your Conversion Rate Is Lying to You
Your conversion rate went up last quarter. Everyone's happy. The checkout redesign worked, the guest checkout flow shaved off a few seconds, the new payment button converted better in testing. Great quarter, right?
Maybe. Or maybe you just moved the problem somewhere nobody's looking.
I've seen this pattern enough times now that I stopped believing conversion rate on its own means anything. It's not that it's a bad metric. It's that it's an incomplete one, and merchants are treating it like the whole scoreboard.
What conversion rate actually measures
Conversion rate captures one moment. A visitor becomes a buyer, or they don't. That's it. It's a snapshot of intent turning into action, taken at the single most optimized, most tested, most obsessed-over point in the entire customer relationship.
Which makes sense why everyone's staring at it. It's easy to A/B test. It moves fast. You can show a chart to your boss on Friday and point at the line going up.
But a customer relationship isn't a moment. It's a duration. And the thing that actually determines whether that customer was worth acquiring doesn't show up in the first ten minutes. It shows up over the next ten months.
That's net revenue retention. And it's playing an entirely different game.
Two metrics, two different jobs
Initial conversion is optimized by removing friction: fewer form fields, saved cards, one-click checkout, guest checkout so nobody has to make an account just to buy a candle.
Net revenue retention is optimized by something almost opposite: durability. Does the card on file still work in month four? Does the customer notice when they get charged? Do you catch the failed payment before it becomes a cancellation? Is the person who converted in ten seconds actually the person who sticks around?
These are different problems, solved by different teams, on different timelines, with different incentives. And most companies only have one team actually getting rewarded for their number. You can probably guess which one.
The part that should worry you
Here's the uncomfortable version of the thesis. Some of the friction you're removing from checkout wasn't just friction. It was doing work you didn't notice, filtering for buyers who were actually going to become customers instead of buyers who were going to bounce off the relationship in six weeks.
Rip that friction out, and yes, more people convert. But you haven't necessarily grown the business. You might have just moved the leak from "didn't buy" to "bought, then quietly disappeared," and relocated it to a part of the funnel nobody on the growth team is watching.
Nobody throws a party for revenue that didn't churn. There's no dashboard alert for a renewal that just... didn't happen. Involuntary churn from an expired card or a declined payment doesn't look like a failure. It looks like nothing. It looks like silence. And silence doesn't show up in a conversion rate.
What this actually means for how you build
I'm not arguing against reducing checkout friction. I'm arguing against treating it as the whole job.
If your growth strategy starts and ends at the moment of purchase, you're optimizing ten minutes of a relationship that's supposed to last years. The real question isn't "did they convert." It's "did they stay converted." And that second question depends on things that have nothing to do with your checkout page: retry logic on failed payments, how you handle expiring cards, whether your dunning emails actually reach anyone, whether your pricing page set the right expectations in the first place.
Growth isn't a checkout problem. It's a lifecycle problem wearing a checkout costume.
So the next time someone shows you a conversion rate chart going up and to the right, ask the follow up question: up and to the right compared to what? Compared to six months from now, are these the same customers, or did you just get better at getting people through the door on their way back out?
Should we talk about dunning next? Seems like a timely fit ;)
-Azsha