Improve These 8 Decisions and Build Great Products Faster

Teresa Torres argues that product builders make constant decisions, but human judgment is error-prone, so teams should deliberately design decision-making processes. She identifies eight critical decisions in building products: setting goals, generating ideas, evaluating ideas, experimenting, prioritizing, building product, measuring success, and communication. For each, she highlights common failure modes such as ad-hoc goal setting, brainstorming in groups, untested intuition, misreading Agile as building the easiest things first, and treating engineering as a handoff. The piece matters because it reframes product work as a set of repeatable processes rather than one-off judgment calls, and it previews a series of deeper follow-ups.

01Key takeaways

• Treat recurring product choices as designable processes instead of relying on in-the-moment intuition.
• Set goals deliberately, since a narrow, well-chosen goal focuses the team while a bad one wastes effort.
• Generate diverse ideas, and gather individual ideas before group discussion to avoid brainstorming's weaknesses.
• Write evaluation criteria in advance and apply them consistently so everyone can contribute and trust the outcome.
• Prioritize building the smallest test that validates your biggest assumption, not the easiest feature.
• Keep talking with engineers throughout the build, treating requirements as the start of a conversation.

02Key sections

Why decisions go wrong

Drawing on the Heath brothers and Kahneman and Tversky, Torres lists common cognitive biases and villains that distort decisions. She argues that research also points to ways of designing better decision processes.

Goals and idea generation

A good goal focuses effort, but bad goals are costly. Torres notes that more diverse ideas improve outcomes and that individuals generate better ideas than brainstorming groups.

Evaluating, experimenting and prioritizing

Clear evaluation criteria create procedural justice and engagement. Experiments need deliberate choice and team standards, and prioritization should target the biggest assumptions rather than the easiest work.

Building, measuring and communicating

Product design should continue as an ongoing conversation with engineers rather than a handoff. Teams should measure outcomes tied to assumptions, avoid vanity metrics, and communicate expectations to internal and external audiences.

03From the post

“What color should this button be? We can only do x and y, but not z, before the deadline. Can we live with that? Sales requested a, b, and c. Which should we do? The CEO had a new idea and asked if it’s worth pursuing. As product builders,”

04Frameworks mentioned

Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.