Last checked
Ask twelve things: who does the work, what similar products they shipped, how they scope, how often you see progress, who owns the code and accounts, how they test, what is excluded, how changes are priced, how payments are staged, what happens after launch and how you can leave. Good answers are specific and written down.
Picture the first meeting: a laptop on a table at a Nairobi cafe, a founder on one side, two people from a studio on the other. The studio shows a slide deck. The founder wants to ask about money and is not sure what else to ask.
This list is for that founder. It covers twelve questions, with what a good answer sounds like and what should make you pause. We are a studio ourselves, so we have answered each one honestly from the other side of the table.
The questions work for any studio, including us. For the budget side, read how much it costs to build an app in Kenya first, so the price conversation starts with line items.
Good: names and roles, and whether they are employees or contractors. Pause if: the person who sells the job is not the person who builds it, and nobody will say who that is.
Good: a live product you can open on your phone, and a client you may speak to. Pause if: only screenshots, or "under NDA" for everything. Some work is private, but not all of it.
Good: a plain story with what changed afterwards. Pause if: "nothing, our clients are always happy". Every studio has a late job somewhere.
The answers to these three are easier to judge if you have seen real products. A few of ours are public: the HackHouse case study shows a client platform with ticketing and M-Pesa payments, and our other products are listed on the same site.
Get weekly developer tips
Join 25,000+ developers. Practical guides, job tips, and new content — straight to your inbox.
No spam. Unsubscribe anytime.
Good: a short discovery step, then a written list of screens, roles and rules, and a list of what is out. Pause if: a price arrives before anyone has asked you a question.
Good: a demo every week or two that you can click, on a real address. Pause if: "you will see it at the end". Our own habit is a first working version in the first week where scope allows, and weekly demos after.
Good: testing in a sandbox, then small real payments, then a check on late and duplicate notices. Pause if: the developer tests only the happy path. For M-Pesa work, read what it costs a business to integrate M-Pesa to see what a full test covers.
Good: a written list: hosting, third-party fees, content, training, store fees. Pause if: "everything is included". Nothing is.
Good: a rate or a process: you ask, they estimate, you approve before work starts. Pause if: changes are free but unlimited with no process. That ends in missed dates.
Good: payments tied to demos you have accepted. Pause if: most of the money is due upfront with nothing delivered, or the final payment has no acceptance step.
| Payment plan | What it protects | What it risks |
|---|---|---|
| Small deposit, then per accepted demo | You pay for what you saw | The studio waits for your sign-off, so respond quickly |
| Half upfront, half at the end | Simple to understand | You carry risk until the end |
| Monthly retainer | Steady team for steady work | You need a clear list of what the month buys |
If you want to see how we would answer these on your project, you can book a scoping call with Mctaba Labs.
Good: you own the code once you have paid, and your business opens the hosting, domain, app store and payment accounts. Pause if: the studio holds the accounts "for convenience", or the contract is silent on code. This is the most common hard lesson for first-time buyers.
Good: a named upkeep plan: bug fixes, monitoring, a response time and a price. Pause if: support depends on one person's phone number.
Good: a handover that includes the code repository, deployment notes and credentials, within a set number of days of your request. Pause if: leaving means a fee, or means starting from nothing.
Ask for the answer to question 10 to be written into the contract. The clause can be short: the client owns the code on payment, third-party libraries stay under their own licences, and all accounts are in the client's name.
A reference call is worth more than any pitch. Ask the studio for two past clients and call both, preferably one whose project ran late.
Listen for pauses. A client who says "yes, but" is giving you the useful half of the answer. Write down the "but" and put it to the studio in your next meeting.
If the studio cannot name a single client who will take your call, treat that as a red flag from the list above, however polished the slides look.
None of these alone is proof of a bad studio. Two or three together is a reason to walk away.
We are not lawyers. For anything large, have a lawyer read the contract. These are the points we would look for as a buyer.
Two payment questions matter for any app that takes money. Ask who holds the Safaricom or Paystack account (it should be yours) and who receives the settlement. Our note on a checkout for M-Pesa, MoMo and cards shows why.
Sometimes the answer is no. If a ready-made till does your job, buy it. Our guide to custom software or off-the-shelf POS gives a decision table. If you will build software every month for years, an employee may cost less than a studio, as hiring an in-house developer or a studio shows.
When a studio is right, use the twelve questions above on every candidate, and score the answers side by side. The studio that is comfortable with specific, written answers is usually the one that will deliver the same way.
We would like you to ask us all twelve. If you are speaking to us, bring the one-page brief from the budget guide and your list of questions. We will answer in writing and tell you where a different choice suits you better.
To start, book a scoping call with Mctaba Labs.
Our Nairobi studio designs and ships software and AI products for African businesses. Tell us what you need on a free scoping call.
Book a scoping call