No-code vs custom build: the trade-offs that bite

The no-code versus custom build decision usually gets made inside a fortnight, by a founder or an operations lead working from one polished demo and no itemised bill. Both options quote the same launch date and the same happy-path workflow, and neither shows you month seven. The costs that actually decide the project — licence creep, automation quotas, the "small" customisation that turns out to be a rewrite — appear only after work is underway, when switching is expensive. Here is what those costs look like, and how to price them before you sign.

Where no-code actually wins

No-code wins when the process is genuinely standard, the user count is under about fifteen, and the deadline is this quarter. Internal approval flows, expense claims, a simple enquiry pipeline, a dashboard over one data source — these are solved problems, and a tool at ₹1,500–3,000 per user per month will beat a six-figure build on both time and total cost. Trouble starts when you cross either threshold: seat pricing scales linearly while custom software scales in steps, and templates stop at the first genuinely unusual rule. If your need is outreach, email campaigns or pipeline tracking, a configured product usually beats both routes — see the products page for what that looks like in practice.

The bill that arrives after launch

The real cost of no-code is the subscription line in year two, not the setup fee. Setup is visible and cheap; what follows is not:

• Per-seat pricing that punishes field teams — 40 sales users at ₹2,000 each is ₹9.6 lakh a year before you have written a single automation.
• Automation task quotas, which a busy order flow can eat through in weeks rather than months.
• Record and API rate limits that throttle exactly the bulk operations your finance team runs at month-end.
• Missing local detail: GST credit notes, TDS adjustments, e-invoice IRN fields, UPI reconciliation references. Most can be added, but each is a consultant invoice.
• Export charges, or throttled bulk exports, priced at the moment you are least able to argue.

Ask for all five in writing before the pilot, not after.

What custom builds hide until month three

Custom builds hide integration and change-management cost, not development cost. The build itself is usually the well-estimated part; the overrun comes from everything the software has to talk to — Tally or Zoho Books, a payment gateway, the WhatsApp Business API with its template approvals, a legacy ERP with no documentation and one retired administrator who still answers the phone. Then there is the change-management tax: staff who used the old tool for six years need training, and the first version will be wrong in ways only they can spot. A realistic Indian mid-market build runs 8 to 14 weeks and ₹8–25 lakh, with 15–20% of that again per year in maintenance. Our case studies show what that scope creep looks like when it is caught in week two rather than week nine.

The exit question nobody asks

Whoever controls the schema controls your exit, so settle data ownership before you choose a tool. With no-code, ask what a full export actually contains — usually records, rarely the automation logic, almost never the attachments at a reasonable price. With custom, ask who holds the repository, whether the database sits in a region that satisfies your DPDP Act obligations, and whether consent and deletion logs are built in or bolted on. Both routes should give you a written exit plan: what you can take, in what format, and at what cost. If a vendor will not put that in the contract, that is your answer.

A worked comparison, with numbers

A 40-user B2B distributor moving order-to-cash off spreadsheets typically faces ₹12–14 lakh a year on no-code versus ₹18–22 lakh in year one for a custom build. The no-code stack is usually a database, a front end and an automation layer, licensed per seat, plus a consultant to wire it together. The custom route costs more upfront and less per user, but carries the integration work described above. Break-even generally lands somewhere in year two to year three — which means the decision is really about how long you intend to run the process unchanged. If the process is still shifting every quarter, do not pay for permanence.

Where to start

Write down the three rules your current process cannot express without a workaround. If all three are standard, buy the tool and move on. If two of them touch money — GST, TDS, credit notes, reconciliation — or your customer-facing WhatsApp flow, the custom route pays for itself faster than the licence maths suggests. Maysan Labs builds both, and we will tell you when the cheaper answer is the right one; a scoping conversation is usually enough to settle it.