Skip to content

Customer Longevity Documentation

Menu Location: Reports > Customers > Customer Longevity

Access Level: Manager / Administrator

Last Updated: 2026-03-01

Overview

The Customer Longevity report shows customer retention patterns by tracking how many orders customers continue to receive after their first order. This cohort-based analysis helps you understand customer lifecycle, identify drop-off points, and measure the long-term value of customers who started in specific weekly cycles.

Primary Functions:

  • Track customer retention week by week from first order
  • View cohorts by weekly cycle (WID) and start date
  • Filter by customer tags for segment analysis
  • Analyze how many orders customers receive over time
  • Identify retention patterns and churn points
  • Export data for further analysis

Page Layout

Header Section

  • Info Alert: Explains how to read the report
  • Filter Form: Date range selectors, tag filter, and Build Report button

Filter Controls

  • From Date: Start of date range (defaults to 5 weeks ago)
  • To Date: End of date range (defaults to today)
  • Customer Tag Dropdown: Filter by promotional tags or VIP status
  • Build Report Button: Generate report with current filters
  • Reset Form Button: Clear all filters and return to defaults

Main Content Area

A data table showing:

  • WID#: Weekly cycle ID number
  • Start Date: Week start date for the cohort
  • First Order: Count of customers who received first order that week
  • Subsequent Columns: Count of customers who went on to receive 2, 3, 4... N orders

Report Data & Columns

Column Description Calculation/Source
WID# Weekly cycle ID number From weekly_cycle table
Start Date First day of the weekly cycle Date from weekly_cycle.date
First Order Number of first-time orders in that cycle Count where cust_order.type = BOX_TYPE_FIRST
1 order Customers who received exactly 1 order total Count of customers with box_count = 1
2 orders Customers who received exactly 2 orders total Count of customers with box_count = 2
N orders Customers who received exactly N orders total Count of customers with box_count = N

How to Read This Report:

  • Each row represents a "cohort" of customers who got their first order in that weekly cycle
  • Numbers in each column show how many of those customers went on to receive that many orders
  • Example: If WID 500 shows "100" in First Order and "75" in "5 orders", that means 100 customers started in week 500, and 75 of them have received at least 5 orders

Data Filtering:

  • Only includes completed orders (excludes cancelled orders, status != 5)
  • Only shows first-order-type deliveries (type = BOX_TYPE_FIRST)
  • Optional tag/VIP filtering applies to all order counts

Search & Filtering

Date Range Filters

  • From: Starting week for cohort analysis (default: 5 weeks ago)
  • To: Ending week for cohort analysis (default: today)
  • Purpose: Limits which weekly cycles (cohorts) are displayed

Customer Tag Filter

Select from dropdown to filter by:

  • VIP: Only VIP-status customers (vip_status = 1)
  • Promotional Tags: Any custom promotional tag from promotions_tags table
  • Blank: All customers (no filtering)

Note: Tag filter applies to the customers being counted, not the weekly cycle dates.

Common Use Cases

Use Case 1: Understand Overall Customer Retention

Goal: See what percentage of new customers become repeat customers

Steps:

  1. Navigate to Reports > Customers > Customer Longevity
  2. Leave default date range (or set to last 3 months)
  3. Leave tag filter blank
  4. Click "Build Report"
  5. Look at first few rows (most recent weeks)
  6. Compare "First Order" column to "2 orders", "3 orders", etc.

Example: 

WID#  | Start Date | First Order | 1 order | 2 orders | 3 orders | 4 orders | 5 orders
500   | 2026-01-06 | 100         | 25      | 18       | 15       | 12       | 10
This shows: Of 100 first-time customers in week 500, 25 only ordered once, 18 ordered twice, 15 ordered three times, etc. This means 75% (75/100) placed at least 2 orders.

Use Case 2: Compare VIP vs Non-VIP Retention

Goal: Determine if VIP customers have better retention

Steps:

  1. Set date range to past 6 months
  2. Select "VIP" from tag filter
  3. Click "Build Report"
  4. Export to CSV and save as "vip_longevity.csv"
  5. Click Reset Form
  6. Set same date range
  7. Leave tag filter blank
  8. Click "Build Report"
  9. Export to CSV and save as "all_longevity.csv"
  10. Compare the two datasets in Excel

Analysis:

  • Calculate average orders per cohort in each dataset
  • Look for higher percentages in 5+, 10+, 30+ order columns for VIPs
  • This shows if VIP designation correlates with better retention

Use Case 3: Identify Drop-Off Points

Goal: Find where customers typically churn in their lifecycle

Steps:

  1. Run report for last 6 months (all customers)
  2. For each recent cohort, look across the row
  3. Note where you see significant drop-offs (large decrease between columns)
  4. Export data for graphing in Excel

Example Pattern: If you consistently see drops between "3 orders" and "4 orders", this suggests customers churn after 3 deliveries. Consider interventions at the 3-order mark.

Tips:

  • Recent cohorts won't have data in later columns (not enough time passed)
  • Focus on cohorts from 3+ months ago for meaningful patterns
  • Look for consistency across multiple cohorts

Use Case 4: Evaluate Marketing Campaign Effectiveness

Goal: See if a specific promotional campaign improved retention

Steps:

  1. Note the week(s) when campaign ran
  2. Set From date to 2 weeks before campaign
  3. Set To date to 2 weeks after campaign
  4. Select the promotional tag used for campaign (if applicable)
  5. Click "Build Report"
  6. Compare the campaign week cohorts to surrounding weeks

What to Look For:

  • Higher numbers in First Order column (more new customers)
  • Higher percentages making it to 5+, 10+ orders (better quality customers)
  • Compare retention rates to pre-campaign baseline

Use Case 5: Long-Term Customer Value Analysis

Goal: Understand how valuable customers from different acquisition periods are

Steps:

  1. Set date range to 1-2 years ago (for mature cohorts)
  2. Click "Build Report"
  3. Export to CSV
  4. In Excel, calculate "average orders per customer started" for each cohort
  5. Plot on graph to see trends over time

Calculation Example: 

WID 400: (25×1 + 20×2 + 15×3 + 10×4 + 5×5) / 100 first orders = 2.25 avg orders per customer

Troubleshooting

Issue: Recent Weeks Show Zeros in Later Columns

Symptoms: Most recent cohorts have numbers only in first few columns, rest are blank

This is Normal: Recent customers haven't had time to place multiple orders yet. A customer who started 2 weeks ago can't have 10 orders.

Solutions:

  • Focus analysis on cohorts from at least 3-6 months ago
  • Use recent cohorts only for first-order and 2-order analysis
  • Wait for more data before drawing conclusions

Issue: Report Shows Unexpected Numbers

Check:

  1. Verify date range includes the weeks you want to analyze
  2. Check if tag filter is applied (might be limiting results)
  3. Remember: columns show exact counts (exactly N orders), not cumulative
  4. Verify customer tagging was done correctly if using tag filters

Issue: "No Data" or Very Few Rows

Symptoms: Report generates but shows very few or no weekly cycles

Check:

  1. Expand date range (default 5 weeks might be too narrow)
  2. Remove tag filter to see all customers
  3. Verify you have first-time orders in the selected date range
  4. Check if your business was active during selected period

If Problem Persists: There may not be any first-time customers in the selected date range. Try a wider range or check order data.

  • Customer Totals (customer-totals.php) - Financial summary by customer
  • Customer Engagement (customer-engagement.php) - Activity tracking by customer
  • Orders (cust_order.php) - Detailed order listing and management

Typical Workflow:

  1. Customer Longevity → Identify interesting cohorts → Orders page (filter by week_id to see specific orders)

Permissions & Access

Required Access Level: Manager

Access Level Capabilities:

  • Customer Service: Cannot access
  • Manager: Full access to view and export
  • Administrator: Full access to view and export
  • Kiva Admin: Full access to view and export

Best Practices

Analysis Frequency

  1. Monthly review: Run report first week of month for trailing 6 months
  2. Quarterly deep dive: Export 12-month dataset and analyze trends
  3. Post-campaign: Run specific reports after marketing campaigns end

Interpreting Retention

  1. Healthy retention: 60%+ customers reaching 3+ orders
  2. Concerning pattern: Steep drop after 1st or 2nd order
  3. Excellent retention: 40%+ customers reaching 10+ orders
  4. Mature cohorts only: Use cohorts 6+ months old for retention benchmarks

Data Export Best Practices

  • Export with descriptive filenames: "longevity_2025-Q4_VIP.csv"
  • Create pivot tables in Excel for easier analysis
  • Chart retention curves to visualize patterns
  • Compare cohorts month-over-month for trends

Things to Avoid

  • ❌ Drawing conclusions from cohorts less than 3 months old
  • ❌ Comparing retention across very different time periods (seasonality matters)
  • ❌ Ignoring tag filters when comparing datasets
  • ❌ Assuming low numbers mean bad retention (check percentage of first orders)

Quick Reference Card

Task Action/Location
Run standard retention report Use default dates, no tag filter, click "Build Report"
Check VIP retention Select "VIP" from tag dropdown, build report
View specific campaign cohort Set dates around campaign period, select campaign tag
Export data for analysis Click CSV button after building report
Reset to defaults Click "Reset Form" link
Analyze specific weeks Enter exact From/To dates for WID range

FAQs

How far back should I look for meaningful data?

At least 6 months for good retention analysis, ideally 12+ months for mature cohorts.

Why do numbers not add up across columns?

Each column shows exact counts (exactly N orders), not cumulative. A customer who's ordered 5 times appears only in the "5 orders" column, not in columns 1-4.

What's a good retention rate?

Industry varies, but generally: 60%+ reaching 3 orders is good, 40%+ reaching 10 orders is excellent. Compare to your own historical data for benchmarks.

Can I see individual customer names?

No, this is an aggregate report. For individual customer order history, use the Orders page or Customer Info page.

What's the difference between "First Order" and "1 order"?

"First Order" = total customers who started in that week. "1 order" = customers who only ever ordered once (and stopped).

End of Documentation

For additional help, contact your system administrator or Kiva Logic support.