Opportunity Sizing · Product Fluency

Opportunity Sizing · 3 min

Situation

“This affects thousands of users.”

A slow export affects 8,000 accounts, but most export once a year. A permissions problem affects 300 admins every week and sometimes blocks new employees from working.

Reach alone can mislead. You also need frequency, severity, and the value of a better outcome.

Mental model

Build a range from explicit assumptions.

Estimate eligible users, the share experiencing the problem, occurrences per period, and value per occurrence. Keep units consistent.

For example, 500 to 1,000 weekly users saving five minutes each suggests roughly 42 to 83 user-hours saved weekly. Those hours are a scale estimate, not revenue and not necessarily time users will spend elsewhere in your product.

Example

Separate user value from business value.

A faster onboarding flow may reduce effort and support tickets. It creates revenue only if it changes a commercially relevant behavior, such as more qualified accounts reaching purchase.

State that link as an assumption. Avoid adding time savings, retention, and revenue into one total when they reflect the same underlying benefit.

Failure case

A precise spreadsheet built on guesses.

A forecast of €184,372 can look authoritative even when adoption is unknown. Show low, central, and high cases with the assumptions that move the result most.

Strategic value can matter even when near-term revenue is modest. Explain the capability or market access it creates rather than hiding it inside an inflated financial estimate.

PM question

What evidence would change the ranking?

If adoption uncertainty dominates the estimate, test demand before refining engineering cost to the day. If every plausible case is small, a detailed forecasting exercise may not help.

Compare opportunities using the same period and definition of affected users. Include the ongoing operational cost, not only development effort.

Remember this

Sizing is a decision aid, not a prediction contest.

Use ranges to expose what must be true for a bet to be attractive. A rough estimate with visible assumptions is easier to improve than false precision.