Cohort Types 2025: Time-Based vs Segment-Ba...

Based on our analysis of hundreds of SaaS companies, cohort analysis is only as useful as the cohorts you define. Time-based cohorts (customers grouped by signup date) answer different questions than segment-based cohorts (customers grouped by characteristics). Using the wrong cohort type leads to insights that don't translate into action. Time-based cohorts reveal whether retention is improving over time—critical for understanding if product changes are working. Segment-based cohorts reveal which customer types retain best—critical for go-to-market and product decisions. The most sophisticated analysis combines both dimensions. This guide explains when to use each approach, how to implement both, and how to combine them for comprehensive understanding.

Time-Based Cohort Analysis

Time-based cohorts group customers by when they signed up—monthly, weekly, or quarterly. This approach answers questions about change over time.

What Time-Based Cohorts Reveal

Time-based analysis shows whether retention is improving or degrading over time. January cohort retained at 80% through Month 6; June cohort retained at 85%. The 5-point improvement suggests product or onboarding changes are working. Without time-based view, you'd miss this trajectory.

Identifying Seasonal Patterns

Some businesses have seasonal retention patterns: Q4 signups during busy season may retain differently than Q1 signups. Time-based cohorts reveal these patterns. Understanding seasonality improves forecasting and resource planning.

Evaluating Product Changes

Launched new onboarding in March? Compare pre-March cohorts to post-March cohorts. Time-based analysis isolates impact of changes by comparing "before" and "after" customer groups.

Standard Time Periods

Monthly cohorts are standard for SaaS—enough granularity to see trends without too much noise. Weekly cohorts for high-volume businesses. Quarterly for enterprise with long sales cycles. Choose based on signup volume.

Trend Visualization

The classic cohort retention table shows months since signup (columns) vs signup cohort (rows). Reading down columns shows retention at specific tenure across cohorts. Reading across rows shows retention curve for each cohort.

Segment-Based Cohort Analysis

Segment-based cohorts group customers by characteristics: plan type, industry, company size, acquisition channel, or behavior. This approach answers questions about customer differences.

What Segment-Based Cohorts Reveal

Segment analysis shows which customer types succeed. Enterprise customers retain at 95%; SMB at 75%. Self-serve signups retain at 65%; sales-assisted at 85%. These insights guide where to focus acquisition and product investment.

Plan and Pricing Segments

Compare retention across pricing tiers: Free→Paid conversion rates, Starter vs Pro vs Enterprise retention. Segment analysis reveals whether higher tiers justify price with better retention or if pricing creates natural selection.

Acquisition Channel Segments

Retention by acquisition source: paid ads, organic search, referral, sales outbound. High-retention channels deserve more investment; low-retention channels may need qualification improvements or should be deprioritized.

Industry and Use Case Segments

Different industries or use cases may have systematically different retention. Horizontal products often discover certain verticals retain 2x better than others. Focus product and marketing on winning segments.

Segment Discovery

Don't assume which segments matter. Test multiple segmentation hypotheses: by plan, by industry, by feature adoption, by company size. Let data reveal which segmentation has strongest retention variance.

Combining Both Approaches

The most powerful analysis combines time and segment dimensions. This reveals not just "which segments perform better" but "is each segment improving over time?"

Two-Dimensional Analysis

Track retention by segment AND by signup time. Enterprise Q1 cohort vs Enterprise Q2 cohort. This shows if segment performance is improving, degrading, or stable—essential for understanding trajectory within segments.

Segment Trend Lines

Plot retention trend line for each segment over time. If SMB retention is improving 2 points per quarter while Enterprise is flat, SMB might deserve more investment despite currently lower absolute retention.

Cohort Size Considerations

Combined analysis creates smaller sub-cohorts. Enterprise + Q1 might be only 20 customers. Small cohorts have high variance. Require minimum cohort sizes (e.g., 30+) for reliable analysis.

Identifying Interaction Effects

Sometimes segment differences only appear in certain time periods. "Marketing-sourced customers retained better than sales-sourced, but only after we changed our qualification criteria in March." Combined analysis reveals these interaction effects.

Complexity Warning

Combined analysis is powerful but complex. 12 monthly cohorts × 5 segments = 60 sub-cohorts. Focus on high-value segments first; don't try to analyze everything simultaneously.

Choosing the Right Approach

Different business questions call for different cohort approaches. Match your analysis method to the question you're trying to answer.

Use Time-Based When

You want to evaluate product changes: "Did new onboarding improve retention?" You need to forecast revenue: "What retention should we expect from recent cohorts?" You're tracking improvement trajectory: "Is retention getting better over time?"

Use Segment-Based When

You're optimizing go-to-market: "Which customer types should we target?" You're making pricing decisions: "Do higher tiers retain better?" You're prioritizing product: "Which use cases succeed most?"

Use Combined When

You're doing strategic planning: "Which segments are improving fastest?" You're evaluating initiatives: "Did our enterprise focus improve enterprise retention?" You need comprehensive understanding for board or investor discussions.

Start Simple

Begin with basic monthly time-based cohorts. Establish baselines. Then layer in 2-3 key segment dimensions. Add complexity only as you exhaust insights from simpler analysis.

Decision Framework

Asking "what's changing?" → Use time-based. Asking "who succeeds?" → Use segment-based. Asking "where should we invest?" → Use both.

Implementation Best Practices

Effective cohort analysis requires consistent definitions, appropriate granularity, and clear visualization. Implementation details matter.

Cohort Definition Consistency

Define cohort assignment rules and apply consistently. If signup date determines cohort, use the same date field always. If segment determines cohort, handle customers who change segments (usually: use original segment).

Retention Metric Selection

Choose retention metric that matches your business: logo retention (customer count), dollar retention (revenue), or both. Track both but be clear which you're analyzing at any moment.

Time Period Alignment

Align cohort periods with business cycles. Monthly for most SaaS. Match billing cycles if relevant. Be consistent—don't compare monthly cohorts to quarterly cohorts.

Visualization Approaches

Retention tables for detailed analysis. Retention curves for visual comparison. Color coding for quick pattern recognition. Choose visualization that matches your audience and purpose.

Documentation

Document your cohort definitions, inclusion/exclusion rules, and calculation methodology. When definitions change, historical comparisons become invalid. Consistency over time matters.

Acting on Cohort Insights

Cohort analysis creates value only when insights drive action. Build workflows that translate analysis into business decisions.

Time-Based Actions

Declining retention trend → Investigate recent changes, prioritize retention initiatives. Improving trend → Document what's working, double down. Seasonal pattern → Adjust resources and expectations by season.

Segment-Based Actions

High-retention segment → Increase acquisition investment, build case studies. Low-retention segment → Investigate fit issues, improve qualification, or deprioritize segment.

Combined Analysis Actions

Segment improving → Increase investment. Segment declining → Investigate, intervene, or reduce focus. Strategic resource allocation should flow directly from combined analysis.

Regular Review Cadence

Monthly: review latest cohort retention vs expectations. Quarterly: comprehensive segment analysis and resource allocation decisions. Annual: strategic planning based on long-term cohort trends.

Action Principle

Every cohort analysis session should end with specific actions: "Based on this analysis, we will do X." Analysis without action is intellectual exercise, not business improvement.

Frequently Asked Questions

Which approach should I start with?

Start with time-based monthly cohorts. They provide the baseline understanding of retention trends and are simpler to implement. Add segment-based analysis after you have 6+ months of time-based data and understand basic retention patterns.

How many cohorts should I track?

Track 12-24 monthly time-based cohorts for trend visibility. Add 3-5 key segment dimensions (plan, channel, size). Combined analysis can create many sub-cohorts—focus on segments with sufficient size (30+ customers) for reliable analysis.

How do I handle customers who change segments?

Standard approach: assign customers to their original segment at signup and keep them there. This prevents survivorship bias where only successful customers remain in premium segments. Alternative: track segment migration as a separate analysis.

What is the minimum cohort size for reliable analysis?

30-50 customers minimum per cohort for directionally reliable retention rates. Fewer than 30 creates high variance—one customer churning significantly changes the percentage. With small cohorts, combine months or segments to reach minimum size.

Should I track logo retention or dollar retention?

Track both—they answer different questions. Logo retention shows customer success rate. Dollar retention shows revenue health (includes expansion and contraction). A company can have 90% logo retention but 110% dollar retention if remaining customers expand.

How far back should I track cohorts?

Track as far back as you have reliable data—2-3 years if available. Older cohorts show long-term retention curves that predict customer lifetime value. But prioritize recent cohorts (last 12 months) for operational decision-making.

Disclaimer

This content is for informational purposes only and does not constitute financial, accounting, or legal advice. Consult with qualified professionals before making business decisions. Metrics and benchmarks may vary by industry and company size.

Key Takeaways

Time-based and segment-based cohort analysis answer fundamentally different questions: time-based reveals whether retention is improving; segment-based reveals which customers succeed. Most sophisticated analysis combines both dimensions to understand improvement trajectory within each segment. Start with monthly time-based cohorts to establish baselines, layer in key segments as you identify meaningful differences, and use combined analysis for strategic resource allocation. QuantLedger automatically generates both time-based and segment-based cohort views from your Stripe data, with the ability to combine dimensions for comprehensive retention intelligence.

James Whitfield