problem | Job cards on paper, the customer phoning for status, the office re-typing the same specification for printing, punching and slitting, and bill quantities collected from the floor by hand |
| Solution | An order-to-delivery label ERP with a client portal: digital job card with artwork upload and reorder, dispatch date at order time, office review and routing, three stage sheets, stage hold and resolve, ready-to-invoice quantities, dispatch with DC and bill PDF, auto-delivered, paper-roll stock, job register, clients, users and roles, six reports to Excel, audit log, notifications, live updates |
| Core mechanics | Expected date from working-day rules (7 days, 3 if urgent; order day and Sundays not counted); route as any subset of printing, punching and slitting in fixed order, or straight to invoice; the next stage becomes Ready when the previous completes; punching impressions from labels divided by die ups; slitting entered as lots of rolls × labels; bill quantities from the last stage that reported output; Delivered two days after dispatch unless recorded earlier |
| Screens | 11 staff modules in the sidebar (Dashboard, All orders, Production board, Stock, Clients, Job register, Master data, Reports & export, Users & roles, Audit log, Settings) plus order detail, “Send to”, a job sheet per stage, My work queue and Notifications; a client portal with Dashboard, My orders, Place new order and Notifications |
| Roles | 7: Administrator, Office staff, Sales, Printing manager, Punching manager, Slitting manager, Client |
| Reports | 6: Order summary, Stage performance, Client-wise, Customer history, Paper consumption, Overdue and on hold — each to .xlsx or CSV |
| Stack | Next.js 16, React 19, TypeScript, Tailwind CSS 4, shadcn/ui, Recharts; Supabase (Postgres with row-level security, Auth, Storage, Realtime, Edge Function) in Mumbai; Vercel hosting deployed by GitHub Actions; service worker for browser notifications |
| Connections | Database, logins, file storage, live updates, browser push and Excel export are real and deployed. WhatsApp Business API, SMS, e-mail notifications, Tally, GST e-invoice and payments are not built; see what this page does not claim |
| Timeline | Built from mid-October 2026; database connected in the last week of October; delivered, deployed on the client’s own domain and invoiced on 3 November 2026; revisions continued under the included support |
| Status | Client build, delivered and handed over to Gunatit Label in November 2026. Daily use, order counts, payment and results are not claimed on this page |
What Gunatit Label’s customers, office and floor work in
The portal needs a login, and it holds the client’s real customers and orders, so this page carries no screenshots. Instead, here is a plain walk through the screens, in the order a label order meets them.
The customer’s dashboard
A customer who signs in lands on “Your orders — where they are now”: a short list of their open orders with the current stage, a strip of “What we make” photo cards (pharma, water bottle, food, cosmetics, home care, lubricant, agro and colour labels), their own paper-wise, colour-wise and size summaries, and the latest updates on their orders. There is one obvious button: Place new order.
Place new order: the digital job card
The form is the job card the press already uses, laid out as fields: label name, paper type (from the eighteen configured), colour, product size, core size, winding, With Gap or No Gap, job colour code, optional transport, rolls and labels per roll, an urgent flag, remarks and an artwork upload. Quantity is worked out from rolls × labels per roll as the customer types. As soon as the order is placed, the screen shows the approximate dispatch date from the working-day rule and the order number from that customer’s own series. A JPG or PNG artwork shows as a thumbnail wherever the order is listed.
My orders and the order timeline
Every order has a timeline the customer can see: placed, reviewed and routed, each stage started and completed, ready to invoice, out for delivery with the DC number, delivered. The customer can add a remark on any order, reorder it in one tap with the artwork carried over, and, once it is dispatched, download the bill PDF the office uploaded.
All orders and the “Send to” screen
Office staff see every order with urgent ones in red and unreviewed ones in orange. Opening a new order shows the job card and a Process screen: tick Printing, Punching and Slitting in any combination (always in that order), or choose straight to invoice for ready stock or trading; set the expected date; and check the stage details that are already pre-filled from the job card, such as the gap type and the die size suggested from the product size.
My work queue and the three job sheets
A printing manager signs in and sees only orders routed through printing, in order of expected date. The printing sheet carries the spec (paper size, running metre, job number, paper type, colour, cylinder teeth, offline or online) and the entry (colour-wise anilox for up to twelve inks, cylinder, rotation, running metre, rolls and labels produced). The punching sheet asks for die size, die number or job number, paper size and type, gap type, die ups and impressions. The slitting sheet asks for label type, ups, core, winding, metre and labels per roll, and takes the output as lots of rolls × labels. Each sheet has Start, Save entry, Report issue and Complete.
Production board and Live floor
The office sees a board of every open order by stage, and a “Live floor” view with a card for the printing press, the die punching machine and the slitter rewinder showing what is on each. Both update live as the managers save their entries.
Ready to invoice, dispatch, delivered
When the last routed stage completes, the order appears on the ready-to-invoice screen with “quantities for the bill”: the lots, total rolls, total labels and paper used in metres, taken from the last stage that reported output. Dispatch asks for a DC number, an optional invoice number and an optional bill PDF; Delivered asks for “received by”, and an order still out for delivery turns Delivered on its own two days after dispatch.
Stock, Job register, Clients, Master data, Reports, Users, Audit log, Settings
The rest of the staff sidebar: paper-roll stock lots with the balance in running metres; the printing job register with date, job number, product, size, cylinder teeth, around and across with gaps, paper and material; the client master with the button that creates the client’s portal login; master data for paper type, colour, paper size, die size, core size and winding with usage counts; six reports with Excel and CSV export; users and a permission matrix; the audit log; and settings for upload limits, allowed file types, notification channels and company details.
In a label plant, the order is a specification that three machines need in three different forms
A self-adhesive label looks like a simple product, and the order behind it is not. A customer who wants labels for a pharma carton, a water bottle, a lubricant can or a bag of seed has to say which paper (chromo, PP white, PET, clear PP, silver, direct thermal, removable, piggy-back), what size, how many colours, which core the roll must sit on, which way it must wind so that it feeds the customer’s applicator, whether the labels are butt-cut or have a gap, how many labels on a roll and how many rolls. Then the artwork. Then the date.
Every stage after that reads the same order differently. The press needs a paper width, a cylinder with the right number of teeth for the repeat, a running metre and an anilox per colour. The punching machine needs a die of the right size, the number of ups across the die and the number of impressions that gives the labels. The slitter needs the core, the winding, the labels per roll and the metre, and gives back lots of rolls. The office needs the rolls and labels that actually came off the slitter for the bill, and a DC number for the transporter. The customer needs one thing: to know where the order is and when it will come.
Gunatit Label runs this flow for pharma, food, cosmetics, home-care, lubricant and agro brands, water-bottle fillers and thermal-roll buyers, with water-bottle stickers and PVC tags following standards of their own. Yukti AI’s job was to put that whole thread — the customer’s job card, the office’s routing, the three stage sheets, the bill quantities, the dispatch and the delivery — into one system that the customer, the office, the sales staff and the three production managers all work in, each seeing only what is theirs.
A label printing ERP has to carry the job card: the same paper, size, core, winding, gap, rolls and labels per roll that the customer typed must pre-fill the printing, punching and slitting sheets and come back as the quantities on the bill. Billing software and WhatsApp groups cannot do that; a job-card portal with stage sheets can.
Who is this label printing ERP and client portal for?
The system was designed around an Ahmedabad label printer that takes repeat orders from brand owners, prints on a flexo press, die-punches, slits into rolls and dispatches by transport, with a small office that reviews every job and three production managers who each own a machine. If you run a self-adhesive label press, a narrow-web converter, a sticker and tag unit, a thermal-roll business or a packaging printer with a similar three-stage floor, the problems on this page will sound familiar. It is also a worked answer to the search for manufacturing ERP software for the printing industry: software built around the job card and the stage sheet, not around a generic item master.
It is a good fit if most of these are true for you:
- Your customers reorder the same label every few weeks and still phone or message the office to place the order and to ask where it is.
- The job card is written by hand, and printing, punching and slitting each copy it again in their own register.
- The office cannot say, without walking to the floor, which orders are on the press, which are waiting for a die and which are stuck.
- The bill is made from numbers collected by phone from the slitter at the end of the day.
- Nobody is sure how many running metres of which paper were printed last month.
- Two or three people need to see everything, and the rest should see only their own work.
The same design suits other print and packaging manufacturers. For a wider label ERP with estimation, artwork gates, plates and dies and QC, see the Jisha Label printing ERP, the sister build in Surat; for corrugated boxes see the corrugated box ERP; for decorative boxes see the Box Factory manufacturing ERP. Yukti AI is in Surat and builds these systems for manufacturers across Surat, Ahmedabad, Gujarat and India; see custom ERP software in Surat.
Why paper job cards, phone calls and WhatsApp groups stop being enough for a label press
These points are our description of the gaps a system of this kind has to close, drawn from how label printers work and from the scope of the build. They are not quotes from the client, and no measurement of the company’s previous way of working is claimed.
- The same specification is written four times. Once by the customer in a message, once by the office on the job card, once by each stage in its own register. Each copy is a chance for the core size or the winding to change.
- The customer has no window into the plant. The only way to know whether an order has been printed is to ask, so the office spends its day answering the same question.
- Dates are promised from memory. Without a rule, two people quote two different delivery dates for the same kind of job, and nobody records when the promise was changed.
- The floor’s numbers live on the floor. Anilox, cylinder, rotation and running metre are useful the next time the same job runs, but only if someone can find the old sheet.
- Impressions and totals are done on a calculator. Labels divided by die ups, lots of rolls added up: small sums, done many times a day, wrong sometimes.
- Bill quantities are collected, not reported. The office asks the slitter how many rolls came off, and the answer depends on who is asked.
- Everyone sees everything or nothing. A shared spreadsheet shows the punching manager every customer’s prices; a paper register shows the owner nothing from home.
- Paper rolls are counted, not metered. A stock of rolls says little when widths and metres per roll differ; the useful unit is running metres per paper type.
How label orders run in the system
Before
- Order received by phone or message, job card written by hand
- Dispatch date promised from experience
- Each stage keeps its own register of the same job
- Customer calls the office for status
- Impressions, roll totals and bill quantities on a calculator
- Delivery challan written separately; bill sent by message
- Paper rolls counted; metres unknown
After (in the ERP)
- Customer fills the digital job card with their own login, or reorders in one tap
- Approximate dispatch date shown at once from the working-day rule
- Office routes the order; stage sheets pre-filled from the job card
- Customer sees the timeline and gets a notification at every step
- Impressions, lot totals and bill quantities worked out by the system
- Dispatch with DC number; bill PDF downloadable by the customer
- Paper-roll lots with the balance in running metres and consumption by paper type
How does a typical label order change, task by task?
The table walks through the real jobs in a label plant and what each looks like in the system. The “before” column is a general description of the manual way of working, not a measurement, and no time savings are claimed.
| Task | Before (manual) | After (in the system) |
|---|---|---|
| Taking an order | Phone call or message, job card written in the office | Customer places it on the digital job card in the portal; quantity from rolls × labels per roll; artwork attached |
| Repeat order | “Same as last time”, with the old card found in a file | Reorder button copies the job card and carries the artwork over |
| Promising a date | From memory | Approximate dispatch date on screen when the order is placed: 7 working days, 3 if urgent, order day and Sundays not counted |
| Numbering the order | Running serial in a register | Per-client series issued by the database, never duplicated |
| Reviewing and routing | Verbal instruction to the floor | “Send to”: printing, punching, slitting in any subset, or straight to invoice; expected date required; stage details pre-filled |
| Changing the job card | Correction on paper, customer not told | Edit while the order is open; the customer sees the change and is notified |
| Printing | Press register | Printing sheet with cylinder teeth, colour-wise anilox, rotation, running metre, rolls and labels produced |
| Punching | Die chosen from memory; impressions on a calculator | Die size suggested from product size and learned from earlier entries; impressions from labels ÷ die ups |
| Slitting | Roll count on a slip | Lots of rolls × labels (up to ten rows); totals worked out; two slitting logins share the stage |
| A problem on the floor | A shout | Report issue puts the stage on hold with a reason; resolve lifts it; the office sees it on the board |
| Knowing where an order is | Walk to the floor | Production board, Live floor cards, My work queue per manager, all updating live |
| Billing | Quantities asked from the slitter | Ready-to-invoice screen shows lots, total rolls, total labels and paper used from the last stage that reported output |
| Dispatching | Challan book | DC number required; optional invoice number and bill PDF; customer downloads the bill |
| Confirming delivery | Nobody records it | Delivered with “received by”, or automatically two days after dispatch |
| Month-end questions | Registers and memory | Six reports with filters, exported to Excel |
Want your customers to place label orders on a job card you designed?Send us one recent job card and tell us how the order ran, and we’ll show you what one system would replace. Free workflow audit, reply the same day.
WhatsApp usThe client job-card portal: online label ordering with a login for every customer
A client job-card portal is a login for each of your customers where they place label orders on your own job card, reorder earlier labels in one tap, see the approximate dispatch date the moment they order, follow the order through printing, punching and slitting on a timeline, add remarks, and download the bill once the order is dispatched. It is not a web-to-print shop with a price calculator and a public catalogue; it is the private order desk of a B2B label printer, open to the customers who already buy from it.
Placing an order
The order form is the press’s own job card. The customer gives the label a name, picks the paper type from the eighteen configured (chromo, chromo HT, removable chromo, Avery chromo, PP white, PET white, metallised and PET films, direct thermal, clear PP, silver, PVC tag, piggy-back chromo, thermal transfer, silver PP, the two water-bottle PP papers and thermal paper roll), the colour, the product size, the core size, the winding and whether the labels are With Gap or No Gap, adds a job colour code and an optional transport, and types the number of rolls and the labels per roll. The quantity is calculated as the customer types. An urgent flag, free remarks and an artwork upload complete the card. Artwork can be a PDF, PNG, JPG, AI, EPS, SVG, CDR or TIF file up to 25 MB; a JPG or PNG shows as a thumbnail wherever the order appears, which is how the office recognises a job at a glance.
The dispatch date, shown at once
The moment the order is placed, the portal shows an approximate dispatch date. The rule is the plant’s: seven working days, or three if the order is marked urgent, with the day of ordering and Sundays not counted. The office can still change the expected date on review, with a reason that is kept in the order’s history, so the customer always sees the current promise and the office always has a record of why it moved. For a customer this replaces the phone call that used to begin “when will my labels come?”
Product rules that save the customer from guessing
Some products have standards, and the portal applies them. For water-bottle labels the customer picks a water-bottle paper and then the bottle size, and the system fills in the label size, the labels per roll, the core, the winding and the standard slitting metre for that bottle; the customer types only the quantity and the rolls are worked out. For PVC tags the size, core and winding are fixed and the customer gives a quantity instead of rolls. Silver PP defaults to its usual labels per roll. These rules live in the system, so a new buyer at a customer cannot order a water-bottle label on the wrong core.
Reorder in one tap
Most label orders are repeats. On any earlier order the customer presses Reorder: the job card is copied, the artwork is carried over, and the customer changes only what is different, usually the rolls. This is where a portal pays for itself in a label business: the brand owner’s purchase clerk orders the pharma label they ordered last month without writing the specification again, and the office receives a complete, correct job card instead of a message.
My orders, the timeline and remarks
My orders lists everything the customer has placed, with its status: Pending review, In production, Ready to invoice, Out for delivery, Delivered or Cancelled. Opening an order shows its timeline, built from the client-visible entries in the audit log: when it was placed, reviewed and routed, when each stage started and completed, when it went out with which DC number, when it was delivered. The customer can leave a remark on any order, and the office sees it. If the office edits the job card while the order is open, the customer sees the change and is notified.
The bill
When the office dispatches an order it can upload the bill as a PDF; the customer downloads it from the order. The bill itself is prepared in the press’s billing software, not in this system — see what this page does not claim — so the portal is where the customer finds it, not where it is made.
Notifications and the phone
The customer has an in-app bell and a Notifications page, and, while the portal is open on a desktop or a phone, browser notifications through a service worker. The portal has a mobile layout, so a buyer can place a reorder and check a status from a phone without an app store. WhatsApp and SMS messages are not sent by this build; they are switches in Settings waiting for a sender, and the WhatsApp section below explains what adding them involves.
How the login is created
The office does not ask customers to sign up. On a client’s card in the Clients module there is a button to create the portal login; the sign-in details appear with copy buttons and a ready-made “Share on WhatsApp” message that opens the ordinary WhatsApp app with the text pre-filled. Forgotten passwords are reset by e-mail through the authentication service.
A worked example: one chromo label order from job card to bill
This example follows an order through the system using the build’s own rules. The order, the customer, the dates and every figure are illustrations chosen to show the arithmetic; they are not Gunatit Label’s orders, standards or business numbers. Where the system calculates a value, the table says so; where a manager types a value, it says that too.
The job card
A customer signs in on Thursday 19 November 2026 and places a new order: label name “Herbal shampoo 200 ml front”, paper CHROMO, four colours, product size 100 × 50 mm, 3-inch core, outside winding, With Gap, 20 rolls × 2,000 labels per roll, artwork as a PDF. Not urgent.
| Step | Working | Result |
|---|---|---|
| Quantity | 20 rolls × 2,000 labels per roll (calculated by the system) | 40,000 labels |
| Approximate dispatch date | 7 working days; Thursday 19 Nov (order day) not counted; Sunday 22 Nov not counted: Fri 20, Sat 21, Mon 23, Tue 24, Wed 25, Thu 26, Fri 27 (calculated) | Friday 27 November 2026 |
| If marked urgent | 3 working days by the same rule: Fri 20, Sat 21, Mon 23 | Monday 23 November 2026 |
| Order number | Next number in this client’s own series, issued by the database (calculated) | e.g. GL/HERB/0007 |
| Overdue if not delivered | Expected date passed, or 10 days after placement (calculated) | From Saturday 28 Nov, or Sunday 29 Nov at the latest |
Office review
The order appears in All orders in orange (unreviewed). Office staff open it, see the thumbnail is missing because the artwork is a PDF, open the PDF, and press Send to. They tick Printing, Punching and Slitting, keep the expected date of 27 November, and check the pre-filled stage details: gap type “With Gap” has already set the punching sheet to gap rather than butt-cut, and a die size of 100 × 50 mm has been suggested from the product size. They save. The order is now In production, the printing stage is Ready, and the customer’s timeline shows “Reviewed and sent to printing, punching and slitting”.
Printing
The printing manager sees the order in My work queue and presses Start. The sheet already shows the spec; the manager picks a paper width of 220 mm from the fourteen configured, a cylinder of 81 teeth, offline printing, and enters the anilox per colour for the four inks, clockwise rotation, and after the run, the output.
| Entry | How it is arrived at | Value |
|---|---|---|
| Labels across the web | The press and die layout (manager’s knowledge, not calculated here) | 4 across |
| Rows of labels needed | 40,000 labels ÷ 4 across (illustrative arithmetic) | 10,000 rows |
| Web needed for labels | 10,000 rows × 53 mm pitch (50 mm label + 3 mm gap) (illustrative) | 530 m |
| Running metre typed on the sheet | 530 m plus make-ready and waste, as the manager records it (typed, not calculated) | 560 m |
| Rolls and labels produced | Typed by the manager | 1 jumbo roll, 40,000 labels |
The manager presses Complete. The printing stage is Completed, the punching stage becomes Ready, and the customer’s timeline shows “Printing completed”. If the manager had recorded the paper used on a stock lot, the stock balance would reduce by 560 running metres.
Punching
The punching manager starts the stage. Die size 100 × 50 mm is suggested; the manager confirms it and types the die number. With Gap is pre-filled. Die ups: 4. The system fills in impressions.
| Entry | Working | Value |
|---|---|---|
| Die ups | Typed by the manager | 4 |
| Impressions | 40,000 labels ÷ 4 die ups (calculated by the system) | 10,000 impressions |
| Die size | Confirmed from the suggestion; a new size would be learned into master data for next time | 100 × 50 mm |
Slitting
The slitting manager (either of the two slitting logins) starts the stage. Core 3 inch, outside winding and 2,000 labels per roll are pre-filled from the job card. The output is entered as lots.
| Lot | Rolls × labels per roll (typed) | Labels (calculated) |
|---|---|---|
| Lot 1 | 12 × 2,000 | 24,000 |
| Lot 2 | 8 × 2,000 | 16,000 |
| Total | 20 rolls | 40,000 labels |
Suppose the slitter had come up short and the second lot were 7 rolls: the total would read 19 rolls and 38,000 labels, and that is what the office would see on the ready-to-invoice screen, because the bill quantities come from the last stage that reported output, not from the job card.
Ready to invoice, dispatch, delivered
With slitting completed the order is Ready to invoice. The screen shows the quantities for the bill: 2 lots, 20 rolls, 40,000 labels, 560 m of paper used. The office makes the bill in its billing software, uploads the PDF, types the DC number and dispatches. The customer is notified, sees “Out for delivery” with the DC number, and downloads the bill. The transporter delivers on Saturday 28 November; if the office records “received by” the order is Delivered then, and if nobody records anything it becomes Delivered on its own on Monday 30 November, two days after dispatch.
What the reports now hold
Order summary shows one more delivered order for this client. Stage performance adds this job’s wait and run time to each of the three stages. Customer history lists it under the client with its rolls and labels. Paper consumption adds 560 printed running metres to CHROMO for the month. The audit log has every save, and the customer’s timeline is the client-visible subset of it. Next month the customer presses Reorder, and the whole card, artwork included, is ready to send again.
Office review and routing: the “Send to” screen
Every order placed by a customer, by sales or by the office itself starts as Pending review, in orange on every list, dashboard, queue and board until someone in the office has looked at it. Review is one screen, called Process or “Send to”, and it does four things.
Choose the route
Printing, Punching and Slitting are three tick boxes, and any subset can be chosen; the order is always the same, printing before punching before slitting, because that is how a label is made. A plain die-cut label on pre-printed stock might skip printing; a printed roll sold unslit might skip slitting. There is a fourth choice, straight to invoice, for ready stock and trading items such as thermal paper rolls bought in and sold on, so that an order with no production does not have to pretend it had any.
Set the expected date
The expected date is required. It starts from the working-day rule the customer already saw and can be moved; every change asks for a reason and keeps the history, so the dashboard’s overdue count is measured against a date somebody chose on purpose.
Check the pre-filled stage details
The stage sheets are pre-filled from the job card so that the managers are not typing the specification again. The gap type on the job card becomes butt-cut or gap on the punching sheet; the product size suggests a die size; the core, winding and labels per roll go to the slitting sheet; the paper type and colour go to the printing sheet. The office can correct any of these before sending.
Edit, cancel, reopen
While an order is open the office can edit the job card; the customer sees the new version and is notified, so a correction made on the phone does not stay a secret. An order can be cancelled with a reason and reopened with a reason. An administrator can skip the remaining stages with a reason and force an order to Ready to invoice, and can correct a stage entry even after the stage is completed, which is what happens when the slitter reports a wrong roll count after the bill has been started.
Order numbers
Each client has its own order series, such as GL/CODE/0001, where the code is suggested from the client’s name when the client is created. The number is issued by a database function the moment the order is saved, atomically, so two people placing orders at the same second cannot get the same number. Stock lots have their own series.
Stage sheets for printing, punching and slitting, and who sees what
The stage sheet is the heart of the floor side of the system. Each of the three stages has a sheet with two halves: the specification, pre-filled from the job card and editable by the office, and the entry, filled by the manager who runs the machine. The fields are the ones that were on the paper registers, so the managers recognise them.
Printing sheet
Specification: paper size from fourteen configured widths between 110 and 220 mm, running metre, job number, paper type, colour, cylinder from seven sizes (64, 72, 78, 81, 96, 99 and 110 teeth), offline or online. Entry: anilox per colour, chosen from nine values between 250 and 1,200 for up to twelve inks, cylinder, rotation (clockwise or anti-clockwise), running metre, and the rolls and labels produced. The next time the same label runs, the previous sheet is in the order history, which is what a press register was for.
Punching sheet
Specification: die size, typed by hand but suggested from the product size and from earlier entries, which are learned into master data; an optional die number or job number; paper size and paper type; No Gap (butt-cut) or With Gap. Entry: die ups and impressions, where impressions are filled in as total labels divided by die ups and can be overwritten if the run differed.
Slitting sheet
Specification: label type, ups (1 to 4, or other), core size from three, winding from two, metre and labels per roll. Entry: up to ten rows of rolls × labels per roll, from which the total rolls and total labels are worked out. Two slitting-manager logins share the one stage, so either operator can pick up a job. The slitting totals are what the office bills, unless the route ends earlier.
Stage control
A stage is Waiting until the one before it completes, then Ready. The manager presses Start (In progress), saves the entry as often as needed, and presses Complete, which makes the next stage Ready and writes a client-visible timeline entry. Report issue puts the stage On hold with a reason, which the office sees on the board and in the Overdue and on hold report; Resolve lifts it. A stage can also be Cancelled or Skipped by an administrator with a reason. That is seven stage statuses, and the order’s own status (six of them) follows from the stages.
My work queue: each manager sees only their stage
The printing manager’s login shows the printing queue, the punching manager’s the punching queue, the slitting managers’ the slitting queue: only orders routed through that stage, in order of expected date, with urgent orders in red. This is not just a filter on the screen; the database’s row-level security returns a manager only the orders routed through their stage, so there is nothing to find by changing a web address. Sales staff see only the orders they entered; clients see only their own orders and files; office and administrators see everything.
Production board and Live floor
For the office and the owner there is a board of all open orders by stage and status, and a “Live floor” page with a card for the printing press, the die punching machine and the slitter rewinder, showing what is running, what is ready and what is on hold. Both pages update as entries are saved, through the database’s realtime channel, without a refresh.
The system records what the managers type. It does not read the press, count impressions from a sensor or measure metres from the slitter. Machine data would need hardware on the machines, which is listed under what AI cannot do yet.
Ready to invoice, dispatch and delivery
The last routed stage completing moves the order to Ready to invoice; so does the office choosing straight to invoice at review, or an administrator skipping the remaining stages. The ready-to-invoice screen is the hand-over from the floor to the office, and its job is to show the quantities for the bill without anyone asking anyone.
Quantities for the bill
The screen shows the lots, the total rolls, the total labels and the paper used in metres, taken from the last stage that reported output. For a slit job that is the slitting lots; for a job that ended at printing it is the printing output. If the floor produced less than the job card asked for, the bill shows what was produced. This single rule is what stops a bill being raised for 20 rolls when 19 came off the slitter.
Dispatch
Out for delivery needs a delivery challan number; an invoice number is optional, and so is a bill PDF, which is uploaded against the order and becomes downloadable by the customer. The GST bill itself is prepared outside the system; this build does not generate invoices, e-invoices or e-way bills, and that is covered honestly under features this build does not have and in the Tally integration page.
Delivered, by hand or by the clock
Delivered is recorded with “received by”. If nobody records it, an order that is Out for delivery becomes Delivered automatically two days after dispatch, a rule that runs while an administrator or office screen is open. The point is that the list of open orders empties itself of jobs that are plainly done, so the overdue count stays honest.
Overdue and highlight rules
An order is overdue when its expected date has passed, or ten days after it was placed if it has not been delivered, whichever comes first. Overdue orders are counted on the dashboard and listed in the Overdue and on hold report. Urgent orders are red and unreviewed orders are orange in every list, queue and board, so the two things that most need a person’s attention are visible from across the room.
The owner’s dashboard
The administrator’s dashboard opens with hero tiles for open orders, orders this year, delivered orders and clients; each tile is a link to the filtered list behind it. Below are charts of orders by stage and by month, and the overdue and on-hold counts. It is a reading of the records, not a separate report somebody maintains.
Paper-roll stock in running metres
A label press buys paper as rolls of a given width and length, and uses it by the running metre. The Stock module records stock-in as lots: supplier, challan number, paper type, paper width, number of rolls and metres per roll (optional, for paper bought without a length). Each lot gets a number in its own series. The stock list shows the balance per paper type and width in running metres as well as rolls.
Paper used is recorded against a lot by the printing or slitting manager on the job sheet, so the consumption of a job is tied to the job and the balance falls as the floor reports it. Returns and adjustments are separate entries with reasons, and every movement is in the audit log. The Paper consumption report then shows printed running metres per paper type for any period, which is the number a buyer needs when ordering the next lot of chromo or PP.
What this is not: a purchase module with purchase orders, goods receipt notes and supplier bills, or a reorder-level alert. Those are listed under what can be added next.
The printing job register: cylinder teeth, around and across, paper and material
Every label press keeps a job sheet: for each job that has ever run, the product, its size, the cylinder teeth, the number of labels around with the gap, the number across with the gap, the paper and the material. It is the press’s memory, and it is how the next run of the same label is set up without starting from scratch. In the system this is the Job register module: date, job number, product, size, teeth, around and gap, across and gap, paper and material.
The client’s existing job register was loaded into the system at handover as part of the delivery, so the press did not start with an empty memory. The register is restricted to administrators by database policy, because it contains customers’ product names and the plant’s setups. Its contents are the client’s business data and are not published here.
Master data sits next to it: paper types, colours, paper sizes, die sizes (which grow as punching entries are made), core sizes and windings, each with a count of how many orders use it and a deactivate switch rather than a delete, so an old paper type disappears from the order form without breaking the orders that used it.
Clients, users, roles and the permission matrix
Clients
The client master holds the code (used in the order series), contact, city, GSTIN and minimum order. From the client’s card the office creates the portal login, copies the sign-in details or shares them through a pre-filled WhatsApp message from the ordinary WhatsApp app. A client’s orders, files and history are one click away from the card, and the Client-wise and Customer history reports read from the same records.
Seven roles
Administrator, Office staff, Sales, Printing manager, Punching manager, Slitting manager and Client. The Users & roles module shows the permission matrix on screen, so the owner can see which role can place, review, route, start, hold, complete, dispatch, deliver, edit masters, read reports or change settings. Logins are created and passwords set by an administrator or office user through a server function that first checks the caller is an active administrator or office user; forgotten passwords go by e-mail.
Enforced in the database, not only on the screen
Row-level security in the database carries the same rules: clients see only their own orders and files; each production manager sees only orders routed through their stage; sales staff see only the orders they entered; the anonymous role is locked out entirely; artwork lives in a private storage bucket keyed by client and is served through short-lived signed links. This is the difference between a portal that looks private and one that is.
Six reports, exported to Excel, and an audit log of everything
Reports read the same records as the screens, so the figure on a report is the figure on the order. Each has a date range and filters, and each downloads as a real .xlsx workbook, with a bold frozen header row and filters already on, or as CSV.
| Report | What it answers |
|---|---|
| Order summary | Every order in a period, filtered by client, status, stage or paper type, with rolls and labels |
| Stage performance | Per stage: jobs completed, average wait before start, average run time, active jobs, jobs on hold and issues reported |
| Client-wise | Orders, rolls and labels by client for the period |
| Customer history | One customer’s totals and every order they have placed, for a review meeting or a credit decision |
| Paper consumption | Printed running metres by paper type, for buying the next lot |
| Overdue and on hold | What is late, what is stuck and why, from the expected dates and the hold reasons |
The audit log
Every action that goes through the service layer, from placing an order to changing a setting, writes an audit entry with who, what and when. The entries marked client-visible form the order timeline the customer sees, so the customer’s view and the owner’s view are the same facts at two levels of detail. Admin corrections to completed stage entries are logged like everything else.
Notifications, browser push and live screens
Notifications are addressed to a role, a user or a client: the office is told when a customer places an order or adds a remark; a manager is told when a stage becomes Ready for them; the customer is told when the order is reviewed, when each stage completes, when it is dispatched and when it is delivered. They appear in the in-app bell and on a Notifications page. While the portal is open in a browser, on a desktop or on a phone, a service worker shows them as browser notifications, so a manager on the floor sees a new job without watching the screen.
Separately from notifications, every open screen updates live. The office’s production board changes when the slitting manager presses Complete; the customer’s timeline changes when the office dispatches. This comes from the database’s realtime channel, not from polling.
Settings holds switches for e-mail, SMS, WhatsApp and Android push channels, stored on each notification record. In this build they are switches only: there is no e-mail, SMS or WhatsApp sender behind them, and no Android app. Adding a sender is the WhatsApp automation layer described next, and the WhatsApp message charges are paid to Meta separately, as explained in WhatsApp Business API pricing in India.
Are your customers still phoning the office to ask where their labels are?We’ll show you how a job-card portal, three stage sheets and a ready-to-invoice screen would answer that for them. Free workflow audit over WhatsApp or at our Surat office.
WhatsApp usAI and automation for a label printing company: what is built here, what exists elsewhere, what is buildable
Honesty first: the system built for Gunatit Label does not contain a language model or any machine-learning feature. A search of its code for AI providers finds nothing, because nothing was asked for. What it has is a layer of automation: rules that do work people used to do by hand. Every AI item on this page is labelled Built here, Built in another Yukti AI system (with a link), Buildable or Not built, and nothing buildable is described as delivered.
What runs by itself in this build
- The approximate dispatch date when the customer orders, and the expected date at review, from the working-day rule.
- Quantity from rolls × labels per roll; rolls from quantity for water-bottle labels and PVC tags; product standards filled in from the bottle size.
- Stage sheets pre-filled from the job card: gap type, suggested die size, core, winding, labels per roll, paper and colour.
- The next stage made Ready when the previous one completes; the order status following the stages; Ready to invoice when the route ends.
- Punching impressions from labels divided by die ups; slitting totals from lots; bill quantities from the last stage that reported output.
- Delivered two days after dispatch; overdue after the expected date or ten days; urgent and unreviewed highlighting everywhere.
- Per-client order numbers and stock lot numbers issued by the database; die sizes learned into master data from punching entries.
- A notification to the right role, user or client on every step, browser push while the portal is open, and live updates on every open screen.
Where AI fits a label press
The most useful AI for a label printer works on the records this system already keeps. A customer’s WhatsApp message such as “same shampoo label, 30 rolls, by Monday” can be read into a draft reorder for the office to confirm. A job card can be turned into a draft estimate once the press has a rate structure in the system. The paper consumption report can become a forecast of running metres per paper type for next month. An owner can get a morning brief: what is late, what is on hold, which customers have gone quiet. These are buildable by Yukti AI on this system; some of them already run in other Yukti AI builds, which are linked below so you can see the real thing rather than a promise. See AI ERP for how we add an AI layer to an ERP and ERP-connected AI agents for what an agent that reads your records can do.
Where do AI and automation help a label printing business?
Status labels: Built here (in the Gunatit Label system), Built in another Yukti AI system (exists in a different Yukti AI build, linked), Buildable (we can add it on your records), Needs hardware (needs sensors or cameras on the machines, not only software).
| Department | Repetitive work today | AI or automation | Status |
|---|---|---|---|
| Customer’s purchase desk | Writing the label specification in a message every time | Job-card portal with reorder and artwork carried over | Built here |
| Customer’s purchase desk | Asking for the delivery date | Approximate dispatch date at order time; expected date with change history | Built here |
| Office | Reading a WhatsApp message into a job card | AI reads the message into a draft order for the office to confirm | Built in another Yukti AI system (Filyarn WhatsApp inbox, Arjun WhatsApp sales agent); buildable here |
| Office | Telling the customer the order moved | In-app and browser notifications on every step | Built here |
| Office | Sending the same update on WhatsApp | WhatsApp Business API messages from order events | Built in another Yukti AI system (Yukti AI CRM); buildable here |
| Office | Copying the job card to each stage | Stage sheets pre-filled from the job card | Built here |
| Office | Pricing a new label | Rate structure plus an AI draft estimate from the job card for a person to approve | Built in another Yukti AI system (Box Factory AI quotation engine); not built here; buildable |
| Printing | Finding last time’s anilox and cylinder | Previous stage sheet in the order history; job register | Built here |
| Printing | Suggesting press settings from past runs of the same label | Recommendation from earlier sheets of the same job | Buildable |
| Punching | Impressions on a calculator; die chosen from memory | Impressions from labels ÷ ups; die size suggested and learned | Built here |
| Punching | Knowing which die is worn or where it is | Plate and die register with life, location and condition | Not built; buildable |
| Slitting | Adding up lots for the bill | Totals from lots; bill quantities on the ready-to-invoice screen | Built here |
| Slitting | Tracing which jumbo a roll came from | QR roll traceability, parent to child rolls | Built in another Yukti AI system (Jisha Label ERP); not built here; buildable |
| Stores | Working out how much paper to order | Forecast of running metres per paper type from consumption and lead time | Buildable (consumption report is built here) |
| Dispatch | Writing the DC, sending the bill | DC number on dispatch; bill PDF downloaded by the customer | Built here |
| Dispatch | GST invoice, e-invoice, e-way bill | Invoice generation and GST provider connection | Not built; buildable |
| Accounts | Chasing payments | WhatsApp reminders and voice AI calls from the due date | Built in another Yukti AI system (Sureja voice AI); not built here; buildable once invoices are in the system |
| Accounts | Posting to Tally | Tally connection | Not built; buildable |
| Owner | Finding out what is late | Overdue flags, dashboard tiles, Overdue and on hold report | Built here |
| Owner | Reading the day’s position | AI morning brief from the records | Built in another Yukti AI system (P. Hitesh owner copilot); not built here; buildable |
| Quality | Catching print defects on the web | Camera-based inspection | Needs hardware |
| Maintenance | Knowing a press part is failing | Predictive maintenance from sensor data | Needs hardware |
WhatsApp automation for a label printing company: order updates, proof approval and reminders
WhatsApp automation for a label printer means the system sends the right message when something happens to an order: order received with the dispatch date, artwork proof sent for approval, printing completed, out for delivery with the DC number and bill, delivered, and a reminder when a payment is due. It runs on the WhatsApp Business API with approved message templates, and the message charges are paid to Meta. It is not built in the Gunatit Label system; the notification events it would hang on are, which is most of the work.
What exists today in this build is honest and limited: an in-app bell, browser push while the portal is open, and a “Share on WhatsApp” button for sending a new customer their login from the ordinary WhatsApp app. The Settings module has a WhatsApp switch per notification, stored on every notification record, with no sender behind it. The client’s invoice records WhatsApp and SMS as options to be opted into later, and that is how the page treats them.
What the WhatsApp layer would send
- Order confirmation with the order number and the approximate dispatch date, the moment the order is placed.
- Date change when the office moves the expected date, with the reason.
- Stage updates, or just the ones the customer cares about: printed, slit, ready.
- Dispatch with the DC number, transport and the bill PDF attached.
- Delivered, asking for confirmation, which would record “received by” without the office doing it.
- Reorder nudge when a customer who orders every few weeks has gone quiet, with a link to the Reorder button.
- Payment reminder from the due date, once bills and due dates are in the system.
- Inbound: a customer’s message read into a draft order for the office to confirm, which is the AI part.
Proof approval on WhatsApp
Many label buyers approve artwork on WhatsApp with a thumbs-up, which is unrecoverable later. A proof-approval step would send the artwork, record the approval or the change request with a timestamp against the order, and hold printing until approval. The Gunatit build uploads and previews artwork but has no versioned approval gate; the Jisha Label ERP has an artwork gate before plates, which is the pattern Yukti AI would reuse.
What it costs and what it needs
A WhatsApp Business number, verified business details, approved templates and a provider connection; the development to connect order events to templates; and the per-message charges, which depend on the category of message and are paid to Meta. See WhatsApp automation for what Yukti AI builds, WhatsApp Business API pricing in India for the charges, and WhatsApp AI chatbot in Surat for the inbound side. Automated business messages go through the API, not through the ordinary WhatsApp app on a phone, which is why this is a separate layer from the “Share on WhatsApp” button that is already there.
AI estimation and quoting for label printing: what is realistic
AI can draft a label quotation, safely, only when it sits on top of a proper rate structure. The AI reads the enquiry (paper, size, colours, quantity, finishing) into a job card and asks for what is missing; a deterministic calculation works out paper by the running metre, ink, plates or dies, machine time and margin; a person approves before the customer sees a price. Neither the estimator nor the AI is in the Gunatit Label build: it was scoped as an order-to-delivery system, and the bill is made outside it. Yukti AI built an AI quotation engine for The Box Factory, where a free-text requirement becomes a bill of materials and a cost, and an estimator with job costing in the Jisha Label ERP; both patterns transfer to a label press.
For a label job the arithmetic the engine would hold is well known: labels across and around from the die and the cylinder repeat; running metres from the quantity, the pitch and the ups, plus make-ready; paper cost from metres and width at the rate per square metre or per roll; ink by colour and coverage; plates or cylinders amortised over the quantity; die charge if new; slitting and packing; margin; GST added on top. The job card the Gunatit customer fills in already carries most of the inputs, which is why estimation is the natural next module, not a different system.
Where AI adds something a calculator does not: reading an untidy enquiry, spotting that a 3-inch core was not mentioned, suggesting the paper the customer used last time, and comparing the drafted price with what the same customer paid before. Where it must not act alone: confirming a price. Every Yukti AI estimation build keeps a person between the draft and the customer.
AI paper forecasting and production insight for a label press
Some ready-made print ERPs now advertise AI inventory forecasting: predicting material needs from past consumption and supplier lead times. The honest version of that for a small label press is modest and useful. The Gunatit build records paper used per job and reports printed running metres per paper type per period. A forecasting layer would take those series, the open orders and their expected dates, and the typical lead time per supplier, and say: at the current rate the 220 mm chromo balance runs out in about this many days, and the next lot should be ordered by this date. That is buildable, and it needs a few months of real consumption records first, which is why the records come before the AI.
The same records support two more insights. Stage performance already gives average wait and run time per stage; a model can flag a job whose run time is far outside the normal band for that paper and size, which is usually a problem worth asking about. And order history per customer gives a reorder rhythm, so a customer who usually orders every five weeks and has not ordered in eight can be surfaced to sales, with the Reorder link ready. None of this is in the delivered system; all of it reads what the delivered system records.
An AI assistant for the label press owner
The owner of a label press wants a few answers every morning: what is overdue, what is on hold and why, what is going out today, which customers placed orders yesterday, which paper is low, which customer has gone quiet. The dashboard and the Overdue and on hold report answer the first three today. An AI assistant would put all six into a short brief on WhatsApp or in the app each morning, in plain English or Gujarati, and answer follow-up questions such as “how many rolls of PP white did we print for this client this year?” by reading the same records that the Customer history and Paper consumption reports read. Yukti AI built an owner’s copilot of this kind on a diamond assorting ERP; the pattern is the same on a label ERP, and it is buildable here. It is not built here.
What AI cannot do for a label press (yet), and what needs hardware
- It cannot count what nobody records. If the printing manager does not type the running metre, there is no consumption to forecast. The stage sheets exist to make the recording quick, and the AI depends on them.
- It cannot inspect print quality from the ERP. Registration, colour and missing-label detection on the web are camera systems installed on the press, sold by inspection vendors; an ERP can record the inspection result, not perform it.
- It cannot predict a press breakdown without sensors. Predictive maintenance needs vibration, temperature or motor data from the machine. The system can log breakdowns and downtime by hand, and that log is a start.
- It should not confirm prices or dates on its own. Drafts, suggestions and reminders, yes; commitments to a customer, a person.
- It cannot replace the job register’s judgement. Which cylinder suits which repeat is press knowledge; the system can suggest from the last run, and the manager decides.
- Your data should stay yours. In Yukti AI builds the AI receives only the text it needs, usage runs on your own provider account, and the records never leave your database for training.
How to add AI and automation to a label press, step by step
Records
Job card, routing, stage sheets, lots, paper used, dispatch, delivery: the order-to-delivery system. This is what Gunatit Label has.
Rules
Dates, numbering, impressions, totals, bill quantities, auto-delivery, overdue flags, notifications, live updates. Also built.
Business API messages from the same events: confirmation, dispatch with bill, delivered, reorder nudge. Buildable.
Billing and books
Invoices in the system or a Tally connection, so payment reminders have a due date to start from. Buildable.
AI reading
Messages into draft orders, job cards into draft estimates, with a person approving. Buildable, proven elsewhere.
AI insight
Paper forecasting, run-time outliers, reorder rhythm, the owner’s morning brief. Buildable once a few months of records exist.
How much does AI automation cost for a label printing company?
It depends on the layer you start from. A custom ERP from Yukti AI starts from ₹30,000, a WhatsApp chatbot from ₹2,000, and workflow automation is priced by the requirement; the general drivers are on AI automation cost in India. Two running costs sit outside any build and are paid to the providers: WhatsApp message charges to Meta and AI usage to the model provider, both on your own accounts. For a label press the sensible sequence is the one in the roadmap above, because each layer makes the next cheaper: the order events are already there for WhatsApp to use, and the consumption records are already there for the forecast to read. Yukti AI quotes each layer after a free workflow audit, not from a price list, and the comparison with ready-made products is in the comparison section.
Features print MIS and label ERP vendors headline that this build does not have, and whether they can be added
The Gunatit Label system was deliberately scoped as an order-to-delivery system with a client portal. It does not have estimation and quotations, job costing or waste analysis, machine scheduling, a plate, die or cylinder register, a QC module, purchase orders, GST invoicing or accounts, barcode or QR roll traceability, WhatsApp sending, credit limits and payment tracking, an online price calculator, versioned artwork approval or AI forecasting. Each of these is listed below with its status, because a buyer comparing this build with a packaged print MIS should see the gaps as clearly as the strengths.
| Feature vendors headline | What it does | In the Gunatit Label build | Can it be added? |
|---|---|---|---|
| Estimation and quotations | Price a job from paper, ink, plates, dies, machine time and margin; send a quotation | Not built; the bill is made outside the system | Yes; the estimator pattern exists in the Jisha Label ERP |
| Job costing and waste analysis | Estimate versus actual per job; wastage by stage | Not built; paper used per job is recorded, cost is not | Yes, once rates and estimation exist |
| Scheduling and capacity planning | Machine-wise plan by shift, make-ready, downtime | Not built; queues are by expected date | Yes |
| Plate, die and cylinder register | Life, location and condition of tooling | Not built as a module; die sizes are learned, the job register holds cylinder details | Yes |
| Quality control | Inspection records, a gate before dispatch | Not built; a stage can be put on hold with a reason | Yes; the QC gate exists in the Jisha build |
| Purchase orders, GRN, supplier bills | Buying paper and consumables through the system | Not built; stock-in lots record supplier and challan | Yes |
| GST invoicing, e-invoice, e-way bill | Generate the tax invoice and file it | Not built; a bill PDF made elsewhere is uploaded | Yes, through an approved GST provider at go-live |
| Accounts and Tally | Receivables, payables, ledgers, posting to Tally | Not built; the client’s invoice records Tally as quoted separately | Yes; see Tally integration |
| Barcode or QR roll traceability | Each slit roll traced to its jumbo and job | Not built; lots are recorded as counts | Yes; built in the Jisha ERP |
| WhatsApp sending | Automated order messages through the Business API | Not built; switches only, plus a wa.me share for logins | Yes; see above |
| Credit limits and payment tracking | Outstanding per customer, reminders | Not built; the client master holds a minimum order, not a credit limit | Yes, with invoices in the system |
| Online price calculator for the portal | Customer sees a price while ordering | Not built; the portal takes the order, not the price | Yes, after estimation |
| Versioned artwork approval | Proof versions, timestamped sign-off, printing held until approval | Not built; artwork is uploaded and previewed | Yes; artwork gate pattern in the Jisha ERP |
| AI inventory forecasting | Material needs predicted from consumption and lead times | Not built; consumption report exists | Yes, after a few months of records |
| Mobile app, SMS, e-mail notifications | Native app push, SMS gateway, e-mail on events | Not built; mobile web layout and browser push exist; e-mail only for password reset | Yes |
And the other side of the ledger, the things this build has that most of those products do not market: a per-customer job-card portal with the dispatch date shown at order time, reorder with artwork carried over, two-day auto-delivery, per-client order series, role-locked queues enforced in the database, browser push, live screens, Excel export, and a delivery in about three weeks on the client’s own domain.
Custom label ERP with a client portal vs ready-made print MIS vs Excel and WhatsApp groups
There are three realistic choices for a small or mid-sized label printer in India. This is a general comparison, not a review of any product, and the “ready-made” column describes packaged print MIS and label ERP products as a group.
| Question | Excel + WhatsApp groups | Ready-made print MIS | Custom build (as built here) |
|---|---|---|---|
| How customers order | Message or call; office re-types | Usually office entry; a portal in some products, often an extra module | Customer logins on your own job card, with reorder |
| Dispatch date | Promised from memory | From scheduling, if configured | Your working-day rule, shown at order time, changes logged |
| Stage sheets | Paper registers | Generic production entry, configurable within limits | Your fields: cylinder teeth, anilox per colour, die ups, lots of rolls × labels |
| Who sees what | Everyone in the group sees everything | Role-based access, usually | Seven roles enforced in the database; each manager only their queue |
| Estimation and quotes | A sheet | Built in, often the main feature | Not in this build; added when needed |
| GST billing and accounts | Tally or a billing app | Built in or via Tally | Bill PDF uploaded; invoicing or Tally link scoped separately |
| WhatsApp updates | By hand, from a phone | Some products, with message quotas | Buildable on the existing events; charges to Meta |
| Fits your floor | Only as well as the person running it | Partly; you adapt to the package | Yes: your stages, your fields, your rules |
| Cost model | Cheap, costly in mistakes | Licence or subscription, per user or per plan | One-time build with support; from ₹30,000 |
| Time to start | Immediate | Implementation and training weeks | This build: about three weeks from first code to handover |
| Data | Scattered, in chats | Vendor’s database; export varies | Your own database; Excel export of every report |
For the general argument see custom ERP vs off-the-shelf ERP and moving from Excel and Tally to one system; for the difference between an ERP, a print MIS and a job-card portal, see the plain explanation below.
Features of label printing ERP software, and which ones this build covers
Label ERP and print MIS products are usually described in the same groups: order intake and customer portal, production and job tracking, materials and tooling, dispatch and billing, reports and administration, and integrations. This checklist uses those groups so you can compare any product, including a custom build, feature by feature. “Covered” means it exists in the Gunatit Label system; “Can be added” means it is not built and would be scoped with the client.
Order intake and customer portal
| Feature | What it does | In this build |
|---|---|---|
| Customer logins | Each customer places and follows orders with their own login | Covered |
| Digital job card | Paper, colour, size, core, winding, gap, rolls × labels, urgent, remarks, artwork | Covered |
| Product standards | Water-bottle and PVC-tag rules fill in size, core, winding, labels per roll and metre | Covered |
| Reorder | Copy an earlier order with its artwork | Covered |
| Dispatch date at order time | Working-day rule, urgent variant | Covered |
| Order timeline and remarks | Client-visible history; customer remarks | Covered |
| Artwork upload and preview | Eight file types up to 25 MB; image thumbnails | Covered |
| Versioned proof approval | Approval gate with timestamp before printing | Can be added |
| Online price calculator | Price shown while ordering | Can be added |
| Sales-entered orders | Sales staff place orders for customers, see only their own | Covered |
Production and job tracking
| Feature | What it does | In this build |
|---|---|---|
| Routing | Printing, punching, slitting in any subset; straight to invoice | Covered |
| Stage sheets | Printing (cylinder, anilox, rotation, metre), punching (die, ups, impressions), slitting (lots) | Covered |
| Stage statuses, hold and resolve | Seven statuses; report issue with reason | Covered |
| Per-manager work queue | Only the manager’s own stage, enforced in the database | Covered |
| Production board and live floor | All open orders by stage; machine cards; live updates | Covered |
| Admin corrections | Correct an entry after completion; skip stages with reason | Covered |
| Machine scheduling | Shift-wise plan, make-ready, downtime | Can be added |
| Quality control module | Inspection records, QC gate | Can be added |
| Waste analysis | Wastage by stage and job | Can be added |
Materials and tooling
| Feature | What it does | In this build |
|---|---|---|
| Paper-roll stock lots | Supplier, challan, type, width, rolls, metres per roll; balance in running metres | Covered |
| Paper used per job | Recorded on the job sheet against a lot | Covered |
| Returns and adjustments | With reasons, in the audit log | Covered |
| Job register | Cylinder teeth, around and across with gaps, paper, material; loaded from the existing sheet | Covered |
| Master data | Paper types, colours, widths, die sizes (learned), cores, windings, with usage counts | Covered |
| Purchase orders and GRN | Buying through the system | Can be added |
| Plate, die and cylinder register | Tooling life, location, condition | Can be added |
| QR roll traceability | Each roll traced to its jumbo and job | Can be added |
| Reorder-level alerts | Warn when a paper type is low | Can be added |
Dispatch and billing
| Feature | What it does | In this build |
|---|---|---|
| Ready-to-invoice quantities | Lots, rolls, labels, paper used from the last output stage | Covered |
| Dispatch with DC number | Required DC, optional invoice number | Covered |
| Bill PDF for the customer | Uploaded on dispatch, downloaded in the portal | Covered |
| Delivered with received-by; auto-delivered | Recorded, or automatic two days after dispatch | Covered |
| Overdue rules | Expected date passed, or ten days after placement | Covered |
| GST invoice generation | Tax invoice from the order | Can be added |
| E-invoice and e-way bill | Through an approved GST provider | Can be added |
| Receivables and payment reminders | Outstanding and due dates; WhatsApp or voice reminders | Can be added |
| Tally | Posting invoices and receipts | Can be added |
Reports, administration and integrations
| Feature | What it does | In this build |
|---|---|---|
| Six reports with Excel and CSV export | Order summary, stage performance, client-wise, customer history, paper consumption, overdue and on hold | Covered |
| Dashboard tiles and charts | Open, this year, delivered, clients; orders by stage and month | Covered |
| Audit log | Every action, with client-visible subset as the timeline | Covered |
| Roles and permission matrix | Seven roles, shown on screen, enforced in the database | Covered |
| In-app and browser notifications | Per role, user and client; service worker push | Covered |
| Live updates | Open screens refresh themselves | Covered |
| Global search and mobile layout | Find any order; use on a phone | Covered |
| Own domain, SSL, hosting | Deployed on the client’s domain from a GitHub pipeline | Covered |
| WhatsApp, SMS, e-mail senders | Automated messages on events | Can be added |
| AI reading, forecasting, owner brief | See the AI section | Can be added |
Benefits of a label printing ERP with a client portal, tied to the part of the build meant to deliver them
Print software sellers usually promise fewer phone calls, no re-typing, on-time delivery, accurate bills and a clear view of the floor. Those are reasonable goals, but they are only worth anything if the design delivers them. This section ties each benefit to the part of the Gunatit Label build meant to deliver it. No measured results are published: none were supplied by the client, and none are claimed.
The customer orders without calling
A login, the press’s own job card, product standards that fill themselves in and a Reorder button mean the office receives complete orders instead of messages to decode.
One specification, written once
The job card pre-fills the printing, punching and slitting sheets, so the core size the customer typed is the core size the slitter sees.
A date that is a rule, not a guess
The working-day rule gives the customer a date at once, and every change to the expected date carries a reason and a history.
The customer sees the status; the office stops repeating it
The timeline and the notifications tell the customer what the office used to tell them by phone, step by step.
The floor’s numbers become the press’s memory
Anilox, cylinder, rotation, metre, die and lots are saved per order and per job, so the next run of the same label starts from the last one.
Each manager sees their own work, and only that
My work queue and database-enforced roles give the printing manager the printing queue and the customer their own orders, nothing more.
The bill is what came off the slitter
Ready-to-invoice quantities come from the last stage that reported output, so the bill matches the rolls, not the plan.
Paper in running metres, not a guess in rolls
Lots with metres per roll, paper used per job and a consumption report by paper type make the next purchase a number instead of an estimate.
The owner sees the press on one page, from anywhere
Dashboard tiles, the production board, the live floor and six Excel reports, on a phone or a laptop, read from the same records the floor writes.
One thread, from the customer’s job card to the delivered roll
Around that thread sit the records that make it work: clients and their logins, paper-roll stock, the job register, master data, users and roles, the audit log and settings. Underneath, one database with row-level security, file storage for artwork and bills, authentication, a realtime channel for live screens and a server function for creating logins. The app is deployed from a GitHub pipeline to the client’s own domain. There is no second system to reconcile: the customer’s timeline, the manager’s queue, the office’s board and the owner’s report are four views of the same rows.
The modules
Eleven staff modules in the sidebar, a client portal with four screens, and the working screens between them — a focused business management system for one job: getting a label order from the customer’s job card to the delivered roll.
Client portal
Dashboard (“Your orders — where they are now”, “What we make”, paper-wise, colour-wise and size summaries, latest updates); Place new order; My orders with the timeline, remarks, reorder and bill download; Notifications.
Orders
Dashboard with hero tiles and charts; All orders with urgent and unreviewed highlighting; New order (office or sales); Order detail; Process (“Send to”) with route, expected date, pre-filled stage details, edit, cancel and reopen; Production board; Live floor; My work queue; a job sheet per stage with start, save, hold, resolve and complete; ready to invoice, dispatch and delivered.
Materials
Stock of paper-roll lots with balance in running metres, paper used, returns and adjustments; Job register (administrators only); Master data for paper type, colour, paper size, die size, core size and winding with usage counts and deactivate-not-delete.
People
Clients with code, contact, city, GSTIN, minimum order and portal login creation; Users & roles with the permission matrix and a server function for logins and passwords; Notifications per role, user and client.
Control
Reports & export (six reports, .xlsx and CSV); Audit log; Settings for upload limit, allowed file types, signed-link minutes, notification channel switches and company details; global search; mobile layout.
Roles and permissions
The build defines seven roles: Administrator, Office staff, Sales, Printing manager, Punching manager, Slitting manager and Client. The sidebar changes with the role: administrators and office staff see everything, with the job register and settings reserved for administrators; sales see the orders they entered and can place new ones; each production manager lands on My work queue for their stage and sees only orders routed through it; clients see their own dashboard, orders and notifications. Two slitting logins share the slitting stage. The permission matrix is shown on screen in Users & roles, so the owner can read who may place, review, route, start, hold, complete, dispatch, deliver, correct, edit masters, read reports or change settings.
One point worth stating plainly, because it is the difference between this build and a working version that only looks secure: the roles are enforced in the database by row-level security and in the server function that creates logins, not only in the screens. A client cannot read another client’s orders or artwork by guessing an address; a punching manager cannot open a slitting entry; anonymous requests are refused.
How it was built and handed over
Start from the job card
The press’s own job card and stage registers became the screens, field for field. Nothing was added or removed without being asked for.
Agree the flow on a working version
Client job card, “Send to”, three stages, ready to invoice, out for delivery, delivered: approved on screen before the database existed.
Rules as code, checked by scripts
Working-day dates, product standards, impressions, lot totals, bill quantities and status moves live in one service layer, exercised by headless flow checks.
Real database, real logins
Nine migrations: tables with row-level security per role, an order-number function, a private artwork bucket, realtime, stock, the job register, and a server function for logins.
Deploy on the client’s own domain
A GitHub pipeline builds and deploys every change; the client’s domain, with SSL, points at the app. Masters and the existing job register loaded.
Train, hand over, keep revising
Office team and production managers trained; delivered and invoiced on 3 November 2026 with twelve months of maintenance and revision support, under which changes continued.
See your own job card as a working portal before you decideWe build on your real paper types, stages and roles, so you judge something that runs, not a slide deck. Free workflow audit, reply the same day.
WhatsApp usWhat was delivered
What the system does and what was handed over — no invented numbers.
The handover included setting up master data, loading the printing job register from the client’s existing job sheet, role-based logins for administrators, office, sales, the three production managers and clients, deployment on the client’s own domain with SSL, and training for the office team and the production managers. WhatsApp Business API and SMS were recorded as options to be opted into later, and a Tally connection as something to be quoted separately. Twelve months of maintenance and revision support from go-live were included, and the revisions made in the days after handover — PVC tag defaults, punching impressions, the invoice summary, a new look and clickable dashboard tiles — were made under it.
What this page does not claim
- No claim that Gunatit Label’s staff or customers use the portal daily, how many customers have logins, how many orders have been placed, or since when. Those facts live in the client’s database, which was not and should not be inspected for this page.
- No claim that the invoice has been paid. The page says the system was delivered, deployed and invoiced, which is what the records show.
- No claim of time saved, calls reduced, deliveries made on time or any other result. There is no client quote, because the client has not written one.
- No WhatsApp, SMS or e-mail notifications are sent by this build (e-mail exists only for password reset). The channel switches in Settings have no sender behind them.
- No Tally, GST invoice, e-invoice, e-way bill or payment connection. The bill is prepared outside the system and uploaded as a PDF.
- No estimation, costing, quotations, QC module, purchase module, plate or die register, scheduling, roll traceability or mobile app.
- No AI model or machine-learning feature anywhere in the system. Every AI item on this page is labelled buildable or built elsewhere.
- The worked example on this page is an illustration using the build’s rules; its order, customer, dates and figures are not the client’s. The client’s job register, customer names, address and identifiers are not published.
Client build, delivered and handed over to Gunatit Label in November 2026, deployed on the client’s own domain with a real database and real logins. Anything not listed as built above is described as buildable, not as delivered.
Technology
The build uses Next.js 16 with React 19 and TypeScript, Tailwind CSS 4, shadcn/ui components on Radix, Lucide icons, Recharts for the dashboard charts, react-hook-form with zod for every form, and a Zustand store with Immer that applies each business rule and then persists the changed records to the database. The business rules live in one service layer: every action from placing an order to completing a stage goes through a function that validates the input, applies the change, writes an audit entry and sends the notifications, which is why the timeline, the board and the reports always agree.
Data lives in Supabase in the Mumbai region: Postgres with row-level security per role across sixteen tables, a database function that issues order numbers atomically, Supabase Auth for logins and password-reset e-mail, a private Storage bucket for artwork and bills keyed by client and served through short-lived signed links, Realtime for live screens, and a Deno Edge Function that creates logins and sets passwords after checking the caller is an active administrator or office user. Nine migrations define the schema, the policies, the realtime publication, stock, and the loading of the job register.
Hosting is on Vercel, deployed by a GitHub Actions workflow on every push to the main branch, with the client’s own domain and SSL pointed at the app. A service worker shows browser notifications on desktop and phone while the portal is open; the Excel export builds a real .xlsx workbook in the browser without a library. Headless check scripts walk the mapping from screen to field and the whole order flow before each release, alongside the TypeScript compiler.
What can be added next
- WhatsApp Business API messages on order events, dispatch with the bill attached, delivered confirmation and reorder nudges; SMS and e-mail senders behind the existing switches.
- A versioned artwork proof-approval gate that holds printing until the customer signs off, with the timestamp on the order.
- Estimation and quotations from the job card, with a rate structure for paper, ink, plates, dies and machine time; then job costing and waste analysis.
- GST invoicing in the system, e-invoice and e-way bill through an approved provider, receivables and payment reminders; or a Tally connection, as the press prefers.
- Purchase orders and goods receipt for paper, reorder-level alerts, and a plate, die and cylinder register with life and location.
- QR roll traceability from jumbo to slit roll, and a QC gate before dispatch, both on the pattern of the Jisha Label ERP.
- A machine-wise scheduling board by shift.
- The AI additions in the AI section: messages read into draft orders, AI draft estimates, paper forecasting, run-time outliers, reorder rhythm and the owner’s morning brief.
How much does a label printing ERP with a client portal cost, and how long does it take?
Yukti AI does not publish a price list, because the price follows the scope. Two things can be stated plainly: a custom ERP from Yukti AI starts from ₹30,000, and this build went from first code to handover in about three weeks, with the database connected in the second week and the client’s domain live at handover. The price of the Gunatit Label system itself is not published on this page. A system of this scope — a client portal, routing, three stage sheets, dispatch, stock, a job register, reports, roles and hosting on your own domain — is quoted on these factors:
- Which modules you need first: the portal and the stage sheets only, or stock, reports and the job register too.
- How many stages your floor has and what each stage sheet must record.
- How many roles, users and customer logins.
- Migration: your job register, client list and master data from registers, Excel or an older system.
- Integrations: WhatsApp Business API, GST invoicing or Tally, which are scoped and quoted separately.
- Hosting, your own domain, backups and support after go-live.
For context, packaged print-shop software in India is sold from a few thousand rupees a year for a small plan up to a one-time licence in the tens of thousands, and full print MIS implementations for larger presses run well beyond that; those products bring estimation and billing and usually leave out a customer portal. A custom build is a one-time price with support, shaped to your floor. The way to start is a free workflow audit: you show us one recent job card and how the order ran, we show what one system would replace, and you get a fixed quote. See custom ERP software cost in India for the general drivers and contact us to book the audit.
Can Yukti AI build this for another label printer, converter or packaging unit?
Yes. The same pattern — a customer-facing job card, office routing, a sheet per production stage, bill quantities from the last stage, dispatch with a challan and a document the customer can download — fits self-adhesive label presses, sticker and tag units, thermal-roll converters, flexible-packaging printers, screen and offset job shops and any plant where the customer’s order is a specification that several machines need in their own form. Each build is shaped around the client’s own job card, stages, roles and rules, starting with a free workflow audit.
See Manufacturing ERP Software, Custom ERP Software, Custom ERP Software in Surat and AI for manufacturing, or read about Yukti AI on the about page.
Label printing ERP software in Surat, Ahmedabad and Gujarat: who it is for
Label printing ERP software in Surat is one system that takes the customer’s job card, routes it through printing, punching and slitting, brings the quantities back to the bill and shows the customer where the order is, built by a team in the same state as your press. Yukti AI, at Leonard Square in Varachha, Surat, built the system on this page for a label printer in Ahmedabad and builds this kind of ERP for printers and converters across Surat, Ahmedabad, Vadodara, Rajkot, the rest of Gujarat and India.
Label demand in Gujarat follows the industries around it. Ahmedabad’s pharma, FMCG, agro-chemical and cosmetics brands need pharma, food, cosmetics and agro labels and water-bottle stickers in repeat runs; Surat’s textile, garment, diamond and jewellery trade needs garment and jewellery tags, barcode labels and thermal rolls; Vapi, Ankleshwar and Vadodara send chemical and lubricant labels; Rajkot and Morbi send engineering and ceramics labels. A label press serving these customers sees short, repeated runs with many sizes, papers and cores, which is exactly the pattern a job-card portal with reorder is built for.
Three kinds of unit usually look for label printing ERP software:
- Self-adhesive label presses with a flexo press, a die-punching machine and a slitter rewinder, selling to brand owners on repeat orders — the shape of the Gunatit Label build.
- Sticker, tag and thermal-roll converters that buy printed or plain stock, cut, slit and pack it, where the route often skips printing and goes to straight-to-invoice trading items.
- Wider label and packaging printers that also need estimation, artwork approval, plates and dies, QC and GST billing, which is the shape of the Jisha Label ERP in Surat.
Whichever you are, the same questions come up. Can my customers order without calling? What date can I promise, and who changed it? Where is this order right now? How many rolls really came off the slitter? How many metres of chromo did we print last month? Who may see what? A good system answers these from one set of records instead of four registers and a WhatsApp group.
This page is a delivered build, not a brochure. The system shown was built for an Ahmedabad label printer and deployed on its own domain, and the sections below turn that build into a plain buying guide for a Surat, Ahmedabad or Gujarat press: what an ERP, a print MIS and a job-card portal each are, what to ask any vendor, and what support from Surat looks like. To see how Yukti AI works with factories more broadly, read custom ERP software in Surat, AI automation agency in Surat and the case studies.
ERP vs print MIS vs job-card portal vs WhatsApp groups for a label press, explained plainly
A print MIS is management software for a printing business whose centre is estimation: it prices jobs, tracks them through production and usually bills them. An ERP is the wider system that also holds stock, purchase, accounts and people. A job-card portal is the customer-facing front of either: a login where customers place orders on your job card and follow them. A WhatsApp group is where most small presses actually run today. The Gunatit Label build is an order-to-delivery ERP with a job-card portal and without the estimation and accounts that a print MIS leads with; those are added when the press wants them.
When a print MIS is the right buy
When estimation is your bottleneck, your jobs vary widely in make-up, and you want quoting, scheduling and billing in one package you can start using with configuration rather than building. Packaged print MIS products from Indian and global vendors do this well, and several now add AI scheduling or forecasting. Most assume the office enters orders; a customer portal is an extra module or absent.
When a job-card portal with stage sheets is the right buy
When your customers reorder the same labels, your floor has a fixed set of stages with fields of its own, your office spends its day on status calls, and your bills are made from numbers collected from the slitter. That was Gunatit Label’s shape, and the system was scoped to match it: no estimator, because prices were already agreed with repeat customers; no accounts, because the billing software already existed.
When to move from WhatsApp groups
When a second person needs to know what the first person knows. A group holds messages; it does not hold a job card, a queue, a stage sheet or a report. The practical test is whether you can answer “how many rolls of PP white did we deliver to this customer this year?” in under a minute. For the general argument about moving off scattered tools, see replace Excel and Tally and ERP vs CRM: what a growing business needs first.
How automation and AI relate to all three
Automation moves information between the records and the people without anyone typing: dates, numbering, hand-overs, totals, notifications. AI reads and suggests: a message into a draft order, a job card into a draft estimate, consumption into a forecast. Neither works without the records, which is why the order of work on this page is records, rules, WhatsApp, billing, AI. See what is an AI agent and AI agent development in Surat for how an agent that reads an ERP is built.
Buying checklist: 12 questions to ask before you choose label printing software
Before you buy any label printing software, ask to see your own job card on screen, find out how your customers would place an order and what date they would be shown, check what each production manager would see and whether that is enforced in the database or only on the screen, and ask where the bill quantities come from. Ask for written scope for GST, e-invoice, Tally and WhatsApp, so you know what is included and what is extra.
- Can my customers order on my job card with their own login? And reorder with the artwork carried over?
- What date does the customer see, and who can change it? A rule with a history beats a promise from memory.
- Can I route a job through any subset of my stages, in my order? And send a trading item straight to invoice?
- Are the stage sheets my fields? Cylinder teeth, anilox per colour, die ups, lots of rolls × labels, or generic production entry?
- Does each manager see only their own queue? Ask whether that is enforced in the database.
- Where do the bill quantities come from? The job card, or what the last stage actually produced?
- What happens when an order is stuck? Hold with a reason, visible to the office, in a report.
- How is paper tracked? Rolls, running metres per paper type, paper used per job?
- What does billing include? Ask separately about GST invoice, e-invoice, e-way bill and Tally.
- Does it send WhatsApp messages, and who pays for them? Automated messages go through the Business API with charges to Meta.
- Who owns the data and where does it live? Ask about export, backups, your own domain and hosting.
- What is not included? A vendor who can list the gaps clearly is easier to trust.
A good test is to ask the vendor to write down what their system does not do. This page does exactly that in its what this page does not claim and not built here sections, and the same discipline applies to any supplier. For a wider comparison of approaches, read custom ERP vs off-the-shelf ERP.
Support from Surat: audit, training and help after go-live
Every Yukti AI project starts with a free workflow audit: over WhatsApp, on a call, or at the office at Leonard Square in Varachha, Surat. For a label press that means one recent job card, the stage registers and a description of how an order moves from the customer’s message to the delivered roll. We map it, say what one system would replace, and give a fixed quote. Presses in Surat can do this in person; the Gunatit Label build was done for Ahmedabad with the same process, remotely, with training for the office team and the production managers at handover.
After go-live, support means the kind of revisions that followed this handover: a default for a product, a field the punching manager wanted filled in, a summary the office asked for, a change to the look. Those came within days, under the included support, because the system is one codebase deployed by a pipeline, not a package waiting for a release. For the wider list of what Yukti AI builds for Gujarat manufacturers, see AI for manufacturing, all industries and workflow automation in Surat.
Label printing ERP, client portals and AI: questions label printers ask
Explore more
Case study: Jisha Label, Surat
The sister label ERP: estimation, artwork gate, plates and dies, QR roll traceability, QC and GST billing.
Case study: corrugated box ERP
Box costing, quotations, job cards, paper stock and dispatch for a Gujarat box plant.
Manufacturing ERP Software
Production, material and costing for Indian manufacturers.
Custom ERP Software in Surat
ERP built around how your Gujarat plant actually runs.
Custom ERP Software
One system shaped to your workflow instead of a package you adapt to.
WhatsApp Automation
Order updates, dispatch alerts and reminders from your system’s events.
ERP-Connected AI Agents
Agents that read your records, draft, remind and brief the owner.
Case study: The Box Factory
Manufacturing ERP with an AI quotation engine for a box maker.
Tally Integration
Connect a custom ERP to Tally, or move off it in stages.
Custom ERP Cost in India
What drives the price of a custom ERP and how to budget.
AI for Manufacturing
Where AI and automation help Indian manufacturers today.
All case studies
Every Yukti AI build across manufacturing, printing, diamond, jewellery and trading.
Summary
Yukti AI, a software company in Surat, built a label printing ERP with a client job-card portal for Gunatit Label, a self-adhesive label printer in Ahmedabad, Gujarat, and delivered it in November 2026, deployed on the client’s own domain with a real database and real logins. Customers place orders on the press’s own digital job card (paper type from eighteen, colour, product size, core, winding, With Gap or No Gap, rolls × labels per roll, artwork), see an approximate dispatch date at once from a working-day rule (7 days, 3 if urgent; order day and Sundays not counted), reorder in one tap with the artwork carried over, follow a timeline, add remarks and download the bill PDF after dispatch. The office reviews each order on a “Send to” screen and routes it through any subset of printing, punching and slitting, or straight to invoice for trading items, with an expected date whose changes are logged. Each production manager sees only their own queue, enforced by row-level security, and fills a stage sheet: printing (paper width, cylinder teeth, anilox per colour, rotation, running metre, output), punching (die size suggested and learned, die ups, impressions calculated as labels divided by ups), slitting (lots of rolls × labels, totals calculated). Stages can be held with a reason and resolved. The ready-to-invoice screen shows the lots, rolls, labels and paper used from the last stage that reported output; dispatch needs a DC number and takes a bill PDF; orders turn Delivered two days after dispatch unless recorded earlier; overdue after the expected date or ten days. Paper-roll stock is held in lots with the balance in running metres and paper used per job; the client’s existing printing job register (cylinder teeth, around and across with gaps) was loaded at handover; six reports export to Excel; every action is in an audit log whose client-visible entries form the customer’s timeline; notifications go in-app and as browser push, and open screens update live. Seven roles. Stack: Next.js 16, React 19, TypeScript, Tailwind 4, Supabase (Postgres with RLS, Auth, Storage, Realtime, Edge Function) in Mumbai, Vercel via GitHub Actions. Not built: WhatsApp or SMS sending, Tally, GST invoicing, payments, estimation, QC, purchase, plate and die register, scheduling, roll traceability, any AI. Daily use, order counts, payment and results are not claimed. A custom ERP from Yukti AI starts from ₹30,000; the final quote follows a free workflow audit.