How to Run a Renewal Forecast Review

What is a renewal forecast review?

It is a monthly working session that calls each upcoming renewal, assigns a risk level with evidence, and commits to intervention while the outcome can still change. It is the retention equivalent of a forecast call.

Most companies run renewals as an administrative task. A list generates 60 days out, someone sends a notice, and the miss arrives as a surprise. That process manages paperwork rather than revenue.

The review needs its own meeting because renewal risk builds on a different clock than new business. A deal in new pipeline moves in weeks. A customer decides to leave over months of accumulated experience, and the signals appear long before the renewal date does.

Why forecast renewals separately from new business?

Because the drivers are unrelated and blending them hides contraction. New business is driven by pipeline creation and win rate. Renewals are driven by product usage, support experience, and executive sponsorship.

Forecast accuracy benchmarks reflect this split. Accuracy targets on new and expansion business are set independently of renewal, because the underlying prediction problem is different. A model trained on new business patterns has no reason to be right about a customer's decision to stay.

The reporting consequence matters more. When both sit in one number, a strong new logo quarter covers a base that is quietly shrinking. You find out four quarters later, when new logo volume flattens and the contraction becomes visible all at once.

Which accounts belong on the agenda?

Every renewal in the next 180 days, sorted by risk rather than by date. Reviewing by date guarantees you spend the meeting on next month's renewals, which are already decided.

Use support case volume as the primary risk cut. It is the most useful early churn signal available, and the relationship is not linear:

The zero-case account surprises people. Silence looks like satisfaction on a dashboard and usually means nobody at the customer is invested enough to ask for help. It is the same pattern that shows up in new business, where the earliest sign a deal is dead is the absence of any signal at all.

Layer two additional filters on top: any account above a revenue threshold, and any account where the executive sponsor has changed since the last renewal.

What questions do you ask per account?

Five, answered with evidence rather than relationship confidence.

- Who signs the renewal, and have we spoken with them in the last 90 days? - What has the customer done in the product in the last 30 days, measured rather than assumed? - What did they buy this for, and can we show it happened? - Which support cases are open right now, and how old is the oldest? - What is the expansion path, and has anyone proposed it?

Answers have to be specific enough to be wrong. "The relationship is strong" is not an answer to any of these. "We spoke with the CFO on the 14th, she confirmed budget is approved, and the security review closed last week" is an answer, and it can be checked.

The product usage question deserves its own discipline. Customer success teams routinely report health scores built on sentiment from the people they talk to, which are the people most invested in the relationship. Measured usage tells you what the rest of the account is doing, and it is usually the first place a renewal problem becomes visible.

The last question changes the meeting from defense to offense. A renewal review that only asks whether accounts will stay produces a portfolio of flat renewals, which reads as a win on a logo retention chart and shows up as a decline in net revenue retention.

How does the renewal review connect to the ARR waterfall?

Every call in the review resolves into a specific line of the monthly waterfall, and the two have to reconcile.

The waterfall runs from beginning ARR to ending ARR through separate contraction and expansion lines:

Reconciling this by month is where the discipline comes from. When beginning ARR always equals the prior month's ending ARR and every renewal decision lands in a named line, contraction cannot hide in a rounding difference. Gross and net retention both fall out of the same table, which removes the arguments about whose retention number is correct.

How often should you run it, and what does it produce?

Monthly for the full portfolio, with weekly touchpoints on high-risk accounts inside 90 days.

The output is a per-account call, an intervention owner, and a date. Cap interventions at the number your CS team can actually execute in a month. A review producing thirty save plans across a team of four produces zero.

Two rules keep the calls honest. First, a renewal cannot be called committed without a confirmed conversation with the person who signs. Second, a call that moves from committed to at risk requires the same written explanation as a slipped deal in the new business forecast, because a late-arriving risk downgrade is the retention version of a close date change.

Then hold the review to a measurable standard. Track called renewal ARR against closed renewal ARR by month and publish the variance next to your new business number, the same way you would track forecast accuracy on the sales side. Teams that measure retention forecasting with the same rigor they apply to sales forecasting stop being surprised by their own base.