Weekly Retention Review Template

Q: How do I set repeat rate thresholds if I don't have historical data?

Start with a 4-week baseline (measure repeat rate for the last 4 complete cohorts). Set the threshold at -2 percentage points from that average. After 8-12 weeks of data, refine the threshold based on observed variance and business impact. For new brands with <4 weeks of data, use industry benchmarks (typically 20-35% repeat rate for DTC at day 14-30) as a temporary floor, then replace with actual baseline.

Q: Should I review retention for all customer segments or just high-value ones?

Review all segments that represent >10% of weekly orders. This typically means top 2-3 traffic sources and top 2-3 product categories. Smaller segments can be reviewed monthly. Segment-level review catches channel-specific or product-specific failures that aggregate metrics would hide.

Core Metrics to Review Weekly

Retention reviews require three metric buckets: repeat rate (percentage of customers who purchase again within a defined window), cohort curves (survival by acquisition date), and segment churn (repeat rate by channel, product category, or customer tier).

The repeat rate window should match the business cycle. For subscription or high-frequency replenishment (weekly or bi-weekly), use a 14-day repeat window. For monthly or seasonal replenishment, use 30 or 60 days. Misaligned windows produce false signals and delay corrective action.

Repeat rate (%) - customers who order again within the cohort window
Cohort retention curve - week 1, 2, 4, 8, 12 repeat rates for each acquisition cohort
Segment repeat rate - repeat rate broken by traffic source, product line, or customer LTV tier
Churn acceleration - week-over-week change in repeat rate (flag if down >2 percentage points)
Return rate - percentage of repeat customers (inverse of churn within repeaters)

Weekly Review Checklist

Execute the review on a fixed day (Tuesday or Wednesday recommended, after weekend order data settles). Assign ownership to one operator; rotate quarterly to prevent blind spots.

The checklist enforces consistency and prevents metric drift. Each item has a pass/fail gate; failures route to a decision tree (see next section).

[ ] Pull repeat rate for the most recent complete cohort (orders placed 7-14 days ago, measured at day 14)
[ ] Compare to the same cohort from 4 weeks prior - flag if down >2 percentage points
[ ] Review segment repeat rates (top 3 traffic sources, top 2 product categories) - flag if any down >3 percentage points
[ ] Check cohort curves for the last 3 acquisition weeks - confirm week 2 repeat rate is 60-80% of week 1 (normal decay)
[ ] Scan for anomalies: product returns, customer service complaints, email deliverability issues, payment failures
[ ] Document findings in a shared template (see template section below)
[ ] Route failures to corrective action owner within 24 hours

Failure Thresholds and Decision Rules

Thresholds define when a metric moves from 'normal variance' to 'investigate.' Set thresholds based on historical baseline and business tolerance, not industry benchmarks.

Decision rules route findings to the right owner and action. A repeat rate drop of 2 percentage points in a 30-day cohort is material; a 0.5 percentage point drop in a 14-day cohort may be noise.

Repeat rate down >2 percentage points from 4-week average - investigate product quality, shipping speed, or email engagement
Segment repeat rate down >3 percentage points - audit that channel's customer profile, traffic quality, or onboarding messaging
Cohort week 2 repeat rate below 50% of week 1 - flag potential product-market fit issue or post-purchase experience failure
Return rate up >5 percentage points - escalate to product and CS; check for defect spike or shipping damage
Churn acceleration (week-over-week decline) for 2+ consecutive weeks - trigger root cause analysis; consider pause on paid acquisition

Root Cause Routing

Once a failure threshold is triggered, route investigation to the owner responsible for that lever. Ownership prevents diffusion and ensures accountability.

The routing matrix maps failure type to owner and typical investigation window (24-48 hours for triage, 3-5 days for corrective action).

Product quality or defect spike - Product/QA owner; check manufacturing, supplier, or packaging changes in the last 2 weeks
Shipping speed degradation - Fulfillment/logistics owner; audit carrier performance, warehouse processing time, and transit delays
Email engagement or deliverability - Email/marketing owner; check send volume, list quality, spam complaints, and unsubscribe rate
Onboarding or post-purchase experience - Customer success/product owner; audit welcome sequence, order confirmation, tracking, and first-touch messaging
Traffic quality or customer profile shift - Paid acquisition owner; audit audience targeting, creative, landing page, and cohort LTV

Weekly Review Template (Spreadsheet or Doc)

Use a standard template to ensure consistency and enable trend spotting over time. The template should live in a shared tool (Google Sheets, Notion, or internal dashboard) and be updated every week at the same time.

Include a summary row for quick scanning and a detailed section for root cause notes.

Date of review | Cohort measured | Repeat rate (%) | vs. 4-week avg | Status (pass/fail) | Segment with lowest repeat rate | Lowest segment repeat rate (%) | Root cause hypothesis | Owner assigned | Target resolution date
Example: 2024-01-16 | Orders 1/2-1/9 | 28% | -1.5pp | Pass | Email traffic | 22% | Possible list quality issue | Email manager | 2024-01-19
Example: 2024-01-23 | Orders 1/9-1/16 | 24% | -4.2pp | Fail | Organic search | 18% | Investigate product reviews, shipping time | Product + Fulfillment | 2024-01-26

Escalation and Cadence

Weekly reviews feed into a monthly retention meeting where trends are reviewed, corrective actions are assessed, and strategy is adjusted. Escalate if two consecutive weeks show failure thresholds or if a single week shows >5 percentage point decline.

The monthly meeting should include product, marketing, fulfillment, and finance to ensure retention levers are aligned.

Weekly review - Tuesday, 30 minutes, one operator
Monthly retention sync - third Thursday, 60 minutes, cross-functional (product, marketing, fulfillment, finance, CEO)
Escalation trigger - two consecutive weeks of failure thresholds OR single week >5pp decline OR segment repeat rate <15%

Common Failure Modes and Prevention

Retention reviews often fail because metrics are misaligned, windows are inconsistent, or ownership is unclear. Anticipate these failure modes and build guards into the process.

The most common failure is comparing cohorts with different acquisition sources or customer profiles without segmenting first. A cohort acquired via paid search will have different repeat rates than one acquired via organic or email.

Metric drift - repeat window changes week to week; prevent by locking the window in the template and auditing it monthly
Cohort contamination - mixing customers acquired via different channels in one cohort; prevent by segmenting before comparison
Delayed data - orders from 3-5 days ago not yet recorded; prevent by measuring only complete cohorts (7+ days old)
Ownership vacuum - no single owner assigned; prevent by naming owner in the template and holding them accountable in monthly sync
Threshold creep - thresholds adjusted after the fact to avoid 'failures'; prevent by setting thresholds in advance and reviewing them quarterly, not weekly

Questions

FAQ

How do I set repeat rate thresholds if I don't have historical data?

Start with a 4-week baseline (measure repeat rate for the last 4 complete cohorts). Set the threshold at -2 percentage points from that average. After 8-12 weeks of data, refine the threshold based on observed variance and business impact. For new brands with <4 weeks of data, use industry benchmarks (typically 20-35% repeat rate for DTC at day 14-30) as a temporary floor, then replace with actual baseline.

Should I review retention for all customer segments or just high-value ones?

Review all segments that represent >10% of weekly orders. This typically means top 2-3 traffic sources and top 2-3 product categories. Smaller segments can be reviewed monthly. Segment-level review catches channel-specific or product-specific failures that aggregate metrics would hide.

What's the difference between repeat rate and retention rate?

Repeat rate is the percentage of customers who purchase again within a defined window (e.g., 28% of customers order again within 30 days). Retention rate is typically used for subscription or cohort-based models and measures the percentage of customers still active at a given time. For DTC e-commerce, repeat rate is the standard metric.

How do I handle seasonal or promotional spikes in the weekly review?

Flag promotional periods in the template and compare cohorts acquired during the same promotional window (e.g., Black Friday cohort vs. prior year Black Friday cohort, not vs. regular weeks). For seasonal businesses, compare cohorts week-to-week within the same season, not across seasons. Document the promotional context so month-over-month trends aren't misinterpreted.

Want this on your account?

Thirty minutes. Bring the number that keeps you up.

Book a demo