RICE vs ICE: which method and when
How to choose RICE or ICE, avoid the effort/ease trap, and score three features side by side.

In brief
- RICE = (Reach × Impact × Confidence) ÷ Effort. Use when you must defend the order.
- ICE = Impact × Confidence × Ease. Use to sort a long list fast.
- Common trap: Effort is divided in RICE; Ease is multiplied in ICE. Do not multiply Effort in ICE.
- Scores are inputs to a decision, not the decision.
How the two frameworks compare
| Framework | Components | Formula | Best for | Notes |
|---|---|---|---|---|
| RICE | Reach, Impact, Confidence, Effort | (Reach × Impact × Confidence) ÷ Effort | Roadmaps that must hold up in debate | Reach anchors size in real counts. Confidence tempers guesses. Effort in the denominator rewards low effort. |
| ICE | Impact, Confidence, Ease | Impact × Confidence × Ease | Quick triage of many ideas | Simple 1–5 scales. Ease is the inverse of effort. Keep it rough and consistent. |
When to choose which
Choose ICE when you face a long idea list and need a sort by the end of the hour. The scales are light. Everyone can give a 1–5 for impact, confidence, and ease. The result is a good first pass. It tells you where to look next.
Choose RICE when your order will be questioned. RICE asks you to name how many people or events will be reached in a time window. It asks you to write the impact scale and stick to it. It makes you state confidence. It keeps effort in the denominator, so low effort rises.
Tip: Write your RICE impact scale up front. For example, 3 = “high”, 2 = “medium”, 1 = “low”, 0.5 = “minimal”. Then use it on every item.
The mistake that flips your list
In RICE, effort sits in the denominator. In ICE, you multiply ease. If you instead multiply effort in ICE, hard work scores higher than easy wins. That inverts the ranking.
If your team prefers to think in effort, convert it to ease before you score ICE. One simple way: define ease on the same 1–5 scale but reverse it from effort. For example, Effort 5 (very hard) becomes Ease 1 (not easy).
Steps to run this well
Write the goal, time window, and constraints. Example: “Next quarter, grow weekly active use by existing users.”
Use ICE for a fast cut on many ideas. Use RICE if you need a ranking you can defend to peers and partners.
Define impact levels. Define the reach window for RICE. Define ease levels for ICE. Note what each number means.
Have the team score. Ask for a brief reason on each dimension. Keep it to minutes, not hours.
Sort by the score. Then read the top and bottom. Fix clear outliers where a single shaky guess drove the result.
Make the call. Note key assumptions. Store scores in your Prioritisation tool so you can revisit with new data.
Worked example: the same three features, both ways
Three backlog items for an existing product. Goal: increase weekly active use this quarter.
- Feature A: One-click share from the main view.
- Feature B: Improve first-load speed on mobile.
- Feature C: In-app nudge to finish onboarding tasks.
Impact scale: 3 = high, 2 = medium, 1 = low, 0.5 = minimal. Confidence: 1.0 = high, 0.8 = medium, 0.5 = low. Ease: 1–5 where 5 is very easy. Effort in RICE is in person-weeks this quarter. Reach is weekly users or events reached this quarter.
| Feature | Reach | Impact | Confidence | Effort | Score |
|---|---|---|---|---|---|
| A | 8,000 users | 2 | 0.8 | 6 wks | (8000 × 2 × 0.8) ÷ 6 = 2,133 |
| B | 12,000 users | 3 | 0.8 | 8 wks | (12000 × 3 × 0.8) ÷ 8 = 3,600 |
| C | 3,000 users | 2 | 0.5 | 2 wks | (3000 × 2 × 0.5) ÷ 2 = 1,500 |
RICE order: B (3,600), A (2,133), C (1,500).
| Feature | Impact | Confidence | Ease | Score |
|---|---|---|---|---|
| A | 2 | 0.8 | 3 | 2 × 0.8 × 3 = 4.8 |
| B | 3 | 0.8 | 2 | 3 × 0.8 × 2 = 4.8 |
| C | 2 | 0.5 | 5 | 2 × 0.5 × 5 = 5.0 |
ICE order: C (5.0), A (4.8), B (4.8). Tie-break by goal fit or risk.
Notice how the same three features land in a different order. In RICE, B rises due to higher reach and impact even with more effort. In ICE, C rises on ease. That is why you pick the model to match the decision you need to defend.
Checklist before you present the order
- Is the goal and time window clear on the page?
- Did you define the impact and ease scales before scoring?
- Did you keep effort in the denominator for RICE and ease in the multiplier for ICE?
- Are the top items still top if confidence drops a level?
- Have you noted at least one risk or assumption per top item?
- Do you have a simple story for each trade-off in the top three?
Where both can mislead
Neither model knows about a critical dependency, a legal must, or a strategic bet that needs proof. If a must-do item scores low, call it out and place it anyway. If a high score rests on weak input, say so and mark it for discovery work before build.
Re-score when you learn more. Keep a change log so you can explain why an item moved up or down. The clarity helps in planning and in trust with partners.


