Opportunity cost

Every backlog is infinite. Capacity is not. Prioritization is the craft of deciding what not to build so that what you do build has a chance to matter. Teams that skip this ritual still prioritize — they just do it by whoever shouted last, whoever closed the biggest deal, or whoever filed the ticket first.

Saying yes feels kind; saying no is the job.

Why this is about saying no

Shipping feature A means you did not ship B, pay down the outage-causing debt, or run the experiment that might 2× activation.

Symptoms of broken prioritization:

• Everything is P0; nothing finishes.
• The roadmap is a chronological dump of sales promises.
• Engineers learn about priorities from Slack, not from a ranked list.
• Tech debt and reliability only get airtime after an incident.
• The same debate repeats every Monday because nothing was written down.

Good prioritization starts from product strategy: a guiding policy that already killed whole categories of work. Frameworks then rank what remains. Without strategy, RICE just ranks a random backlog of requests with false precision — you get a scientific-looking order of the wrong work.

Stakeholders often experience prioritization as personal rejection ("you said no to my customer"). Reframe publicly: we are choosing the highest expected learning or value per week of capacity, given what we believe now. Invite them to improve the inputs (better reach data, a cheaper version) rather than lobby for a higher rank with no new evidence.

Walking through a stolen week

The team has 6 engineer-weeks this quarter. Putting all 6 on unpaid-invoice chasing beats splitting 2/2/2 across invoicing, pretty themes and a slightly faster list page. Every week on themes is a week not spent on collections, where willingness to pay is already proven. That is opportunity cost in one picture — the same arithmetic as scoring problems worth solving.

Common mistakes

• Everything is P0. Then nothing is.
• Frameworks without strategy. You rank the wrong pile scientifically.
• Treating no as personal. Invite better inputs, not louder lobbying.

Check yourself

A two-week "medium" project blocks a high-confidence activation bet. What's the real cost? The delayed growth, not two weeks of salary. That is opportunity cost.

RICE on a backlog with no strategy. What do you get? A precise order of the wrong work. Strategy kills categories first; scores rank what remains.

Key takeaways

• Prioritization is choosing what not to do. Opportunity cost is the real price.
• Strategy kills whole categories; frameworks rank the rest.
• Invite better evidence, not a higher rank with no new inputs.

Further learning

Next: score remaining work with RICE. Map: prioritization.