The build-or-buy conversation usually starts in the wrong place. Someone demonstrates a platform, someone else says it does not fit the way the business works, and within twenty minutes the discussion is about features instead of about which choice the company can live with for the next five years. The decision is genuinely difficult, but it becomes tractable once you separate what is distinctive about your operation from what is simply familiar.
Start with what makes you different
Every business has processes that are genuinely its own and processes that are ordinary work done in a particular habit. Payroll, accounting and email are not competitive advantages. The way you price a complex job, schedule field teams across emirates, or handle a bilingual approval chain might be. Buy the ordinary, build the distinctive — and be ruthless about which is which, because most organisations overestimate how unusual they are.
The customisation trap
A ready-made platform heavily customised is often the worst of both worlds: you pay licence fees and development costs, you cannot upgrade cleanly, and you own the maintenance burden without owning the product. If you find yourself planning extensive modifications before go-live, that is a signal the fit is wrong, not a plan. Configuration is fine; deep customisation is a warning.
Cost behaves differently, not just differently sized
Off-the-shelf front-loads speed and back-loads dependency — costs scale with users, and pricing changes are outside your control. Custom front-loads investment and back-loads control — the build costs more initially but per-user growth does not carry a licence multiple. Neither is cheaper in general; they are cheaper under different conditions.
| Consideration | Off-the-shelf | Custom build |
|---|---|---|
| Time to first use | Weeks | Months |
| Fit to unusual processes | Partial, via configuration | Exact, by design |
| Cost as headcount grows | Rises with seats | Largely flat |
| Who carries the risk | Vendor roadmap and pricing | Your team and delivery partner |
| Exit difficulty | Data export and migration | Handover of code and knowledge |
Questions worth answering before committing
- If this vendor doubled its pricing or discontinued the product, what would we do?
- Can we export our data in a usable form, on demand, without vendor assistance?
- Does it handle Arabic content, right-to-left interfaces and local invoicing formats properly?
- Where is the data hosted, and does that satisfy our obligations and our clients’ expectations?
- Who maintains the integrations we depend on, and what happens when one breaks?
A decision process that avoids regret
- Document the current process as it truly runs, including the workarounds nobody admits to.
- Separate requirements into essential, valuable and merely familiar.
- Trial two credible off-the-shelf options against your essential list, using real data.
- If neither clears the essentials without deep customisation, scope a custom build for that gap only.
- Consider the hybrid: buy the platform, build the one component that is genuinely yours, integrate carefully.
Security and continuity apply either way
A custom build does not automatically mean better security, and a large vendor does not automatically mean safe. Ask the same questions of both: how access is controlled, how backups are taken and tested, how updates are applied, and who is accountable when something fails at midnight. Data protection obligations follow your business regardless of who wrote the software, so confirm your responsibilities with a qualified adviser rather than assuming a supplier has covered them.
What to do next
Most mid-market UAE businesses end up buying their core platform and building a thin, well-defined layer around it — a pricing engine, a client portal, a reporting tool. That approach keeps you on a supported product for the ordinary work while giving you exact control where it matters commercially. It is usually the honest answer, and it is a good place to aim.
Write down the three processes that would be hardest to explain to a competitor, and the three that any similar business would recognise instantly. Take that single page into your next vendor demonstration and judge the platform only against the first list. If it handles those without modification, buy it. If it does not, you now know precisely what to scope.
