How long does it take to build an MVP?
By Bonaventure Ogeto, Founder and CEO, Mctaba Labs
Published
An MVP takes as long as its scope, its approvals and your own replies take. Building one flow is the short part. Fixed waits, such as Google's 14 day closed test for new personal accounts and a payment provider's checks, are where the weeks go. Begin those on day one and ask for a weekly demo.
You already have a date in your head. The end of your leave, a trade fair in Mombasa, the start of the school term, the month your savings run out. So you ask a developer how long the MVP will take, they name a period that fits, and everyone feels better.
That answer was a guess shaped by your hope. Nobody can tell you how long "an MVP" takes, for the same reason nobody can tell you how long "a journey" takes. What they can do is size a written list of work. And what you can do is remove the waiting, which is where most of the calendar goes.
Why is there no standard timeline for an MVP?
Because three different things set the time, and they vary enormously between projects.
The scope. One flow for one kind of user is a small job. Each added flow, user type and rule adds building and, more than people expect, testing.
The waiting. Parts of every project depend on companies that do not work for you: a payment provider checking your documents, an app store reviewing your app, a domain being pointed. This time passes whether or not anyone is coding.
You. How quickly you answer questions, approve designs, supply the words and prices that go on the screens, and resist adding things.
Only the first of the three is about the developer's speed. This is why two founders with similar products and similar builders can finish far apart. One of them had the Paybill and the content ready on the first day and replied within hours. The other did not.
Where does the time in an MVP actually go?
Every build passes through the same six stages. The lengths are yours to find out. The order is not.
- Agreeing the scope. Turning the idea into a one-page list of what a user can do. This is the cheapest stage to spend time on and the most expensive to rush.
- Design. Deciding what each screen shows. For an MVP this should be sketches and a standard layout, not weeks of polish.
- Building the main flow. The part everyone pictures when they think of software.
- Connecting outside systems. Payments, SMS, WhatsApp. The code is small. The accounts, approvals and tests around it are not.
- Testing with real people and real money. Finding what breaks on a cheap Android phone with a weak connection.
- Release and the first fixes. Putting it where customers can reach it, then repairing what the first users trip over.
Founders tend to budget time for stage 3 and are surprised by all the others. A good builder's estimate names every stage.
What will I be waiting for, and can I begin it early?
Yes, and this is the most useful thing you can do for your timeline. Almost everything below can begin before a line of code exists.
| What you wait for | Who you wait on | Begin it |
|---|---|---|
| A business payment account, such as a Till, a Paybill or a payment provider account in live mode | Safaricom, your bank or the provider, who will ask for business documents | The day you decide to build |
| Business registration and KRA PIN, if you do not have them and the provider asks | The government registration services | Before you look for a builder |
| Domain name and hosting account, in your own name | You | The first day. It is a short job |
| Words, prices, photos, terms and privacy text for the screens | You | During design, not at the end |
| Sender name for SMS, or approval for business WhatsApp messages | The messaging provider | As soon as you know you will send messages |
| Apple developer membership | Apple, which asks organisations for a D-U-N-S Number to verify identity | Only if you truly need an iPhone app in version one |
| Google Play testing period | Google, which requires new personal accounts to run a closed test first | Only if you truly need an Android store app in version one |
The last two rows deserve a second look. Apple's enrolment page sets out the identity checks for organisations. Google's Play Console help says a personal developer account created after 13 November 2023 must run a closed test with at least 12 testers opted in for at least 14 days in a row before the app can go to production. Those are fixed waits that no developer can shorten. A website that opens from a WhatsApp link has neither, which is the main reason to make it your first version.
On payments, using a provider that already supports M-Pesa (Paystack lists it for Kenyan businesses) usually means less to build than a direct connection to Safaricom's Daraja platform. Either way, the provider will want to verify your business before real money moves, so open the account first.
How do I work out my own timeline?
Build it from the bottom up with your builder, in four steps.
Step 1. Ask for working days per item. Send your scope sheet and ask the builder to put a number of working days beside each line, including testing. Do not accept one date for the whole thing.
Step 2. Add the waiting lanes. List each approval from the table above with the date you applied. These run beside the build, not after it.
Step 3. Add your own feedback time. Decide how fast you will respond, and write it down as a promise: for example, "I answer questions the same day and review each demo within two days."
Step 4. Add room for what you have not thought of. Something will be missed. Agree a margin with the builder and state it openly instead of hiding it inside the estimates.
Here is the layout, filled in with made-up round numbers so you can see how it reads. They are an illustration of the method, not typical durations.
| Item | Builder's estimate (working days) | Waits on | Done when |
|---|---|---|---|
| Sign-up with phone number | 2 | Nothing | I can register on my own phone |
| Choose a service and time | 3 | My price list | I can make a booking |
| Pay deposit by M-Pesa | 3 | Live payment account | I pay KES 10 and see it recorded |
| Confirmation by SMS | 1 | Sender name approval | The message arrives on two networks |
| Admin list of bookings | 2 | Nothing | I can see today's bookings |
| Testing with five outsiders | 3 | Me finding five people | All five finish without help |
| Margin for the unknown | 3 | Nothing | Used or handed back |
Add the days and you have the building time. Then look down the "Waits on" column. Any row that waits on an approval you have not applied for is where your real launch date will come from.
What makes an MVP late?
The usual causes, most of them avoidable:
- The scope grew. "Can we also add" is the most expensive sentence in software. Each addition is small, and together they can double the job. Keep a "later" list and put every new idea there.
- Feedback was slow. A builder who waits four days for your answer does not sit idle. They pick up other work, and getting back to yours costs more time again.
- Content was missing. The screens are ready but nobody wrote the terms, set the prices or took the photos.
- Approvals began late. The product is finished and cannot take money.
- The builder is part-time on your project. Ask plainly how many hours a week are yours.
- Nobody tested on real phones until the end. So the problems all arrived in the last week.
Four of those six are in your hands.
How can I shorten the time without lowering quality?
Cut what the product does, not how well it does it.
Remove a flow. Do sign-ups by hand. Send statements yourself on Fridays. Launch to one estate, one school or one market before the whole town. Choose the website over the store app. Use the payment provider your builder has connected before.
Do not shorten testing of the payment step, and do not add developers to a build that is running behind. Fred Brooks made the second point famous in The Mythical Man-Month: new people must be brought up to speed by the people already doing the work, so adding them late tends to make a project later.
How do I know early that the build is slipping?
Ask for a working demo every week, on your own phone, from the first week. Not a report, not a screenshot, not "it is 80 percent done". Something you can tap.
A message you can send every Friday:
Hi Brian, please share this week's link so I can try it on my phone. Which lines on the scope sheet are now done, which are in progress, and is anything waiting on me?
Three answers tell you where you stand. A link that works and a short list means you are on track. A link with less than promised, plus a clear reason, means adjust the plan together. No link for two Fridays running means stop and have a direct conversation before another payment is made.
The demo habit also fixes your own side. When you can tap the product every week, you give feedback while it is cheap to act on, and you stop imagining features, because you can see how much work each small one already is.
Key takeaways
- Ask a builder for working days per item on your scope sheet, not a single delivery date.
- Begin every approval that depends on another company on the first day, before any code.
- A website that works on a phone avoids app store steps entirely.
- Answer your builder's questions within a day. Slow feedback is a delay you control.
- Insist on a working demo every week. A missed demo is the earliest warning you will get.
- When the date is at risk, cut scope. Adding people late makes a build slower.
Frequently asked questions
- Can an MVP be built in a weekend?
- A manual version, with bookings taken by message and written in a notebook, can be running in a weekend, and so can a simple no-code page. A coded product with real payments usually cannot, because the provider's checks and proper testing with real money do not fit inside two days no matter how fast the builder types.
- Why does connecting M-Pesa take longer than the developer said?
- The code is rarely the slow part. Going live needs a business payment account, documents the provider must approve, and tests with real money on real phones. If those steps begin only when the code is finished, the product sits ready and unable to take payment.
- Should I set a launch date before the build begins?
- Set a date for the first working demo, not for the launch. Once you have seen two or three weekly demos arrive on time, you will know the builder's real pace and can name a launch date you are able to keep.
- Does a bigger team build an MVP faster?
- Not usually. An MVP is small enough that one or two capable people can hold all of it in their heads. Extra people need briefing and coordination, and joining a project that is already late tends to slow it further.
- How long should I test before launching?
- Long enough for five people outside your circle to complete the whole flow, including payment, on their own phones without your help. If any of them gets stuck, fix that and test again with someone new.
Sources
- The Mythical Man-Month: Essays on Software Engineering, Frederick P. Brooks Jr. (book, 1975). Checked on 19 November 2026.
- App testing requirements for new personal developer accounts, Google Play Console Help. Checked on 19 November 2026.
- Enrollment, Apple Developer Program. Checked on 19 November 2026.
- Payment Channels, Paystack Developer Documentation. Checked on 19 November 2026.
