MVP scope worksheet

The one job

• You can state what the product does in one sentence, without "and"
• You know the single user this first release is for If the answer is "everyone", the scope has not been cut yet.
• You can name the thing they do today instead
• You know why today's workaround is bad enough to switch away from

The shippable slice

• One complete path exists from a user arriving to getting value
• Every feature outside that path is written on a separate list Write the list. Unwritten deferrals come back as scope creep.
• Nothing in the release exists only because a competitor has it
• The release could go to real users even if it looks unfinished

What you are deliberately not building

These are usually the right things to defer in a first release.

• Admin panel — can you use the database or a spreadsheet for now?
• Role and permission system — can everyone be the same role?
• Bulk import/export — can the first users be onboarded by hand?
• Notification preferences — can there be one sensible default?
• Native mobile app — is a responsive web app enough to test the idea?
• Internationalisation — is one language enough to learn what you need?

Proving it works

• You have named the number that tells you this is working Sign-ups are usually the wrong one — pick something closer to value delivered.
• Analytics are instrumented before launch, not after
• You know what result would tell you the idea is wrong
• You have agreed how long you will wait before judging

Constraints worth writing down

• The date, and what actually happens if you miss it
• The budget, and who signs off on changes to it
• Regulatory or data-residency requirements that apply from day one
• Systems this must integrate with, and who owns access to them
• Who can make a scope decision without escalating