← Archive

Case #009

High ROAS campaigns brought in customers. The dashboard looked profitable. But the cohort data told a different story — customers who never returned.

ROAS vs LTVCohort CollapseMargin Destruction

How the Illusion Builds

Day 1

Ad campaign shows 4x ROAS. Team celebrates. Budget increases.

Day 14

New customers are not returning for a second purchase.

Day 30

Cohort analysis shows 8% day-30 retention across acquired segments.

Day 90

Blended margin per acquired customer is negative. The ads looked profitable. The cohort is not.

ROAS vs LTV:CAC Reality

Margin Gap
ChannelROAS (Day 1)D30 RetentionLTV:CAC (90d)
Paid Social4.2x7%0.68x
Search Ads3.8x11%0.81x
Display5.1x4%0.44x
Email Acq.2.9x24%1.42x

Avg ROAS

4.0x

Looks profitable

Avg D30 Retention

12%

Tells the truth

Avg LTV:CAC (90d)

0.59x

Negative margin

Evidence Detected

High ROAS dashboard

Ads reported strong return on spend — measured at day 1, not day 90.

Low day-30 retention

Most acquired customers never placed a second order.

Cohort collapse pattern

Every new acquisition cohort showed the same disappearance curve.

LTV:CAC below 1x

When measured over 90 days, the customer was worth less than what was paid to acquire them.

Root Cause

The business measured success at the point of first purchase. No system tracked what happened after the ad was paid for. The acquisition funnel was optimized. The retention funnel did not exist.

"

ROAS measures the ad. It doesn't measure the customer.

Case Conclusion

Acquisition Illusion

Solved

Problem

High ROAS campaigns masking low-LTV customer acquisition.

Cause

Measuring acquisition success at day 1, not day 90.

Action

Implement cohort-level LTV:CAC tracking per channel.

Investigator's Takeaway

A 4x ROAS means nothing if the customer was worth 1x. The acquisition looks efficient. The cohort is telling you the truth.

Don't be a victim of the illusion. Let's analyze your cohorts and uncover the truth.

Book a Retention Investigation →