Breaking the rules of growth: Why Shopify bans KPIs, optimizes for churn, and prioritizes intuition
Executive Summary
Lenny's Podcast episode, sponsored by Explo and Dovetail, in which Lenny Rachitsky talks with Archie Abrams, VP of product for growth at Shopify leading over 600 people, about why Shopify breaks the standard growth rules. Shopify, roughly 10 percent of US e-commerce with 235 billion dollars of GMV in 2023, bans KPIs on core product teams - decisions run on taste, intuition and a hundred-year vision from CEO Tobi Lutke. Abrams explains why the growth team effectively optimizes for churn: the mission is more entrepreneurship on the internet, so they lower barriers and let many first attempts fail, knowing some merchants become the next Allbirds or FIGS. He shows the funnel trap - teams boost their own conversion rate by making the previous step harder, which shrinks the whole - and why getting more people in the door always hurts conversion rates while helping the business, plus long-term holdouts on every experiment revisited years later. PARTIAL COVERAGE: this summary highlights the philosophy; org design and experiment mechanics are not covered. These are Abrams's own views and company practices, not independently verified.
Key Takeaways
KPIs are banned
Core product teams decide on taste and a hundred-year vision, not metric targets.
Optimizing for churn
Lowering barriers means many merchants fail fast - and a few become the next Allbirds.
The funnel trap
The easiest way to raise one step's conversion is to make the previous step harder - a net loss.
More in the door
Widening the top always hurts conversion rates while growing the total that matters.
Long-term holdouts
Every experiment keeps a holdout group and gets revisited years later.
Finland-sized GMV
Shopify did 235 billion dollars of GMV in 2023, roughly the economy of Finland.
Six hundred in growth
Abrams leads 600-plus across product, design, engineering, data, ops and growth marketing.
The Conversation Breakdown
KPIs are banned
⏱ Untimed topicCore product teams decide on taste and a hundred-year vision, not metric targets.
Optimizing for churn
⏱ Untimed topicLowering barriers means many merchants fail fast - and a few become the next Allbirds.
The funnel trap
⏱ Untimed topicThe easiest way to raise one step's conversion is to make the previous step harder - a net loss.
More in the door
⏱ Untimed topicWidening the top always hurts conversion rates while growing the total that matters.
Long-term holdouts
⏱ Untimed topicEvery experiment keeps a holdout group and gets revisited years later.
Finland-sized GMV
⏱ Untimed topicShopify did 235 billion dollars of GMV in 2023, roughly the economy of Finland.
Six hundred in growth
⏱ Untimed topicAbrams leads 600-plus across product, design, engineering, data, ops and growth marketing.
Core Frameworks & Ideas
Funnel-whole thinking - Abrams's antidote
Judge changes on total successful users, not any single step's conversion.
Churn-tolerant growth - Shopify's model
When the mission is more attempts, cheap failure is a feature, not a leak.
Holdout memory - their experiment rule
Keep long-term holdouts and re-litigate wins years later.
Actionable Next Steps
- ✓Check which team metric could be gamed by hurting a neighbor's step.
- ✓Run one experiment with a permanent holdout.
- ✓Ask what your equivalent of lowering the drawbridge would be.
- ✓Review one win from two years ago and see if it still holds.
Notable Quotes
“it's actually almost always easier to just make it harder to do the thing right before your step in the funnel”
“can we lower the barriers to getting started”
“increase the amount of Entrepreneurship”
The One Thing to Remember
Shopify's growth heresy: ban KPIs, lower the drawbridge, accept churn - and never let a funnel team win its own metric at the expense of the whole.