Renewal Forecasting for SaaS: A Step-by-Step Guide

Most renewal forecasts are a spreadsheet of optimistic guesses that gets revised the week before quarter end. This guide shows how to build a renewal forecast you can defend: a weighted method based on account evidence, a copyable template, a worked example and a weekly routine to keep it honest.

Why renewal forecasting matters more now

Retention is getting harder to protect. In its 2026 SaaS benchmarks takeaways, Aleph reports that median gross revenue retention (GRR) fell to 84% in 2025, down from 90% three years earlier. If the typical company is losing more revenue at renewal than it used to, finance and leadership will lean harder on your forecast, and surprises will cost more.

A good forecast does two jobs. It tells the business how much recurring revenue to expect, and it tells your team where to spend time before the renewal date, while there is still room to change the outcome.

Step 1: Define what you are forecasting

Before you assign any probabilities, agree on the basics with finance:

• The unit. Forecast in dollars of ARR, not logos. A 5-account miss can be a rounding error or a disaster depending on size.
• The window. Pick the period (quarter or month) and include every contract with a renewal date inside it.
• The outcomes. Each renewal ends in one of four ways: renews flat, renews with expansion, renews with contraction, or churns. Forecast the base renewal and treat expansion separately so a big upsell does not hide a lost account.
• The start date. Velaris recommends starting renewal forecasting 90 to 180 days before the renewal date and continuing throughout the customer lifecycle, so there is time to intervene. A 120-day start is a reasonable default for most mid-market teams.

Step 2: Gather the evidence for each account

A forecast is only as good as the signals behind it. For every renewal in the window, collect the same short list so accounts are compared fairly:

1. Product usage trend: growing, flat or declining over the last 90 days.
2. Engagement: when you last spoke to a decision maker, not just a day-to-day user.
3. Sentiment: recent NPS or CSAT responses and the tone of recent conversations.
4. Commercial status: renewal quote sent, procurement involved, any discount or price-increase discussion.
5. Support: open escalations, reopened tickets and missed response targets.
6. Billing: overdue invoices.
7. Champion and sponsor status: are they still at the company, and still engaged?
8. Competitive or budget risk: any mention of a competitor, an audit of tools or a budget freeze.

Velaris lists the same families of inputs (health scores, usage, sentiment, financial data and historical trends) as the building blocks of an accurate forecast. If you do not have a health score yet, our guide to the customer health score formula shows how to build one.

Step 3: Assign probabilities with a simple rubric

Avoid assigning percentages by gut feel. Instead, map evidence to a small set of confidence tiers. The percentages below are starting points, not industry benchmarks; you will calibrate them in Step 5.

RENEWAL CONFIDENCE RUBRIC (starting values, calibrate with your own history)

Commit (90%)
- Usage stable or growing
- Exec sponsor engaged in the last 30 days
- Renewal quote accepted or verbally agreed
- No open escalations, no overdue invoices

Likely (75%)
- Usage stable
- Contact with day-to-day owner in last 30 days
- Renewal conversation started, no objections yet

Uncertain (50%)
- Usage declining OR sponsor unresponsive OR pricing objection
- Needs a recovery plan with an owner and a date

At risk (25%)
- Two or more red signals (declining usage, detractor score,
  champion left, competitor in play)
- Escalated to management

Churn expected (5%)
- Customer has given notice, or is clearly not renewing

Rules that keep the rubric honest:

• Any account with two or more red signals can never be "Commit," whatever the CSM feels.
• An account cannot be "Commit" until the customer has said yes in writing or in a recorded conversation.
• The tier must be reviewed any time a signal changes, not only at the weekly meeting.

Step 4: Build the weighted forecast

The core calculation is simple: for each renewal, multiply the ARR up for renewal by its probability, then add up the results.

Expected renewed ARR = sum of (ARR up for renewal x probability)

Forecast GRR = Expected renewed ARR / Total ARR up for renewal

A worked example

For example, imagine a company with eight renewals due next quarter (illustrative numbers, not real data):

Total ARR up for renewal is $410,000 and expected renewed ARR is $290,500, so the forecast GRR is about 70.9%. Notice that the single largest account (A) drives over a third of the expected total. If A slips from Commit to Uncertain, the expected total drops by $48,000. That is exactly the kind of exposure your forecast should show, and it tells you where to put executive attention.

Present a range, not a single number

Leadership should see three figures:

• Floor: only Commit accounts at their probability, with everything else at zero.
• Expected: the weighted forecast above.
• Ceiling: every account except "Churn expected" renews at full ARR.

Showing best-case and worst-case scenarios alongside the most likely outcome also protects you from false precision: the weighted figure is an estimate, not a promise.

Step 5: Calibrate against what actually happened

This is the step most teams skip. Each quarter, compare what you forecast with what renewed.

1. Export last quarter's forecast as it stood 90 days, 60 days and 30 days before the quarter ended.
2. Compare each tier to its actual outcome. If "Likely" accounts renewed 90% of the time, raise that tier's probability. If "Commit" accounts churned, find out which signal you missed and add it to the rubric.
3. Check for patterns by CSM, segment and contract size. Consistent over-forecasting suggests optimism, and consistent under-forecasting suggests sandbagging.

Rework's renewal metrics guide suggests that forecast accuracy 30 days before quarter end should be 90 to 95%, and that you should compare forecast against actual every week. Treat those as targets to aim for rather than guarantees, and measure your own baseline first.

Step 6: Run a weekly forecast review

A forecast only stays accurate if it is reviewed on a schedule. A 30-minute weekly meeting is enough:

• Review every account whose tier changed since last week and record why.
• Review every renewal inside 60 days that is not "Commit," and confirm an owner and a next step.
• Look at any account with a new red signal (usage drop, escalation, overdue invoice, detractor score).
• Compare this week's total to last week's and explain the change in one sentence.
• Update the floor, expected and ceiling numbers and send them to finance.

The point of the meeting is action, not reporting. If you want a structure for the at-risk part, use our churn risk review template.

Common mistakes

• Using sales-style stages. Pipeline stages like "proposal sent" say little about whether a customer is getting value. Anchor probabilities in health and engagement evidence.
• Blending segments. Rework warns that retention differs a lot between SMB, mid-market and enterprise accounts, so a single blended number can mislead. Forecast each segment separately and then combine them.
• Never changing probabilities. An account that sits at "Likely" for three months while usage falls is a warning sign, not stability.
• Counting expansion in the base. Keep renewal and expansion separate so you can see retention on its own.
• Skipping the final 30 days. Late changes are the most expensive to explain. Keep updating until the contract is signed.

Where a CRM helps

You can run this process in a spreadsheet, but the evidence in Step 2 lives in several systems, which is why spreadsheets drift out of date. In XPI CRM, the renewals board holds each renewal with a stage probability and a forecast, marking a renewal as Won updates the client's ARR, and the health score (0 to 100, built from usage trend, engagement, sentiment, NPS, billing, open work and support tickets) gives you an evidence-based starting point for the tier. The dashboard shows ARR at risk, GRR and NRR so the forecast and the actuals sit side by side. See the help guides for setup.

Try it in XPI CRM

If your renewal forecast currently lives in a spreadsheet nobody trusts, try moving it to a renewals board where health, usage and billing sit next to every contract. Start free — every feature included, no credit card.