Software is cheap to start and expensive to keep. The second number is the one that decides.

Software development South Africa businesses can actually live with: websites, online stores, web and mobile applications, subscription products, and prototypes that test an idea before you commit to building it properly.

What we build

Why this matters

The quote is for the build. The bill is for the decade

Every software conversation is anchored on one number, which is what it costs to build. It is the wrong anchor, and it is wrong by a large margin. The build is a one-off; the operating systems it runs on update every year, the libraries underneath it go out of support, the payment provider changes its API, and none of that asks whether you have budget this quarter.

That is why so much custom software ends up abandoned rather than replaced. It was scoped as a project and it is actually a commitment, so the first year looks affordable and the third year looks like a decision nobody remembers making. The way out is not to build less carefully. It is to scope honestly - fewer things, chosen deliberately, built so they can be maintained by somebody other than whoever wrote them.

The other half is deciding what to build at all, and it is where most of the money is genuinely lost. The largest single cause of new products failing is not the engineering. It is having built something the market did not want - which is knowable beforehand, cheaply, and is skipped by most of the people it would have saved.

So we do two unfashionable things. We tell you when a smaller build, an existing platform, or no software at all is the right answer. And we would rather sell you a prototype that costs a fraction of the build and might tell you not to proceed, than the build itself.

43%

Of failed startups died of poor product-market fit. Not the code. The decision about what the code was for.

70%

Ran out of money, which reads as the cause and is the symptom. The capital went into building the wrong thing well.

40%

Of founders formally validate anything before launching. The cheapest step in the whole process is the one most often skipped.

Three things that turn a build into a write-off

Each of these is a decision that feels prudent at the time.

Let us get everything in the first version.

Scope is the single biggest driver of whether software ever gets used. A longer first version arrives later, costs more to maintain forever, and pushes the moment a user gets something useful further away - which is precisely where products lose people. The strongest version one is almost always smaller than the one that was requested.

We will validate it once it is built.

That is the most expensive possible order. Whether anybody wants this is answerable in weeks for a fraction of the build, by putting something crude in front of real people and watching what they do. Building first means you find out with the budget already spent, which is the shape almost every failed product has.

Custom software will be cheaper than the subscription.

Sometimes, and we will do the arithmetic with you honestly. But the comparison has to include what it costs to keep running, not just to build - and against a mature platform that is maintained by somebody else, custom often loses. We would rather set you up properly on something that exists than sell you a build you will regret in year three.

Also from LAUNCHT

Building it is half of it. Getting anybody to find it and use it is the other half, and it is what we do the rest of the time.

Marketing

Tell us what you are trying to build, and we will tell you if you should

Fifteen minutes with the people who would do the software development South Africa work, not a salesperson. What it has to do, what it replaces, and whether a smaller version, an existing platform or a prototype gets you the answer faster. We will say so if it does.