Pivot or persevere
When the metrics aren't lying anymore.
Pivot-or-persevere is the highest-variance call a founding team makes, and the one most distorted by who speaks first. The founder's conviction, the loudest investor's last comment, or a single bad week of metrics can anchor the whole room. Running it async forces every co-founder and early operator to commit to a threshold privately before they hear anyone else's, so the decision reflects the team's real read instead of its most confident voice.
Should we pivot the product, or push another two quarters with the current bet?
Surface the data each contributor sees as decisive — not the rollup, the raw numbers. Note any retention cohort that's clearly working and any segment that's clearly not. The aim is to get every person's threshold for 'persevere' on the table before we agree.
What each contributor weighs
Ask everyone to put these on the table privately, before they see another contributor's position.
- The raw retention and engagement numbers you personally find decisive — not the board-deck rollup.
- Your honest probability that the current bet hits its next milestone, stated as a number.
- What you would need to see in the next 60 days to flip your vote.
- The opportunity cost: the strongest alternative bet the team would pursue instead.
- Three quarters of flat growth
- Founder gut says one thing, dashboards say another
- Investor check-in next month
How this decision usually goes wrong
- Sunk-cost framing — counting the months already spent as a reason to keep going.
- Anchoring on the founder's gut before the raw data is on the table.
- Confusing a distribution problem (nobody knows the product) with a product problem (the product is wrong).
- Deciding in a single emotional meeting right after a bad metrics week.
How to run this decision in 48 hours
- Open a room seeded with the question above. Set a 48-hour deadline.
- Paste the opening context — keep it short. Engineers will not read more than 300 words.
- Invite by email. Each contributor submits position + reasoning + risks privately, before seeing anyone else.
- Read the synthesis. Claude maps consensus, dissent, blind spots, and confidence.
- Decide and log. The synthesis is the decision-log entry — no separate write-up.
A synthesis that separates the people persevering on evidence from those persevering on hope, surfaces the one cohort everyone overlooked, and names the explicit 60-day kill criteria the team will hold itself to.
Related reading
- Async vs sync engineering decisions: how to choose
- Why every distributed engineering team needs a decision log