IncykitScope Decider
A Keep, Defer or Remove verdict for a candidate feature, plus one clear next step.
Score a candidate feature as Keep, Defer or Remove — and catch the ones you could fake instead of building.
- Approximate effort
- 5–5 min
- Privacy
- Runs in your browser. Nothing is saved or sent anywhere.
Before you start
- Takes about 5 minutes (answer for one candidate feature at a time).
- Nothing is saved or sent anywhere (this runs entirely in your browser).
- You'll get a Keep, Defer or Remove verdict and one concrete next step.
Worked example
For example: scoping Shift Swap, Priya runs "automated calendar sync" through this Tool. It's rated nice-to-have — the product still works without it — and doesn't touch her riskiest question (will people actually use the board instead of texting). Verdict: Defer.
About this tool
Answer four questions about one candidate feature — how necessary it is, how relevant it is to your riskiest open question, how much effort it would take, and whether you could deliver it manually instead — and get a Keep, Defer or Remove verdict plus one concrete next step. Usable anonymously, with no account required.
How the result is worked out
Start from your riskiest open question — the one assumption you're least sure of — and let it drive scope more than anything else. For each candidate feature, separate genuine necessity (the product fails to deliver its core value without it) from nice-to-have (people just wouldn't notice). Weigh build effort honestly, especially for nice-but-cheap features, where most scope creep actually hides. Before committing engineering time to anything, ask whether it could be delivered manually first — a concierge process, a spreadsheet — and reserve real build effort for what you've already confirmed matters.
Who it's for
Founders and small product teams deciding what belongs in a first release.
When to use it
Use once per candidate feature you're unsure about, while scoping a first release.
When not to use it
Not a replacement for the one-page scope itself — it scores one feature at a time, it doesn't produce the full page.