There's a specific moment in every high-growth ecommerce brand where the numbers stop making sense. Revenue is at an all-time high. The acquisition team is celebrating. Board decks look incredible.

But the CMO has a feeling they can't quite explain. Q4 feels harder than it should. The repeat purchase numbers look soft. Customer support tickets are up. Discounting keeps creeping into more campaigns.

I've seen this exact scenario three times in the past two years. And every time, the answer was hiding in a report that nobody in the leadership meeting had pulled.

It's a cohort table. Not a dashboard metric. Not a retention percentage averaged across the entire customer base. A table that shows repurchase behavior broken out by the month each customer was acquired.

Most ecommerce brands don't build this. They track aggregate repeat purchase rate, maybe segment by channel, and call it a day. But aggregate repeat purchase rate is one of the most dangerous numbers in ecommerce because it blends every customer you've ever acquired into a single figure.

That's like averaging the grades of every student who ever attended a university and using it to evaluate this year's freshman class.

I audited an apparel brand last year that had scaled aggressively in the first half of the year. New customer acquisition went from around 8,000 per month in January to nearly 22,000 by May. Revenue looked phenomenal.

But when we built the cohort table, the story fell apart.

Customers acquired in January had a 14.5% repurchase rate within their first 60 days. February was similar, around 15%. Healthy for mid-market apparel.

By March, when acquisition started ramping, repurchase dropped to 12.8%. April, 12.1%. May, the peak month, came in at 11%.

A few percentage points. Doesn't sound catastrophic. But here's what made this different from a normal seasonal dip.

Repurchase rate kept falling even after acquisition volume came back down.

By August, they'd pulled back to 14,000 new customers a month. Half the May peak. But the August cohort repurchased at just 10.2%. September was 9.4%. By December it was under 7%.

The assumption was always "we scaled fast, quality dipped a little, it'll normalize when we pull back." It didn't normalize. The damage was structural.

This is what most growth teams miss entirely. When you scale acquisition hard and fast, you're not temporarily lowering customer quality. You're changing the business underneath.

You're pulling from wider, colder audiences. The high-intent buyers who convert from organic and warm paid are a finite pool. Scaling 3x means reaching people who bought because the ad was persuasive, not because they had genuine purchase intent.

At the same time, you're flooding your lifecycle systems with low-quality data. Your email flows, your segmentation, your product recommendations are all learning from a customer base with a much higher percentage of one-time buyers. The system starts optimizing for the wrong people.

And then there's the part that never shows up in any report. Scaling acquisition almost always involves heavier discounting. First-time codes, flash sales, aggressive retargeting. You attract customers whose primary buying trigger was price. Price-driven buyers don't come back at full margin. They wait for the next sale or they don't come back at all.

That's why repurchase doesn't recover when you pull back. The customers you acquired during the ramp are already in your base. Their behavior is already dragging the averages. And your lifecycle system has already adapted to serve them.

In May, this brand's gross revenue hit $1.4M. Record month. New customer revenue was $1.2M. Repeat purchase revenue was $190K.

By November, gross revenue had dropped to $450K. New customer revenue was $390K. Repeat was $42K.

That's not a seasonal dip. That's a business that spent six months buying customers who never came back. The entire revenue line was being carried by acquisition spend, and when acquisition pulled back, there was almost nothing underneath it.

We calculated what the brand lost. If the May through October cohorts had repurchased at January's rate (14.5% instead of the 7 to 11% they actually delivered), the brand would have generated roughly $340,000 more in repeat revenue over six months. No additional acquisition spend. No new campaigns. Just customers coming back a second time.

That $340K wasn't a rounding error. It was the difference between a profitable second half and a cash-negative one. And the first time I showed that number to the CEO, he went quiet for about ten seconds and then said, "So our best quarter was actually our worst quarter?" Which, in terms of what it did to the customer base, was exactly right.

There's a secondary cost that doesn't appear on any dashboard. When your customer base is full of one-and-done buyers, everything your retention team does gets harder.

I've watched retention teams spend six months trying to re-engage customers who were acquired on a 40% off flash sale and never had any product affinity to begin with. The campaigns don't work. Results stay flat. Leadership starts questioning whether retention is even worth investing in.

The problem was never retention strategy. It was acquisition quality. And that distinction only becomes visible when you look at cohort data.

You can't un-acquire bad customers. But you can stop the pattern from repeating.

Build the cohort table. Each row is an acquisition month. Columns track repurchase rate at 30, 60, and 90 days. Two hours in a spreadsheet if your data is clean. If you can't build it, that's its own signal.

Set a repurchase floor. Decide what minimum 60-day repurchase rate is acceptable. For mid-market apparel, anything below 10% should trigger a review of what changed in acquisition that month. Channel mix, creative, discount depth, audience targeting.

Report acquisition quality alongside volume. Right now, most brands report new customer count and new customer revenue. Start also reporting the 60-day repurchase rate of each month's cohort. If volume is going up but repurchase is going down, you're buying one-time customers at increasing cost.

Stop measuring retention teams on aggregate repeat purchase rate. They can only work with the customers acquisition gives them. Measure them on performance within cohort. Did the March cohort repurchase at a higher rate than it would have without intervention? That's a fair question. "Why is overall repeat down?" is not.

If you're heading into annual planning or a post-holiday review, answer one question before that meeting: is each new month's cohort repurchasing at a higher, lower, or equal rate to the one before it?

Export your customers by acquisition month. Count how many made a second purchase within 60 days. Calculate the percentage for each month.

If you see three or more consecutive months where the rate is declining, you're in cohort compression. And the fix isn't more retention campaigns. It's a conversation with acquisition about who they're bringing in the door.

If the decline started during a period where you scaled volume hard, you already know the cause. The question is whether anyone in the business is willing to connect that growth quarter to the retention problem they're trying to solve right now.

What is cohort compression in ecommerce?

Cohort compression is when each new month's group of acquired customers repurchases at a lower rate than the previous month's group. Revenue can continue growing because acquisition volume increases, but per-customer value is declining underneath.

The aggregate metrics like total repeat purchase rate won't show the deterioration until the newer, weaker cohorts are large enough to drag down the averages. By then, the damage is typically 6 to 12 months deep.

Why does repurchase rate drop when ecommerce brands scale acquisition?

Three factors compound during rapid acquisition scaling. First, the brand pulls from wider, colder audiences with less purchase intent. Second, heavier discounting attracts price-driven buyers who don't return at full margin.

Third, lifecycle and retention systems learn from a customer base increasingly dominated by one-time buyers, which degrades personalization and re-engagement effectiveness. The critical insight is that repurchase rate often continues to decline even after acquisition volume returns to normal, because the structural damage to the customer base persists.

How do you measure ecommerce cohort quality?

The simplest method is a cohort table where each row represents an acquisition month and columns track repurchase rate at 30, 60, and 90 days. For mid-market apparel brands, a healthy 60-day repurchase rate is 10 to 15%.

Any cohort below 10% warrants a review of what changed in acquisition that month, including channel mix, creative strategy, and promotional depth. The goal is to make cohort repurchase rate a standard metric in growth reviews alongside new customer volume and acquisition cost.