Setu vs Marg ERP: Free Tools or a Licensed Retail ERP in 2026?

A shop reaches a point where the notebook stops working. Usually it arrives with the second counter, or the first employee who is trusted to sell but not to remember, or the day a supplier claims a payment was never made and nobody can prove otherwise. At that point the owner starts searching, and two very different answers come back: buy a proper retail package, or string together free tools and keep the money.

Marg ERP is the first answer and it has been for a long time. It is one of the oldest Indian retail and distribution packages, with deep roots in pharmacy wholesale, FMCG distribution and general trade. Setu is the second answer, and it is free browser tools with no account behind them. They end up on the same shortlist more often than their makers would expect, so here is where the line between them actually falls.

What Marg ERP is built to do

Marg is on-premise software: a full retail and distribution package that runs on a Windows machine in your shop. The core of it is billing plus inventory plus accounting in one database: you raise a bill, stock moves, the ledger posts, and the accounting side of the business stays in step with the physical side without anyone retyping anything. That single-database behaviour is the thing a bundle of separate tools cannot imitate.

Around that core sit the features its traditional customers depend on. Batch and expiry tracking, which is the reason it dominates pharmacy retail and distribution. Barcode scanning at the counter. Supplier ordering built from your own sales history. Multi-godown stock. Scheme and margin handling for distributors. GST returns prepared from your own billing data, plus e-invoicing and e-way bill generation, which are portal-connected operations no browser tool performs. There is also a cloud edition marketed separately for businesses that would rather not run their own machine.

The honest summary is that Marg is a system of record. It expects to be the place your business lives, and it is priced and implemented accordingly.

What it costs, as publicly listed

Four editions appear consistently across Indian software marketplaces, as publicly listed in October 2026: Nano at about ₹5,550, Basic at about ₹10,300 for one user with full rights, Silver at about ₹13,900 for one user with full rights, and Gold at about ₹26,000 for unlimited users. Every one of those figures is quoted exclusive of tax.

Two cautions, and they matter more than the numbers. First, the listings do not agree on what the figure buys. One marketplace presents the four prices under a yearly selector; another presents the same four as a perpetual licence plus 18% GST, with annual support quoted separately by the vendor. A subscription and a perpetual licence are very different purchases, and the difference is the whole basis of a five-year budget. Second, the traditional Marg model in the trade is a licence plus a yearly maintenance contract, with the dealer who sells and installs it also billing for setup, data entry and training. None of that appears on a marketplace card.

So treat any figure you read, including the ones above, as a starting point for a quote rather than a price. Get it in writing from the vendor's own price list or an authorised dealer, and ask three specific questions: is this a licence or a subscription, what does renewal cost in year two, and what is quoted separately for installation and training.

What Setu's tools do instead

Setu splits the same work into separate free browser tools. There is no account, no seat count, no document cap and no renewal. The Invoice Generator produces a GST-formatted bill with the tax split worked out. The Stock Register tracks what is on the shelf. The Purchase Register records what came in and from whom. The Barcode Generator handles labelling for a counter that scans.

What you are not getting is the join between them. Each tool holds its own working data in the browser that opened it. Raising a bill does not move stock; receiving a purchase does not post to a ledger. You are the integration. For a single counter doing modest volume that is a completely workable arrangement, and it is how a very large number of small shops already run, just with paper in the gaps instead of browser tabs.

The honest comparison

The three questions that actually decide it

Does stock have to move when you bill? This is the cleanest split and it is not close. If you sell a few hundred SKUs and need the shelf count to be right at the end of the day without a manual stock take, you need one database doing both jobs. A billing page and a stock page that do not talk to each other will drift within a week, and the drift is invisible until it is large. If your stock is small enough to eyeball, or you count it weekly anyway, the join is a convenience rather than a requirement.

Do you sell anything with a batch and an expiry? Medicines, food, cosmetics, agri inputs. If yes, this is effectively decided. Expiry-tracked stock is the thing Marg and its direct rivals were built for, and handling it on a spreadsheet is how shops end up writing off boxes they could have returned to the distributor in time.

Does the e-invoicing mandate reach you? Notification 10/2023-Central Tax set the threshold at aggregate annual turnover above ₹5 crore, measured at PAN level across every registration you hold, applied from 1 August 2023. Crossing it in any financial year from 2017-18 counts, and there is no exit if turnover later falls. A separate restriction from 1 April 2025 stops businesses at ₹10 crore and above from reporting a document to the portal more than thirty days after its date. Several categories are carved out regardless of turnover, and B2C supplies sit outside the mandate. If any of this reaches you, a browser-only generator is not the tool. None of this is tax advice and notifications change — confirm your own position with your CA.

What free and one-time each hide

Both sides advertise a price that is not the whole cost, in opposite directions, which is why total cost of ownership is the only comparison worth making.

A licensed package carries a tail. Setup and data entry at the start, training for whoever uses it, a renewal or support contract, and a machine that has to keep running. The bigger item is harder to see: the business becomes the shape of the software. Three years in, your history, your supplier records and your working habits are inside it, and moving to anything else means a data migration nobody wants to pay for. That is not a criticism of Marg specifically, it is simply what a system of record does, and it is the reason the purchase deserves a proper evaluation rather than a dealer visit and a decision in the same week.

A free browser tool has the opposite problem. Nothing is stored for you, so nothing is locked in — and nothing is protected either. The history lives on one machine, in one browser, under one person's habits. The cost is paid in discipline: export, back up, and accept that a reinstalled laptop takes your records with it. You also pay in retyping, because the tools do not post to each other.

Neither of those is the worse deal. They fail in different directions, and the right question is which failure your business can absorb.

Where each one wins

Marg wins when inventory is the business: hundreds of SKUs, batches and expiry, more than one counter, a distributor relationship with schemes and margins, portal filing obligations, and enough volume that an hour saved a day pays the licence in a month. Pharmacy and distribution are its home ground for good reason.

Setu's tools win when the document is the job and the shelf is small. A single-counter trader, a service business with stock on the side, a new shop that has not settled its process yet, a seasonal operation that cannot justify a renewal. Getting the tax split right on a bill and keeping an honest purchase record covers a great deal of ground before software becomes the limiting factor. The Zoho Inventory comparison lands on a similar line from the inventory side, which suggests the line is real rather than a pitch.

One option neither column covers. If the actual problem is not which software raises the bill but that nobody is reconciling, categorising or closing the month behind it, then no licence fixes it. That is what a bookkeeping service is for, and Setu's consultancy works inside whichever tool you have already chosen rather than asking you to move.