Prioritization, from a 2x2 to research-backed
Five frameworks for picking what to build next, in the order you should learn them. Start with the Impact/Effort Matrix when you need a sketch on a whiteboard. Add numbers with ICE for quick triage, RICE when reach matters, and Weighted Scoring when your team has its own criteria. Use Kano when the question isn't which feature, but which kind. Pick the lightest one that actually produces a decision.
Suggested learning order
Start with the visual sketch, add numbers as the stakes get higher, customize the criteria when generic ones don't fit, then validate with real customers.
- 1
Impact/Effort Matrix
Spot quick wins hiding in your backlog instantly
- 2
ICE Scoring
Make roadmap decisions in minutes, not meetings
- 3
RICE Scoring
Build features that move metrics, not just ship code
- 4
Weighted Scoring
Align competing stakeholders with custom scoring they helped create
- 5
Kano Model
Build features that create raving fans, not just satisfied users
All Prioritize calculators
Build features that move metrics, not just ship code
10 min · beginner
Make roadmap decisions in minutes, not meetings
8 min · beginner
Align competing stakeholders with custom scoring they helped create
15 min · intermediate
Spot quick wins hiding in your backlog instantly
5 min · beginner
Build features that create raving fans, not just satisfied users
20 min · advanced
Go deeper in the Learning Hub
Frequently asked questions
The one that fits the decision. A 50-item backlog with mostly small bets needs ICE or Impact/Effort. A roadmap with cross-team dependencies needs RICE. A multi-stakeholder strategic call needs Weighted Scoring with everyone's criteria on the table. The wrong question is "which framework is best." The right one is "what decision am I making, and what's the minimum rigor to make it well?"
Don't score 100 items. Triage first. Use Impact/Effort to push obvious low-impact and obvious huge-effort items into separate piles. Score what's left with ICE for a quick ranking. Only run RICE on the top of that list, where the cost of being wrong is high. Two passes are faster than one and the output is better, because the easy filters happen visually before you waste time scoring nonsense.
ICE for quick gut-check ranking, growth experiments, or anything you'll iterate weekly. RICE when stakeholders need a defensible Reach number and the decision is quarterly or bigger. RICE was created at Intercom by Sean McBride. ICE came out of Sean Ellis's growth-hacking work. The short version: use ICE first, RICE second, only on what survives. Full comparison at /compare/rice-vs-ice.
Three habits keep scores honest. Anchor the scale with examples ("a 9 on Impact looks like X, like our last big launch"). Score independently first, then compare. Re-score historical items that already shipped, and check whether the score predicted what actually happened. If your top-scored items from last quarter didn't move the metric, your scoring is broken.
The PM owns the final ranking. The team scores. You want at least 3 people scoring independently for items that affect their work. Engineering scores Effort. PM scores Impact and Reach. Designers and CX score Confidence. If only the PM scores, you'll see PM bias every time, and you lose the side benefit of getting the team aligned during the conversation.
Don't score them in the same list. They optimize different things. Carve out a fixed share of capacity for tech debt and prioritize within that bucket separately. Public guidance lands in the 10-20% range, and many teams use 15% as a working number. The only time tech debt and features compete is when debt is causing user-visible pain, and at that point it's not really debt anymore. It's a bug.
Ready to dive into a calculator?
Start with the first lesson in the curriculum or explore the full toolkit.