Small studio. Direct work.
TreeInAPool is a product development agency: a small, senior team that takes prototypes and rough operations and turns them into software people rely on.
Origin
Most software projects don't fail in the code. They fail in the gap between what a client thought they were buying and what the agency thought it was building. It is a gap that only shows up at the end, when it's expensive.
TreeInAPool came out of building. Vuvu.ng set the standard. Bill payments are unglamorous and unforgiving: the flows are short, the tolerance for error is zero, and the product only counts once it sits on the phone people already pay with, which meant defining every journey from onboarding to receipt and taking the web app into both stores rather than stopping at a URL.
Farm Buddy was the opposite discipline, a question about input rather than output. Its users were never going to type a description in English, so the camera became the primary input, a voice note became a valid one, and a multimodal model turned a photograph into a diagnosis and something to do about it. That is a product decision before it is a technical one, and getting it wrong would have produced a demo nobody could use. AfroMadeIt Global was a third kind of problem, where the build is not the hard part and deciding what the business actually claims is.
The pattern across all three was the same. The interesting work was never the demo. It was the distance between something that runs and something a stranger can depend on. Closing that distance is what this studio sells.
A prototype is a cut flower. It looks alive, it photographs well, and it is already dying, because nothing underneath is feeding it. A tree is the other thing: rooted, slower to get going, and still standing next season. The pool is where we keep finding them, a contained space nobody planted for, holding something that has to grow anyway. The name is a reminder of which of the two we are being paid to produce.
Written, not implied
Price, timeline and scope are agreed in a document before work starts. Changes are quoted and approved before they're built. Every conversation about money happens before the work, not after it, because that is the conversation that ends most agency relationships badly.
Show, don't report
A status report is easy to write and impossible to verify. A live URL is neither. Every sprint ends with something you can open, click, and dislike out loud while there is still time to change it.
Ownership is not a bargaining chip
Code, accounts, domains and infrastructure are yours and transfer at launch. Some agencies keep hold of these to guarantee a retainer. We would rather earn the next engagement by having done the last one well.
The honest no
Some projects don't need an agency yet, and some need a different one. Saying so on the call costs us a proposal and saves you a quarter. We would rather be the studio that told you the truth early.
How we work
Every engagement runs the same four stages, whether it's a three-week automation or a fourteen-week build. The durations change; the shape doesn't.
Olaseni Otusanya
TreeInAPool is run as a small, hands-on studio rather than a large agency with account layers between you and the people building your product. On most engagements the founder is in the discovery call and the sprint demos, not just the sales call.
Olaseni is a product and operations lead by trade. Roadmaps, PRDs, acceptance criteria, release plans, and a backlog prioritised by impact against effort rather than by whoever asked most recently. He has run that work across fintech, ecommerce and media, which is a polite way of saying he has shipped in the three categories where being wrong is visible to everybody by Tuesday.
At Baniri Technologies he owned the definition and launch of Vuvu.ng, writing the flows and acceptance criteria for every bill-payment journey from onboarding through to the receipt, and taking it from a web app to a listing in both stores. He is Co-Founder and COO for product at AfroMadeIt Global, a strategic HR consultancy operating across US, UK and African markets. His own engineering runs to computer vision, forecasting and language automation: decision-support systems, where a confidently wrong answer costs somebody real money.
He arrived at product from an evidence discipline rather than from design or engineering, which shows up as an unfashionable interest in whether a thing is actually being used, by whom, and instead of what. He reads more papers than is strictly good for him, holds opinions about empty states he is glad to defend at length, and will tell you your project does not need an agency yet when that is the true answer. The thread through all of it is a mild allergy to software that demos beautifully and folds the moment a stranger touches it. Most of this job is declining to be impressed by your own prototype.
Core discovery, design and architecture stay with the founder on every project. Build capacity flexes with a small bench of vetted engineers and designers brought in per engagement, briefed by, and accountable to, the same person you talked to on the discovery call.
Frequently asked
You do. Repositories and accounts are transferred to you at launch, not licensed back to you.
About two hours a week: one demo, plus decisions when a question is genuinely yours to answer. Everything else is our job.
Often, yes. It starts with an audit of what exists so nobody promises a timeline before reading the code.
Talk to the person building it
Thirty minutes, direct with the founder, no account manager relay.