Startups

We work with founders turning an idea into something real, in Arusha, across Tanzania, and with Tanzanian founders building from abroad.

What founders usually get wrong first

The most common mistake is building too much before anyone has paid for anything. It is understandable. Building feels like progress, and specifying features is more comfortable than asking strangers whether they want the thing.

But scope added before validation is the most expensive scope there is, because it is the most likely to be wrong. Almost every founder we work with ends up removing something they were certain about at the start, and the ones who built least before finding that out lost the least.

Our job in the early stage is often to argue you out of features, not into them.

The smallest thing that proves the idea

We start by identifying the single assumption the whole business rests on, then building the smallest thing that tests it honestly.

Sometimes that is not software at all. If your assumption is that people will pay for a delivery service, a landing page, a WhatsApp number and two weeks of doing it manually will teach you more than three months of development, and cost almost nothing. We will tell you when that is the case, even though it is less work for us.

When it is software, we build the version that does one thing properly, get it in front of real users quickly, and let what they actually do reshape the plan. See custom software development for how we run that.

Looking established before you are

Early-stage companies face a credibility gap that has nothing to do with quality. Customers, partners and investors are all quietly assessing whether you will still exist in a year.

A great deal of that judgement is formed from surfaces: a coherent identity, a domain that is yours, email on that domain rather than a free account, a site that answers the obvious questions, and materials that look like they came from one company rather than four freelancers. None of this is expensive at the start, and all of it is expensive to retrofit once you have printed things and signed clients. See brand identity.

Building so the next decision stays cheap

Early technical decisions are made with the least information you will ever have, so the goal is not to predict correctly. It is to stay cheap to change.

In practice that means mainstream technologies another developer can pick up, no architecture built for a scale you have not reached, and a clear separation between the parts likely to change and the parts unlikely to. It also means you own the code, the accounts and the infrastructure from day one, in your name, so that a future co-founder, employee or investor is not inheriting a dependency on us.

Being honest about money

Founders are frequently sold more than they need, because the person quoting benefits from the larger number. We would rather have a client for six years than a large invoice once.

So we scope in stages with fixed prices per stage, we say plainly when something can wait, and we will tell you when an off-the-shelf product does the job well enough that building your own would be a waste. If your budget does not fit the ambition, we would rather help you sequence it than quietly build a worse version of everything at once.

Related work

Most founders start with a website and an identity, then move to product. If you are building for tourism, see our work in tourism and safari. If it is restaurants, hotels or lodges, see hospitality.