The RICE Scoring Prompt

The problem with prioritization is that most teams score features emotionally, not analytically. The loudest stakeholder, the newest idea, or the most visible customer request often gets pushed to the top of the roadmap without a structured evaluation framework.

That’s where the RICE model becomes valuable.

RICE helps product managers evaluate initiatives using four variables:

1. Reach → How many users will this affect?
2. Impact → How meaningful will the change be?
3. Confidence → How certain are we about these assumptions?
4. Effort → How expensive is it to build?

The goal is not to create a mathematically perfect roadmap. The goal is to force clearer thinking, expose weak assumptions, and make tradeoffs visible.

Using Claude for RICE scoring becomes especially useful when you’re evaluating multiple competing initiatives at once. Claude can help estimate assumptions, identify weak reasoning, compare tradeoffs, and surface hidden risks that teams often ignore during prioritization discussions.

Here’s the prompt structure:

For example:

1. A feature may have massive projected impact, but only because the team is overestimating adoption.
2. Another initiative may look smaller, but the assumptions are backed by behavioral data and user research.
3. Some roadmap items score well simply because effort estimates are unrealistically low.

Claude is particularly good at identifying these inconsistencies. It can detect when the logic behind the scoring feels weak, contradictory, or overly speculative.

The real value of this workflow is not the spreadsheet itself. It’s the discussion quality it creates.

Instead of roadmap conversations sounding like:

“This feels important.”

They become:

“We’re assuming this impacts 40% of users, but we only have evidence from a small power-user segment.”

That shift from intuition to explicit reasoning is what makes prioritization significantly more reliable.