RetentionAdvanced7 min read

Renewal Forecasting

Q: What are common mistakes with Renewal Forecasting?

The trap is treating renewal forecasting like sales forecasting — assigning probability based on CSM gut feel ('feels like an 80%'). CSM intuition is wrong as often as it's right; they over-weight recent positive interactions and under-weight quiet disengagement. The other trap: forecasting only the dollars at renewal in the current quarter. The leading indicators of Q3 renewal show up in Q1 — if you don't have visibility 6 months out, your forecast is reactive. Real renewal forecasting tracks the full forward pipeline 12 months out with weighted probability.

Q: How do you apply Renewal Forecasting in practice?

Build a renewal forecast that combines (1) signal-driven scoring (account health, usage trend, sponsor activity, support sentiment) and (2) CSM judgment (recent conversations, known executive changes, competitive threats). Weight each renewal dollar by a probability score (0-100) computed from both inputs. Surface the pipeline by quarter: Committed (>80%), Likely (60-80%), At-Risk (30-60%), Best-Case (<30%). Review weekly with CS leadership; review monthly with CFO. Every at-risk renewal must have a documented save plan with owner and deadline. Forecasting that doesn't drive action is a vanity report.

Renewal forecasting is the process of predicting, account-by-account, what % of upcoming contracted ARR will renew, expand, downsize, or churn — typically projected over the next 1-2 quarters. The output isn't a single number; it's a weighted pipeline view: 'Q3 has $4.2M up for renewal, $3.1M is committed (>80% likely), $700K is at-risk (40-80%), $400K is best-case (<40%).' Without forecasting, finance gets surprised by Q-end churn and CS doesn't know where to focus. With forecasting, every at-risk dollar gets surfaced 90+ days before renewal, when there's still time to intervene.

Also known asRenewal PipelineRenewal ForecastGRR ForecastingRetention Forecasting

Challenge a friend Browse library

The Trap

The trap is treating renewal forecasting like sales forecasting — assigning probability based on CSM gut feel ('feels like an 80%'). CSM intuition is wrong as often as it's right; they over-weight recent positive interactions and under-weight quiet disengagement. The other trap: forecasting only the dollars at renewal in the current quarter. The leading indicators of Q3 renewal show up in Q1 — if you don't have visibility 6 months out, your forecast is reactive. Real renewal forecasting tracks the full forward pipeline 12 months out with weighted probability.

What to Do

Build a renewal forecast that combines (1) signal-driven scoring (account health, usage trend, sponsor activity, support sentiment) and (2) CSM judgment (recent conversations, known executive changes, competitive threats). Weight each renewal dollar by a probability score (0-100) computed from both inputs. Surface the pipeline by quarter: Committed (>80%), Likely (60-80%), At-Risk (30-60%), Best-Case (<30%). Review weekly with CS leadership; review monthly with CFO. Every at-risk renewal must have a documented save plan with owner and deadline. Forecasting that doesn't drive action is a vanity report.

Formula

Forecasted Renewal ARR = Σ(Account ARR × Renewal Probability_i) for all accounts in the forecast window

In Practice

Salesforce's customer success organization runs a quarterly renewal forecast that's updated weekly. Each renewal dollar is weighted by a model that combines product usage trends, executive sponsor activity, NPS history, and CSM-reported risk. Their Q-3 (3 quarters out) forecast accuracy is within 4% of actual outcomes. The system surfaces 'Best-Case' accounts — those with <40% renewal probability — to the CS leadership team as 'red list' accounts that get exec sponsor intervention. Internal data shows red-list accounts that received exec intervention within 60 days of identification renewed at 55%, vs 18% for those identified <30 days from renewal date.

Pro Tips

01
Always forecast 4 quarters out, not 1. The accounts at risk in Q4 show signals in Q1. A 90-day renewal forecast is reactive; a 12-month forecast is strategic. The CS team that knows their Q4 risk in Q1 has 9 months to fix it — and 9 months is enough to materially change the outcome.
02
Track 'forecast accuracy' as a CS leadership KPI. Compare forecasted renewal % to actual at quarter-end. If you're consistently optimistic by 5%+, your CSMs are sandbagging in the wrong direction. If you're pessimistic, they're over-flagging risk to look like heroes when accounts renew.
03
Renewal forecasting MUST distinguish 'gross renewal' from 'net renewal'. Gross = % of dollars retained. Net = retained + expansion. A team forecasting net renewal can hide gross churn behind expansion — losing logos but growing wallets. Both numbers must be visible separately.

Myth vs Reality

Myth

“Renewals are predictable — most customers renew on autopilot”

Reality

Even at 90% gross retention, 10% of renewals are decisions, not autopilot. In a $20M ARR book, that's $2M of dollars that depend on active intervention. The 'auto-renew' assumption is what creates Q-end surprises — the 10% that DON'T autopilot disproportionately churn.

Myth

“CSM forecast is the most accurate signal”

Reality

CSM forecasts overweight recent positive interactions. The customer who had a great QBR last week gets forecasted at 90%; the model that sees their usage declining for 60 days flags them at 55%. The right forecast is a weighted blend — never CSM judgment alone.

Try it

Run the numbers.

Pressure-test the concept against your own knowledge — answer the challenge or try the live scenario.

🧪

Knowledge Check

Your Q3 renewal forecast shows $5M up for renewal: $4M committed, $700K at-risk, $300K best-case. Total forecasted renewal: $4.5M (90% GRR). Two weeks before quarter-end, the CFO asks 'are we hitting the number?' What's the right answer?

Industry benchmarks

Is your number good?

Calibrate against real-world tiers. Use these ranges as targets — not absolutes.

Gross Revenue Retention (GRR) by Segment

Annual GRR by customer segment, not net of expansion

Enterprise SaaS

92-98%

Mid-Market SaaS

85-92%

SMB SaaS

70-85%

Consumer Subscription

60-80%

Source: OpenView SaaS Benchmarks 2024

Renewal Forecast Accuracy

90-day-out forecast vs actual quarter-end renewal

Excellent

Within ±3%

Good

Within ±5%

Acceptable

Within ±8%

Unreliable

>±10% off

Source: Hypothetical: KnowMBA composite from CS leadership surveys

Real-world cases

Companies that lived this.

Verified narratives with the numbers that prove (or break) the concept.

☁️

Salesforce

2020-2023

success

Salesforce's CS organization runs a renewal forecast updated weekly, with each account scored by a model that blends product usage trends, exec sponsor activity, NPS history, and CSM judgment. Their Q-3 forecast accuracy is within ~4% of actual. Accounts scored as 'Best Case' (<40% renewal probability) are flagged as 'red list' and receive exec sponsor intervention. Internal data: red-list accounts intervened on within 60 days of identification renewed at 55%; intervention <30 days from renewal date dropped save rates to 18%. The forecast wasn't just a measurement — it was the trigger for the highest-leverage retention activity in the company.

Q-3 Forecast Accuracy

Within ±4%

Red-List Save Rate (60+ day intervention)

55%

Red-List Save Rate (<30 day intervention)

18%

Gross Revenue Retention

>95%

A renewal forecast is only valuable if it triggers intervention 60+ days out. Forecasts produced for the CFO without driving CS action are reporting theater — the leverage is in early identification + decisive response.

Source ↗

📊

Hypothetical: Series C SaaS

2024

failure

A $50M ARR Series C SaaS company forecasted Q4 renewals at 92% GRR based purely on CSM judgment. Actual GRR came in at 81% — an $5.5M variance. Post-mortem revealed CSMs were overweighting recent positive QBRs and underweighting silent disengagement signals (sponsor logins down 40%, usage trending down 25%). The board demanded a model-driven forecast going forward. Within 6 months, forecast accuracy improved to within ±3% and the CFO could plan cash with confidence — but the credibility damage from the original miss took a full year to rebuild.

Forecasted GRR

92%

Actual GRR

81%

ARR Variance

-$5.5M

Post-Fix Forecast Accuracy

±3%

CSM gut-feel forecasting consistently overshoots reality. The first time you miss your renewal forecast by 10%+ is the last time the CFO trusts CS leadership without a model behind the number.

Source ↗

Decision scenario

The Q-End Renewal Crisis

It's October 1. You're VP CS at a $30M ARR SaaS. Q4 has $7M up for renewal. Your forecast model just kicked out: $5M committed, $1.2M likely, $600K at-risk, $200K best-case → forecasted $6.16M (88% GRR). Your CRO needs to commit to a renewal number for the board on October 15.

Q4 Renewal ARR Up

$7M

Committed

$5M (90% prob)

Likely

$1.2M (70% prob)

At-Risk

$600K (45% prob)

Best-Case

$200K (20% prob)

Decision 1

You can commit to $6.16M (the model number), $6.5M (slightly aspirational, with a save plan on at-risk), or $5.8M (conservative, sandbagged). Your CRO is pushing for $6.5M.

Commit to $6.5M — be aggressive, motivate the team to save the at-risk accountsReveal

Q4 closes at $6.05M — you missed by $450K. The CFO and board lose confidence in CS forecasting. Worse, the team burned out trying to save accounts that were structurally lost (sponsor changes, M&A in customer base) — and demoralized themselves chasing impossible saves. The next quarter, half the team disengages from forecasting because 'the number is whatever the CRO wants anyway.'

Q4 Actual GRR: Forecast 93% → Actual 86%Board Trust: Damaged

Commit to $6.16M (the model number). Build save plans on every at-risk account with documented owner and deadline. Update the forecast weekly.Reveal

Q4 closes at $6.22M — you beat the forecast by $60K. The save-plan discipline recovered $300K of at-risk dollars that would otherwise have churned. The CFO trusts the forecast going forward; the board sees CS as a reliable function. The team's morale is high because they hit a realistic, model-grounded number with effort.

Q4 Actual GRR: Forecast 88% → Actual 89%Board Trust: Strengthened

Related concepts