The rule of thumb

Buy off-the-shelf for anything that is not your competitive advantage — accounting, email, HR. Build custom only where your process is genuinely different from everyone else's, and that difference is why customers choose you.

This decision gets made emotionally more often than analytically. Founders who like control build everything; founders who fear cost buy everything. Both are expensive mistakes in different directions.

The honest comparison

FactorOff-the-shelfCustom built
Up-front costLow or zero₹2L – ₹25L+
Ongoing costPer user, per month, foreverHosting + maintenance only
Time to runningDays2–8 months
Fit to your processYou adapt to itIt adapts to you
SupportVendor SLAYour developer or team
Data ownershipOn their serversYours
Price increasesVendor decidesNot applicable
Risk of shutdownReal — vendors get acquiredNone

Where the crossover actually happens

The maths is simpler than people expect. A SaaS tool at ₹800 per user per month:

₹1.9L
20 users, 1 year
₹9.6L
20 users, 5 years
₹24L
50 users, 5 years

At 50 users over five years you are spending roughly ₹24 lakh on subscriptions — comfortably more than most custom builds cost. That is the arithmetic that makes custom development sensible for growing companies, and pointless for small ones.

Do not run this calculation naively. Custom software is not a one-time cost — budget 15–20% of build cost annually for maintenance, plus hosting. Include that before declaring a crossover point, or you will be unpleasantly surprised in year two.

Buy when…

  • The function is standard — accounting, payroll, email, video calls.
  • Regulations change frequently and the vendor absorbs that burden (GST filing is a good example).
  • Your team is small enough that per-seat pricing stays cheap.
  • You need it running this month, not this year.
  • The process is not a source of competitive advantage.

Build when…

  • Your workflow genuinely differs from the industry norm — and that difference is why customers choose you.
  • You are paying for a large platform but using 10% of it.
  • Your team maintains elaborate spreadsheets to bridge gaps between tools.
  • Per-seat costs have become a meaningful line item.
  • You need deep integration between systems that do not talk to each other.
  • Data residency or confidentiality requirements rule out hosted options.

The option most people skip

There is a third path that is frequently the right one and rarely considered: extend a platform you already pay for.

  1. Custom modules on an existing platform Zoho, Salesforce, and Odoo all support bespoke modules. You keep the vendor's maintenance burden and add only what you need.
  2. Integration layers Often the pain is not any single tool but the gaps between them. A focused integration connecting your CRM to your accounting system can eliminate the problem for a fraction of a rewrite.
  3. A custom front end on standard back ends Keep the proven system of record; build only the interface your team actually needs.

Most "we need custom software" conversations end with a ₹3 lakh integration instead of a ₹20 lakh rebuild — and the client is better off.

A five-question test

Score one point per "yes". Three or more suggests custom is worth costing properly:

  • Have you evaluated at least three off-the-shelf options and found genuine gaps?
  • Are staff maintaining manual workarounds weekly because the software cannot do something?
  • Would per-seat costs exceed roughly ₹5 lakh a year within two years?
  • Is the process in question a genuine competitive differentiator?
  • Do you have someone internally who can own the software long-term?

That last one matters more than people expect. Custom software without an internal owner drifts into disrepair regardless of how well it was built.

If you would like an honest assessment — including being told to just buy something — describe your workflow. See also our website vs web application guide.

Frequently asked questions

Up front, yes — usually significantly. Over three to five years it often costs less, because per-user subscription fees compound while a custom build is a one-time cost plus modest maintenance. The crossover point for most Indian SMBs is somewhere between 25 and 50 users.
Run a genuine trial with your real data and your real workflow for two weeks. If your team is inventing spreadsheet workarounds within days, it does not fit. If they only complain about cosmetics, it does.
Often, and it is usually the cheapest good answer. Zoho, Salesforce, and most serious platforms support custom modules and integrations. This gets you 80% of custom behaviour at 20% of custom cost — and should be ruled out before you commission a full build.