We are focusing on retention-curve shapes rather than just benchmark percentages, then map common patterns across SaaS types like SMB self-serve, enterprise, vertical, usage-based, and freemium/PLG.
Check out these real SaaS financial model templates with various types of retention frameworks.
For signup-month cohorts, the biggest trap is assuming SaaS sector alone explains retention. In practice, these are better ways to segment customer behavior patterns:
B2C / prosumer / low-ARPA self-serve
SMB SaaS
Mid-market SaaS
Enterprise SaaS
Usage-based / infra SaaS
Vertical SaaS
Mission-critical system-of-record SaaS
ACV (contract sizes) and buyer type often explain cohort shape better than industry. SaaS Capital’s 2025 benchmark work makes this point directly: retention benchmarking by ACV is usually a better starting point than benchmarking by company age, ARR, or industry, because companies at similar ACVs tend to have similar GTM (go-to-market), implementation, support, and customer behavior. Higher NRR (Net Revenue Retention) also tends to correlate with higher ACV.
The common retention curve shapes
1. The “early cliff, then plateau” curve
This is common in:
| SaaS type | Examples |
|---|---|
| Consumer SaaS | learning apps, wellness apps, creator tools |
| Prosumer SaaS | design tools, writing tools, AI assistants |
| Self-serve SMB SaaS | scheduling, social posting, simple CRM, invoicing |
| Freemium / trial-heavy PLG | individual productivity, browser tools, lightweight dev tools |
The cohort pattern usually looks like this:
Month 0 100%
Month 1 55–75%
Month 3 35–60%
Month 6 25–50%
Month 12 15–40%
For user-level retention, the drop can be even steeper. Pendo’s 2025 global product benchmark found that software products retain about 39% of users after one month and about 30% after three months on average, which is a useful reminder that user-retention curves and paid-account-retention curves can look very different.
What it usually means: a large share of signups are “tourists.” They tried the product but never reached durable value. In your model, the most important columns are often month 1, month 2, and month 3. After that, the retained base may become relatively stable.
For these businesses, signup-month cohorts are useful, but I’d also create cohorts by:
first value moment
first paid conversion
first repeated-use event
first team invite
first integration connected
Signup month alone often mixes serious users with casual trial users.
2. The “steady monthly decay” curve
This is common in:
| SaaS type | Examples |
|---|---|
| Low-touch SMB SaaS | email tools, appointment tools, simple project management |
| Marketing tools | SEO tools, ad tools, outbound tools |
| Light back-office tools | simple invoicing, bookkeeping add-ons |
| Horizontal productivity tools | note-taking, lightweight task tools |
The curve does not just drop early; it keeps leaking.
Month 0 100%
Month 1 85–90%
Month 3 75–82%
Month 6 65–75%
Month 12 45–65%
Month 24 25–50%
This usually indicates low switching costs, weak habit formation, a broad customer base with mixed fit, or customers buying for short-lived projects.
Marketing SaaS often shows this pattern because customers cancel when campaigns end, budgets tighten, channel performance disappoints, or tools overlap. Sales-tech can behave similarly: customers may buy during a growth push, then churn or contract when headcount, outbound volume, or pipeline targets change.
For these businesses, I’d model:
logo retention
MRR retention
reactivation
plan downgrades
acquisition channel retention
Signup-month cohorts are useful, but channel cohorts are often more revealing. Paid search, affiliate, AppSumo-style deals, and influencer-driven signups often create very different curves from organic, sales-led, or referral cohorts.
3. The “annual renewal cliff” curve
This is common in:
| SaaS type | Examples |
|---|---|
| B2B annual-contract SaaS | CRM, HRIS, ERP modules, compliance tools |
| Enterprise SaaS | workflow automation, data platforms, security |
| Mid-market SaaS | sales enablement, customer success, RevOps |
| Vertical SaaS | practice management, field service, restaurant systems |
The month-to-month logo curve may look deceptively strong until renewal:
Month 0 100%
Month 1 99%
Month 3 98%
Month 6 97%
Month 11 96%
Month 12 84–92%
Month 24 72–85%
Month 36 62–80%
This is not a smooth churn business; it is a renewal-event business. If you model it using signup month only, you may miss the fact that churn is concentrated around contract anniversaries.
For annual-contract SaaS, I’d add:
contract start month
renewal month
contract term length
implementation completion month
customer success owner
renewal cohort
ChartMogul’s billing benchmark also found that annual plans consistently drive stronger retention across ARR and ARPA levels, while monthly billing can help early-stage companies grow faster by lowering buying friction.
4. The “land-and-expand” curve
This is common in:
| SaaS type | Examples |
|---|---|
| Enterprise SaaS | CRM, data platforms, security, observability |
| Seat-based SaaS | collaboration, design, support, sales tools |
| Usage-based SaaS | API platforms, infrastructure, data warehouses |
| Multi-product SaaS | platforms with modules, add-ons, cross-sells |
Logo retention may decline modestly, while revenue retention rises:
Logo retention:
Month 0 100%
Month 12 85–95%
Month 24 75–90%
Revenue retention:
Month 0 100%
Month 12 105–130%
Month 24 110–150%
This is the classic SaaS compounding pattern: some customers leave, but retained customers add seats, usage, modules, or volume.
ChartMogul’s retention research found that higher ARPA businesses have much better net retention; only 2.7% of SaaS businesses with ARPA under $10/month had NRR over 100%, compared with 41.1% of businesses with ARPA over $500/month. It also notes that best-in-class mid-market and enterprise B2B SaaS NRR is often in the 115–125% range.
In your model, do not just look at customer count. For land-and-expand SaaS, create parallel cohort views:
logo retention
seat retention
MRR / ARR retention
usage retention
module adoption
expansion MRR by customer age
A healthy land-and-expand company may show declining customer counts but rising revenue per retained cohort.
5. The “implementation cliff, then high durability” curve
This is common in:
| SaaS type | Examples |
|---|---|
| Vertical SaaS | dental, legal, salon, construction, field service |
| ERP-like SaaS | finance ops, inventory, procurement |
| Healthcare SaaS | practice management, patient workflows |
| Compliance SaaS | GRC, audit, risk management |
| Payroll / accounting / fintech SaaS | payroll, payments, tax, bookkeeping |
The curve often has two risk zones:
Risk zone 1: before go-live
Risk zone 2: first renewal
A typical pattern:
Signed customers: 100%
Implemented customers: 80–95%
Month 12 retained: 75–92%
Month 24 retained: 65–88%
Once fully embedded, retention can become very strong because the product becomes part of the customer’s operating system. But failed onboarding, data migration, integrations, or workflow disruption can cause early churn.
For these businesses, signup month is often the wrong anchor. Better anchors are:
contract signed month
implementation start month
go-live month
first transaction processed
first payroll run
first invoice sent
first patient / client / job managed
Tidemark’s 2025 Vertical & SMB SaaS benchmark noted that fintech-led companies had the strongest retention profile among vertical SMB SaaS categories, closely followed by back-office control points.
6. The “usage ramp” curve
This is common in:
| SaaS type | Examples |
|---|---|
| API SaaS | messaging, payments, identity, maps |
| Cloud infrastructure | compute, storage, data pipelines |
| Developer tools | CI/CD, observability, testing |
| Data SaaS | warehouse, reverse ETL, event pipelines |
| AI infrastructure | model hosting, inference, vector databases |
Customer count may be stable, but usage and revenue can behave very differently.
A common revenue cohort pattern:
Month 0 100%
Month 1 60–90% pilot / testing noise
Month 3 90–140% production ramp
Month 6 110–180%
Month 12 120–250%
But there can also be sharp contraction if the customer’s own usage falls, if a large project ends, or if the customer optimizes spend.
Usage-based SaaS often has strong expansion potential because pricing scales with customer value. OpenView describes usage-based and hybrid pricing as increasingly common across infrastructure, horizontal apps, and vertical apps, and notes that usage-based public SaaS companies have shown strong scale performance driven by best-in-class net retention. Metronome’s 2025 usage-based pricing survey similarly found broad UBP adoption across SaaS categories and highlighted expansion within existing accounts as a key benefit.
For this type, model cohorts by:
first production usage
first API call
first successful integration
committed contract start
usage tier
customer workload type
Signup month is usually too early. A developer might sign up months before production usage begins.
7. The “seasonal scallop” curve
This is common in:
| SaaS type | Examples |
|---|---|
| Education SaaS | school-year cycles |
| Tax / accounting SaaS | tax season, fiscal year-end |
| Retail SaaS | holiday season, inventory cycles |
| Hospitality / events SaaS | seasonal demand |
| HR / benefits SaaS | open enrollment |
| Construction / field service | weather and project cycles |
The curve may appear to decline and recover depending on the activity definition.
Example:
Month 0 100%
Month 1 75%
Month 2 60%
Month 3 52%
Month 4 58% seasonal return
Month 5 65%
Month 6 55%
This is where strict “active this month” retention can mislead. For seasonal SaaS, consider rolling retention or “returned by next relevant season.”
Better metrics:
active in same season next year
renewed for next academic year
filed another return
ran next payroll cycle
processed next event
completed next compliance cycle
For these sectors, calendar-month cohorts can be more useful than signup-month cohorts because external seasonality dominates customer behavior.
8. The “macro-sensitive contraction” curve
This is common in:
| SaaS type | Examples |
|---|---|
| Recruiting SaaS | ATS, sourcing, interview tools |
| Sales SaaS | outbound, enablement, prospecting |
| Marketing SaaS | campaign tools, attribution, content tools |
| Real estate / construction SaaS | market-cycle exposed tools |
| Fintech for lending / investing | tied to transaction volumes |
The pattern may look fine for several cohorts, then suddenly worsen across all cohorts during a macro shift.
In a cohort table, this shows up as a vertical stripe, not a single bad cohort:
M1 M2 M3 M4 M5 M6
Jan 92 88 84 72 70 68
Feb 93 89 73 71 69
Mar 94 76 74 72
Apr 80 78 76
May 82 80
That “all cohorts got worse in the same calendar month” pattern usually means a market event, budget freeze, pricing change, billing issue, outage, competitor launch, or product change.
For macro-sensitive categories, model:
customer headcount
customer hiring volume
customer ad spend
customer transaction volume
customer industry
renewal quarter
budget cycle
HR and recruiting tools are especially prone to this: logo retention might remain acceptable while seat count, usage, or revenue contracts sharply during hiring slowdowns.
9. The “mission-critical flatline” curve
This is common in:
| SaaS type | Examples |
|---|---|
| Payroll | payroll processors, HR/payroll systems |
| Accounting systems | GL, AP, AR, close management |
| Security | identity, endpoint, cloud security |
| Compliance | audit, risk, regulatory reporting |
| Core vertical systems | practice management, field service, POS |
These products often show high durability once implemented:
Month 0 100%
Month 3 96–99%
Month 6 94–98%
Month 12 88–96%
Month 24 80–93%
The interesting part is often not monthly churn. It is:
failed implementation
vendor consolidation
M&A
business closure
compliance cycle
ERP migration
security breach / incident response
For these companies, low product usage does not always mean low retention. A compliance product, security product, or payroll product may be used heavily only around specific workflows but still be highly retained.
Use job-completion metrics, not just logins:
payrolls run
compliance reports filed
incidents resolved
audits completed
invoices reconciled
claims processed
10. The “network effect / team adoption” curve
This is common in:
| SaaS type | Examples |
|---|---|
| Collaboration SaaS | Slack-like tools, async video, docs |
| Design tools | design collaboration, whiteboards |
| Project management | tasks, workflows, planning |
| Support / success tools | shared inboxes, ticketing, CS platforms |
| Knowledge bases | internal docs, wikis |
Individual user retention may be mediocre, but account retention improves sharply once a team threshold is crossed.
Typical pattern:
1 user only: poor retention
2–4 users invited: moderate retention
5–10 active users: strong retention
Cross-functional use: very strong retention
For these products, signup month is less useful than activation state. Segment by:
number of invited users by day 7
number of active users by day 30
teams / departments created
shared objects created
collaborative actions
integrations connected
The retention curve often bifurcates: accounts that do not invite teammates churn early; accounts that create shared workflows become sticky.
Sector-level patterns I’d expect
Here is the practical version.
| Sector / SaaS type | Common logo retention pattern | Common revenue retention pattern | Main modeling caution |
|---|---|---|---|
| Consumer / prosumer SaaS | Very steep early drop, then low plateau | Usually decays unless strong annual plans or premium tiers | Signup cohorts mix serious users with casual trials |
| Low-priced AI tools / AI wrappers | Big trial churn, high substitution risk | Can decay quickly unless embedded in workflow | Cohort by repeated output/workflow completion, not signup |
| SMB horizontal SaaS | Early cliff plus steady decay | Often below 100% NRR unless expansion or annualization works | Acquisition channel quality matters a lot |
| Marketing SaaS | Moderate-to-high churn; budget-sensitive | Expansion possible, but contraction common | Cohort by customer spend/channel maturity |
| Sales SaaS | Sensitive to headcount and pipeline cycles | Seat contraction can dominate logo retention | Track seats and active reps, not just accounts |
| HR / recruiting SaaS | Renewal-driven, cyclical | Contracts during hiring slowdowns | Separate recruiting from core HR/payroll |
| Core HR / payroll | Durable after go-live | Stable, with expansion via employees/payroll volume | Cohort by first payroll/open enrollment |
| Accounting / finance ops | Strong once embedded | Stable to expanding | Usage may be periodic; login retention can mislead |
| Security / compliance | High retention, renewal-driven | Strong if compliance/security mandate persists | Churn is often consolidation, not daily usage |
| Developer tools | Early trial drop, then strong if productionized | Can expand strongly with engineering usage | Cohort by first production usage |
| API / infrastructure SaaS | Logo retention can be strong after integration | High expansion, but usage volatility | Model committed vs pay-as-you-go separately |
| Vertical SaaS | Strong if it becomes operating system | Strongest when payments/fintech/back office are embedded | Business failure can look like product churn |
| Education SaaS | Seasonal and renewal-driven | Often tied to school/fiscal cycles | Use academic-year cohorts |
| Enterprise platforms | Flat monthly, renewal cliffs at 12/24/36 months | Often 100%+ NRR via seats/modules | Signup month is often less useful than contract/go-live month |
How I’d improve your cohort model
Since you already cohort by signup month, I’d keep that but add a few “second dimension” cuts.
The most useful additions are:
1. Signup month × acquisition channel
2. Signup month × customer segment / ACV
3. Signup month × plan type: free, monthly, annual, enterprise
4. Signup month × activation status
5. Signup month × first-paid month
6. Signup month × first meaningful-use month
7. Signup month × industry / vertical
8. Signup month × seats or team size by day 30
For B2B SaaS, I’d especially separate:
self-serve SMB
sales-assisted SMB
mid-market
enterprise
ChartMogul’s research shows why this matters: low-ARPA SaaS tends to have much weaker net retention, while higher-ARPA B2B SaaS has more informed buying, lower gross churn, and more expansion opportunity.
The cohort-table signatures to look for
Early activation problem
M1 bad, then curve stabilizes
Usually onboarding, acquisition quality, weak activation, or too many low-intent signups.
Product-market fit problem
M1 okay, but every month keeps leaking
Usually ongoing value is not strong enough, or the category has low switching costs.
Renewal problem
M1–M11 strong, M12 falls sharply
Usually price-value mismatch, CS failure, budget issue, or annual-contract churn.
Expansion engine
Logo retention declines, revenue retention rises
Healthy land-and-expand pattern.
Seat contraction
Logo retention stable, revenue retention falls
Common in sales, HR, support, collaboration, and any seat-based SaaS during downturns.
Usage ramp
Logo retention stable, revenue/usage retention rises after month 2–6
Common in API, infra, data, developer, and usage-based SaaS.
Seasonal usage
Activity retention falls, then returns later
Common in education, tax, compliance, events, retail, and seasonal operations.
Calendar-period shock
All cohorts worsen in the same calendar month
Usually macro, pricing, product change, outage, billing issue, competitor, or budget freeze.
Default Mental Model to ConsiderFor SaaS retention curves, I’d read them like this:
- Low ARPA / self-serve:
- Retention is mostly about activation and habit.
- SMB SaaS:
- Retention is about onboarding, business quality, and whether the product is essential.
- Mid-market SaaS:
- Retention is about implementation, measurable ROI, and renewal management.
- Enterprise SaaS:
- Retention is about stakeholder depth, integrations, workflow ownership, and expansion.
- Usage-based SaaS:
- Retention is about production adoption and customer workload growth.
- Vertical SaaS:
- Retention is about becoming the operating system, ideally with payments or back-office control points.
- Mission-critical SaaS:
- Retention is about switching cost, compliance, embedded workflows, and avoiding failed implementation.
For your current setup, the most impactful upgrade is probably this: keep signup-month cohorts, but add activation-based cohorts. In many SaaS businesses, the real retention curve begins not at signup, but at the moment the customer first gets durable value.
Hire me to build custom financial models here.
Article found in SaaS.
