Before You Build Anything, Check These 3 Numbers

Merchants come with a list. New feature, new page, new app. Sometimes the right answer is that none of it will move the needle until something already broken gets fixed first.

From the engineering side, three numbers make this obvious before a single ticket gets written.

AOV

Average order value. If this is low, the store has a monetization problem, not a traffic problem. No new feature fixes that on its own. Post-purchase flows, smart upsells, coupon logic surfaced on the product page, that’s what actually moves AOV. I’ve built all three, more than once, and the pattern holds every time.

Conversion rate

People landing but not buying. That’s a trust or friction problem sitting somewhere in the stack: checkout UX, page speed, payment options, form complexity. All of it is solvable on the engineering side. But you have to know which one is the actual problem before you touch anything, or you’ll fix the wrong friction and wonder why the number didn’t move.

LTV

Do customers come back? If not, every new customer costs full acquisition price, every time, forever. Email triggers, repeat-purchase flows, loyalty hooks, these are technical implementations with direct business impact. Almost nobody asks for them until the acquisition math stops working, which is exactly when it’s most expensive to start.

Most feature requests are symptoms

A merchant asking for a specific new page or app has usually diagnosed a symptom, not the disease. These three numbers point at the disease. Check them first, and half the roadmap conversation changes before a single line of code gets written.