Insights / Article
How to scope an MVP in the first two weeksWhat ships and what waits
The first two weeks end with four short documents and a line through your feature list. This is how we get there with you.

To scope an MVP, we take two weeks. Week one settles who the product is for and what problem it solves, on one page. Week two maps the core flows and draws a line through your feature list: what sits above it ships first, and what sits below waits, with a written reason. You end with four signed-off documents, the same ones our product strategy service delivers, and the build starts from the last of them.
What you bring to the first call
Bring what you have: a pitch deck, notes from talks with people who might use the product, a list of features, sometimes a quote from another studio. We ask for all of it before kickoff and read it before we meet.
That leaves the first call for what is not written down yet. Who do you picture opening the product on the first day? What do they use today instead? What made you start? We record the call, with your permission, so nobody has to take notes while they listen. A lot of the positioning page later comes straight from that recording.
Week one: the people and the problem
In the first week we try not to propose anything. We look at what your future users do today, which is usually a spreadsheet, a competitor, a group chat, or nothing at all. We read the reviews people leave for the tools they use now. Complaints in reviews tend to be specific in a way that interview answers often are not.
At the end of the week we write the positioning on one page: who the product is for, the problem it solves for them, and why they would move over from what they use now. We keep it to one page on purpose. If it does not fit, we do not understand it well enough yet, and week one is the cheapest time to find that out.
You read it, mark it up and sign it off. Everything after this gets checked against it.
Week two: the flows and the line
With the positioning agreed, we map the core flows. These are the few paths that make the product worth using, drawn end to end with every state in them: the empty screen before there is any data, the error when a payment fails, the person who comes back after a week away.
Drawing the states is where the real size of the work shows. A feature that sounded like one screen often turns into five once you draw what happens when something goes wrong. It is much easier to see that in a diagram than halfway through the build.
Then we draw the MVP line. Every feature goes on one list, and we draw a line through it together. What sits above the line ships in the first version. What sits below waits, and next to each item that waits we write down why.
Drawing the states is where the real size of the work shows.
Three months later, when someone asks why the product has no team accounts yet, the answer is on the page, with the reasoning you agreed to at the time. If the reasoning no longer holds, you move the item up with your eyes open.
The questions we use to scope an MVP
We use a handful of plain questions, and we ask them about every feature:
- Does the main flow break without it?
- Would the first users notice it is missing in their first week?
- Could we do it by hand for the first users, and build it later?
- Does it depend on something we do not have yet, like data from real users?
A feature has to earn its place on the first two. The last two usually move things below the line. Admin dashboards, detailed analytics and most settings screens end up waiting, because for the first users they can be handled by hand or by a short email. Paul Graham makes the same point in Do things that don’t scale: with a small number of users, you can do by hand what you plan to automate later.
We hold our own apps to the same questions. Galance, our Mac app for culling photos, runs fully offline and needs no account. Leaving accounts out took sign-in, password resets and a server out of the first version, and the job people open Galance for needs none of them. Secda, our prayer app, was made for Turkish speakers first. Picking one audience that early answers a lot of questions about language and content before anyone starts building.
The build plan
The fourth document turns the line into an order of work: which slice we build first, what the first weekly build will let you click through, and what depends on what.
We want the first week of the build to end with something you can open, so the plan puts the thinnest working version of the main flow first. Sign-in, settings and polish come later, once there is something worth signing in to. If the main flow is wrong, we would like you to see it in week three of the project, while it is still cheap to change. Our note on why we show you a working build every week covers what each of those builds looks like.
What you have after two weeks
Four documents, each signed off with you:
- The positioning, on one page.
- The core flows, mapped end to end with their states.
- The MVP line, with what waits and why.
- The build plan, in the order we will build it.
They are yours to keep whether or not we build the product together. Inside Zero to launch they carry straight into the Build stage with the same team, and the first working build follows a week later. Branding runs alongside these two weeks, so the product has a name and a look by the time there are screens to put them on. The Launch Path guide shows where these two weeks sit in the whole project.
When two weeks is not enough to scope an MVP
Sometimes the line does not settle. Usually that means the problem is not clear yet: two groups of users want different things, or nobody has asked the people the product is for. When that happens we say so, and we suggest what would make it clear, which is often a week of talking to a few more people before anyone draws a screen.
It is a slower start, and it is still much cheaper than finding the same thing out after eight weeks of building. The product strategy page lists every document the stage produces and what is in each one.



