Cohort Retention Reporting for D2C — Beyond Simple ROAS

Aggregate ROAS hides everything important. It mixes new customers and repeat customers, low-AOV and high-AOV, profitable cohorts and unprofitable ones. Cohort retention reporting separates the layers — and reveals which customer groups are actually building your business versus which are flattering your dashboards.

Quick Answer

Cohort retention reporting groups customers by their acquisition month and tracks revenue from each group over time. A healthy D2C cohort shows 25-40% repeat purchase within 90 days, with cumulative gross profit exceeding CAC by month 6-9. Aggregate ROAS can't tell you this; cohorts can.

What a cohort retention table looks like

Acquisition Month	Customers	Month 1 Revenue/Customer	Month 3	Month 6	Month 12	Total LTV
January 2025	1,200	₹1,400	₹1,950	₹2,400	₹3,200	₹3,200
February 2025	1,450	₹1,380	₹2,100	₹2,580	₹3,400	₹3,400
March 2025	1,650	₹1,500	₹2,200	₹2,700	₹3,600	₹3,600
April 2025	1,800	₹1,420	₹2,050	₹2,520	—	(incomplete)

Each row is a cohort. Each column is months since acquisition. The diagonal shows aging customers; the bottom-right is incomplete (those cohorts haven't matured yet).

Why cohort tables matter for D2C

Three things only cohorts can show you:

1. Whether new customers are getting better or worse over time If January's cohort hits ₹3,200 LTV but April's cohort is on track for ₹2,500, your acquisition quality is declining. Could be channel mix, creative fatigue, or scaling into worse audiences.

2. When customers actually pay you back Aggregate metrics blend everyone. Cohorts reveal exact payback timing — "Customers acquired in January took 5.5 months to break even; April customers are at 7 months and counting."

3. Which channels deliver lasting customers Segment cohorts by acquisition channel (Meta vs. Google vs. organic). Some channels look great on Day 1 ROAS and terrible on Day 180 cohort LTV.

How to build a cohort table

Step 1: Define the cohort Group by acquisition month (or week if volume is high). First-order date defines cohort membership.

Step 2: Track revenue per customer per month For each cohort, calculate average revenue per customer in months 1, 2, 3, etc. Use gross revenue (not gross profit) for the base table; add a margin column for profit cohorts.

Step 3: Track cumulative LTV Each column is incremental; cumulative LTV is the running sum. This is what compares to CAC.

Step 4: Compute payback Cumulative gross profit / customer ÷ CAC. The month it crosses 1.0 is the payback month.

Step 5: Visualize A heatmap with cells colored by intensity makes patterns obvious. Newer cohorts performing worse than older = red flag.

Cohort segmentation that surfaces hidden truths

Segment cohorts by:

• Acquisition channel: Meta vs. Google vs. organic vs. email

• First product purchased: Hero SKU vs. category-introductory SKU

• AOV bracket: under ₹999 vs. ₹999-2,000 vs. ₹2,000+

• Geography: Tier 1 vs. Tier 2-3 cities

• First offer: Full price vs. 10% off vs. 20% off

A Meta cohort acquired with a 20% off offer typically has 35-50% lower 90-day repeat rate than a Meta cohort acquired at full price. Aggregate numbers hide this; cohort segmentation surfaces it.

What healthy retention looks like

Indian D2C cohort retention benchmarks (% of customers who repurchase):

• Day 30: 12-22%

• Day 60: 22-35%

• Day 90: 30-45%

• Day 180: 40-55%

• Day 365: 50-65%

If your Day 90 is below 25%, retention is the bottleneck — not acquisition.

Common cohort patterns and what they mean

Pattern 1: Steady cohorts, slight decline Normal — newer cohorts haven't fully aged yet. Wait 6 months before judging.

Pattern 2: Sharp decline in newer cohorts Acquisition quality dropping. Audit recent channels, creatives, audiences.

Pattern 3: Older cohorts plateau early, newer ones still growing Likely a product improvement or onboarding change is working.

Pattern 4: Cohorts diverge by channel One channel (often Meta retargeting or email) is producing way better LTV. Reallocate budget.

When to start tracking cohorts

The day you cross 100 first-time customers/month. Below that, monthly cohorts are too small to be meaningful. Use weekly cohorts when monthly is too lumpy.

Cohort tracking should be a weekly review for the founder, not a quarterly board exercise.

Common Questions

How is cohort analysis different from LTV calculation?

LTV is a single number (average). Cohort analysis shows the distribution — which acquisition month delivers what LTV. You need both, but cohorts catch problems LTV averages away.

What tools build cohort retention reports for D2C?

Shopify's built-in cohort reports are usable for small brands. Tools like Lifetimely, Glew, and Triple Whale add depth. Bach AI auto-generates monthly cohort tables with channel segmentation.

Should I include discounted purchases in cohort revenue?

Yes — that's gross revenue. For profit cohorts, apply your blended gross margin (which already accounts for average discount). Don't try to filter discounts out manually.

How long should I wait before judging a new cohort?

At minimum 60-90 days for consumable categories, 180 days for considered purchases. Earlier signals are noisy.

What to do next

Try Bach AI on your account at app.wittelsbach.ai. Bach AI builds your cohort retention table automatically — by month, channel, AOV bracket, and first-purchase offer — and flags when a cohort is breaking pattern before it shows up in aggregate metrics.