The custom software vs off-the-shelf question lands on most South African founders eventually, usually dressed up as ambition. Before you spend a cent on a build, hear the unglamorous truth: for most needs, off-the-shelf wins, and if you are asking should I build custom software for a standard problem, the answer is almost always no. A studio that only ever says "build" is selling, not advising. So here is the honest version.

Default to off-the-shelf

Accounting, email, a basic online store, a standard CRM, payroll, project boards — these are solved problems. A monthly subscription tool is cheaper, faster to switch on, and someone else maintains it, patches it, and keeps it running while you sleep. You inherit years of other people's hard-won product decisions for the price of a seat. If a well-known tool already does the job, use it. Building your own version of Xero or Shopify is not clever; it is expensive and slow, and you will spend years catching up to features they ship for free.

We say this plainly because it is true more often than not. Most businesses do not need custom software. They need to configure the tools they already pay for properly.

The four honest tests for when custom is worth it

Custom earns its keep only when specific, measurable pain shows up. Run your business against these:

One — the copy-paste tax. Your off-the-shelf tools do not talk to each other, so a person spends part of every day moving the same data between them by hand. That salary is a hidden software cost.

Two — your edge lives in a process no generic tool models. If the thing that actually makes you money is a workflow no off-the-shelf product understands, you are bending your advantage to fit a template instead of the other way round.

Three — the per-seat creep. You are paying rising monthly fees across a stack of tools, and every new hire multiplies the bill across all of them at once.

Four — you need to own it. Lock-in risk, price hikes you cannot escape, or POPIA and compliance obligations that mean the data genuinely has to be yours — not a foreign vendor's to hold hostage.

None of these? Stay on the subscriptions. One or two, sharply felt? The maths starts to move.

The trap: the franken-stack

The failure mode we see most is not too little software — it is too much of the wrong shape. Ten subscriptions that do not integrate, held together by a shared spreadsheet and one long-suffering person who "knows how it all works". Each tool is fine alone. Together they are a tax you pay in hours, errors, and key-person risk. Adding an eleventh subscription rarely fixes a franken-stack. At some point the glue costs more than a system built to fit.

What custom actually buys you

Done properly, a custom build is not a fancier tool — it is a system shaped around how you already work, instead of a template you contort yourself to fit. At Mirembe Muse that also means you own it: the code, the data, and the logins ship to your accounts. No lock-in, no vendor holding your business ransom. It is POPIA-registered work (Reg 2026-005658) with row-level security, so the right people see the right records and no one else. And we scope it deposit-first at a fixed price — you know the number before we start, not an open-ended hourly meter that rewards us for going slowly.

Proof it is not theory

Two builds make this concrete. BB-MotherShip is a 74-table restaurant operations OS — one system that replaced a scatter of disconnected tools and the copy-paste between them. The Selves of Us is an 87-page publishing platform, custom-built for a workflow no off-the-shelf CMS modelled cleanly. In both cases the client was not chasing novelty. They had hit the copy-paste tax and the integration gaps, and the sums finally favoured owning the system over renting ten of them.

The honest closer

Buy off-the-shelf. Keep buying it — right up until the copy-paste tax, the integration gaps, or the lock-in risk cost you more than a custom build would. That crossover is a real number you can estimate, not a feeling. When you reach it, build the thing that fits your business — and make sure you own it outright. Until then, the subscription is the smart money, and we will be the first to tell you so.