ERPNext for Retail in Bangladesh: Why We Built MondayPOS on It

When we started Mondaysys in Uttara, we had a choice that every software company in Bangladesh faces at some point: write a POS from scratch, or build on something that already exists. Writing from scratch is tempting because it is "ours". We chose the other path. MondayPOS is built on ERPNext and the Frappe framework, an open-source ERP that thousands of businesses around the world run, and we have never regretted it.

This post explains that decision honestly. It covers what ERPNext actually is, why open source matters to a shop owner who will never read a line of code, what vanilla ERPNext is missing for a counter in Mirpur or Chattogram, what we added, and, just as importantly, who should skip us and self-host ERPNext themselves. If you are searching "erpnext bangladesh" because you are weighing exactly that, this is written for you.

What ERPNext and Frappe are

ERPNext is a full business system: accounting, inventory, purchasing, sales, point of sale, CRM, HR and payroll, manufacturing and projects, all in one database. It is developed by Frappe Technologies in India and by a worldwide community of contributors, and it has been in active development for well over a decade.

Underneath ERPNext sits the Frappe framework, the toolkit ERPNext is written in. Frappe handles the things every business app needs: user roles and permissions, forms, reports, workflows, an API, a print format engine, and a background job system. ERPNext is the business logic on top; Frappe is the engine underneath.

Both are open source projects, published openly and developed in public. In practical terms that means anyone can download, run, inspect and modify the software without paying for a licence, and that the code your business depends on is not a black box owned by one company. Check the terms on the official repositories yourself before you build anything on top of them.

That openness is not a footnote. It is the reason a company our size in Dhaka could ship a retail ERP with proper double-entry accounting, batch and expiry control and multi-warehouse stock in a couple of years instead of a decade.

Why open source matters to a shop owner

Most owners we meet do not care what a licence says. They care whether bills print, whether stock matches the shelf, and whether the VAT consultant is satisfied. Fair enough. But three consequences of open source reach the counter directly.

No lock-in

With a closed POS, your supplier owns the software, the format your data lives in, and often the server it runs on. If the relationship sours, or the company closes, or the price doubles at renewal, you have limited options. With ERPNext, the software itself belongs to nobody and everybody. If Mondaysys disappeared tomorrow, your MondayPOS database would still be an ERPNext database that any ERPNext partner, in Bangladesh or abroad, could host and maintain. We think a vendor should earn renewal every year, not hold it hostage.

Your data is exportable

ERPNext stores data in a standard MariaDB database with a documented structure. Every list in the system, items, customers, invoices, stock ledger, general ledger, can be exported to CSV or Excel by a user with the right permission, and the full database can be backed up as a file. When a prospective customer asks "how do I get my data out if I leave?", the answer is a button, not a negotiation. We put that question in every demo checklist we publish because we are comfortable answering it.

A community larger than any vendor

Bugs in the accounting core, security patches, new features like better batch handling or faster reports come from a global community and a company much larger than ours. We pull those improvements into MondayPOS rather than re-inventing them. For an owner, it means the ledger and stock engine underneath the till has been tested by far more businesses than we alone could ever serve.

What vanilla ERPNext lacks for Bangladesh retail

Here is the honest part. If you download ERPNext today and install it for a supershop in Bashundhara, it will not work the way you need on day one. We know, because we tried. These are the gaps we found, and they are the reason MondayPOS exists as a product rather than a service contract.

None of this is a criticism of ERPNext. It is a global product; it cannot know that NBR wants a specific invoice layout, or that a Nagad payment needs to sit in its own bucket at day-end so the cashier can match it against the app balance. That localisation is the job of a local company, and that job is ours.

Mushak forms

A VAT-registered shop needs to issue a Mushak 6.3 invoice and keep the purchase and sales registers NBR prescribes. ERPNext gives you a perfectly good tax engine and a print-format designer, which means a capable developer can build Mushak 6.3 in it. But "can be built" and "is there" are very different things when the VAT consultant is standing at the counter. We built it, and we maintain it as the forms change. Our Mushak 6.3 guide explains which forms a shop actually needs.

Mobile financial services

bKash and Nagad are not an "integration" in Bangladesh; they are cash. A counter that cannot ring up a bKash payment as bKash and show it separately in the day-end report forces the cashier into a side notebook, and side notebooks are where shrinkage starts. MondayPOS treats each MFS as a tender type with its own closing line.

Offline-first

ERPNext's point-of-sale screen caches items and can take bills while the server is briefly unreachable. For an office in Bengaluru with a stable connection that is enough. For a counter in Bogura during evening load-shedding, it is not. We rebuilt the counter so that the terminal is the primary place a bill lives, and the server is where it goes afterwards. Our load-shedding post shows what that looks like when the router goes dark mid-bill.

Bangla and local hardware

Cashiers should not have to learn English to work a till. MondayPOS ships a Bangla interface and Bangla receipt formats, and we have tested the thermal printers, barcode scanners, cash drawers and scales that shops actually buy from Multiplan and Elephant Road. ERPNext is hardware-agnostic, and so are we; the difference is that we have already done the setup.

Who should self-host ERPNext instead of using MondayPOS

We would rather you self-host ERPNext and succeed than buy MondayPOS and resent it. Self-hosting is a good choice if most of the following are true:

• You have at least one developer or system administrator on staff who is comfortable with Linux, Python and MariaDB, or a long-term relationship with a Frappe partner.
• You are not VAT-registered yet, or you already have a VAT consultant who will produce Mushak forms outside the system.
• Your connectivity is reliable enough that brief offline gaps are an inconvenience, not a daily event.
• You need modules MondayPOS does not focus on, such as manufacturing, projects or a complex subscription business.
• You want to customise deeply and own every change, and you accept that you will maintain those changes through every ERPNext upgrade.

If that is you, ERPNext is free to download, the documentation is good, and Frappe Cloud offers hosted plans priced in US dollars for those who want the software without the server work. There is also a handful of Frappe partners in Bangladesh who do implementation work on a project basis. We are happy to point you to the community resources at a demo even if you do not buy from us.

Who should use MondayPOS

MondayPOS makes sense when the list flips:

• You run a shop, a pharmacy, a supershop, a restaurant or a small chain, and you have no one on staff whose job is servers.
• You are VAT-registered or expect to be, and you want Mushak 6.3 printed from the same screen that prints the receipt.
• bKash and Nagad are a meaningful share of your takings and you want them reconciled automatically at day-end.
• Load-shedding and broadband drops are a normal week, not an exception.
• You want the setup, migration, training and support in Bangla, from a team in Dhaka, at a published price in taka.

Pricing is public: Lite at ৳1,500 per outlet per month for a single small shop, Business at ৳3,500 (POS, sales, purchasing, stock and accounting, with ৳2,450 for each extra outlet), Growth at ৳6,500 (adds CRM, HR and payroll, ৳4,550 per extra outlet), all billed yearly with a 14-day refund window. The full module list is on the features page.

The honest trade-offs

Building on ERPNext gives us a deep foundation, but it comes with costs we should name.

It is heavier than a simple till. ERPNext is a full ERP, and even though MondayPOS hides most of it behind a retail-first interface, a single corner shop that only ever needs a bill and a stock list may find a lighter, cheaper tool sufficient. We say so in our free POS software post.

Upgrades take work. ERPNext releases major versions regularly. Every localisation we have built has to be tested against each new release before we roll it out. That is our job, not yours, but it does mean we are sometimes one version behind the newest community release, on purpose.

Openness cuts both ways. Building on a public project means a competitor can see the same foundation we did, and in principle build something similar. We think that is fair, and it is the deal we accepted going in. We compete on support, setup and the speed with which we adapt to NBR and MFS changes, not on hiding code.

We are a younger brand. Some local vendors have been selling POS for two decades. Our foundation is mature; our company is not. Ask us for references and check them.

Annual billing. Our plans are billed yearly. That keeps the price per outlet low, but it is not right for everyone, and we would rather say that up front than at the invoice.

Questions to ask any ERPNext-based vendor at a demo

Several companies in Bangladesh, not only us, sell systems built on ERPNext or Frappe. Whoever you talk to, these questions separate a real localisation from a fresh install with a logo on it.

1. 1Print a Mushak 6.3 invoice for this sale and show me the VAT registers for the month. Which version of the form is it?
2. 2Unplug the router and complete three bills. Reconnect. Show me where they landed and that the stock ledger moved.
3. 3Take a bKash payment, a Nagad payment and a card payment. Run day-end closing. Show me each tender on its own line.
4. 4Switch the interface and the receipt to Bangla.
5. 5Which ERPNext version is this, and when did you last upgrade? What happened to your customisations?
6. 6Export all customers and all sales invoices for the last month to Excel, right now.
7. 7If I stop paying, what do I receive, and could another ERPNext partner host it?

If a vendor hesitates on the last two, the open-source foundation is not working in your favour.

The bottom line

We built MondayPOS on ERPNext because the hard, boring parts of a retail ERP, the ledger, the stock engine, permissions, audit trails, were already solved in the open, tested by a global community and licensed so that nobody, including us, can lock you in. What ERPNext could not know is Bangladesh: Mushak 6.3, bKash and Nagad at the counter, six hours of load-shedding in a bad week, cashiers who work in Bangla, and printers bought from Elephant Road. That is the layer we added, and it is the layer we keep maintaining.

If you have a technical team and the time, self-host ERPNext; it is an excellent system. If you would rather run the shop, book a demo and run the seven questions above against us, or start free.