How to build a CRM-driven renewal process

Before you touch cadences or health scores, make sure the renewal exists in the CRM as a real opportunity record, not a date field on the account. Create the renewal opportunity automatically the moment the original contract is countersigned, with a close date set to the contract end date, a stage of "Renewal: Not Started," an amount equal to current ARR, and the account owner and customer success manager both populated. Treat the renewal exactly like a new-business opportunity for pipeline, forecasting, and review purposes. Teams that store renewals as a date on the account inevitably lose them. The record has no owner, no stage, no next step, and no forecast weight, so it drifts until someone notices it is thirty days out and panic sets in. A renewal is a sales motion. Give it the same system support you give new business and the whole process gets cleaner.

• Auto-create a renewal opportunity on contract signature, dated to the contract end date, amount set to current ARR
• Assign both the account executive and the customer success manager so dual ownership is visible from day one
• Set the initial stage to a dedicated renewal pipeline, not the new-business pipeline, so forecasts stay clean
• Backfill renewal opportunities for every active contract before you launch the new process

Renewals need their own pipeline because the stages are different from new business. A reasonable default is six stages: Not Started, Discovery, Value Review, Commercial, Negotiation, and Closed. Not Started is the parking lot where every renewal lives until T-minus 120. Discovery is where the CSM runs the health conversation and surfaces usage gaps. Value Review is where the business case is quantified and shared. Commercial is where pricing, term, and uplift are proposed. Negotiation covers legal, procurement, and signature logistics. Closed splits three ways on the close reason field: Renewed, Downsold, or Churned. Keep the stage definitions short and the exit criteria tight. Reps and CSMs will not read anything longer than four bullets. Store the stage definitions where the forecast lives so nobody has to hunt. The goal is a renewal that can be forecast with the same discipline as any other deal.

• Create a dedicated renewal pipeline with six stages and distinct exit criteria for each
• Make close reason a required picklist on Closed, with values for Renewed, Downsold, and Churned
• Add a secondary field for renewal type: like-for-like, downsell, flat with term change, or expansion
• Train CSMs and AEs on the stage definitions before enabling the pipeline for forecasting

The 120/90/60 cadence is the backbone of every renewal process that holds up at scale. At T-minus 120 days the renewal moves from Not Started to Discovery, the CSM is prompted to open a health review, and a task is created to confirm the economic buyer and procurement contact are still in seat. At T-minus 90 the Value Review stage opens: the CSM is prompted to deliver a quantified value recap, and the account executive is looped in to flag any expansion signals. At T-minus 60 the Commercial stage opens: pricing is proposed, uplift is justified, and legal review is initiated for any non-standard terms. At T-minus 30 the Negotiation stage opens and finance is pulled in for revenue recognition timing. The automation does not do the work. It makes sure the work gets started. Reps who have run this cadence for a year will tell you the single most important moment is the T-minus 120 ping, because it is the only one that catches silent churn before it hardens.

• Create a time-based automation that advances renewal stage based on days-to-close-date
• Fire a task to the CSM at T-minus 120, 90, 60, and 30 with a stage-specific checklist
• Flag any renewal that has not advanced within seven days of its scheduled trigger as at-risk
• Escalate any renewal that is still in Not Started at T-minus 90 to the segment leader

Health scores fail when they are built from vibes. Build yours from four categories of hard signal. Product usage: active users, weekly active days, feature adoption against the baseline the account signed up to use, and the trend over the trailing ninety days. Commercial: time since last expansion, days until renewal, open credits, outstanding balance, and payment on time rate. Relationship: executive sponsor last engaged, number of named stakeholders in the account, open support tickets by severity, and NPS or CSAT trend. Lifecycle: has the account completed onboarding milestones, hit the usage threshold their plan was sized for, and attended at least one business review this term. Weight the categories by what historically predicted churn in your own data. Do not borrow a competitor's formula. Your product, your contract shape, and your buyer behavior are different, and borrowed health scores produce false confidence until the first bad quarter rips the illusion open.

• Pull last year of churned accounts and run a backward regression against candidate signals
• Weight the top four to six signals by predictive power, not by what feels important
• Score every account weekly and store the score history so trend becomes visible, not just the point value
• Publish the formula to CSMs and AEs so health is a shared language, not a black box

Auto-renew is a two-edged tool. Done well it protects revenue on long-tail accounts and reduces renewal friction for happy customers. Done badly it generates refund requests, chargebacks, and churn-with-a-vengeance when a customer finds the surprise charge. The honest version of auto-renew has five rules. First, auto-renew clauses are disclosed at signature in plain language, not buried on page nineteen. Second, the system sends a renewal notice at T-minus 60, T-minus 30, and T-minus 7, no matter what the contract says about notice. Third, any uplift above a published threshold (we use five percent) requires explicit opt-in, not silent application. Fourth, downgrades and cancellations are available self-serve in the admin console with no retention gates designed to trap. Fifth, finance reconciles the auto-renew list seven days before invoice to catch any account flagged as at-risk or in active negotiation. Teams that follow these rules see their auto-renew revenue hold up because the customers who get auto-renewed are the ones who actually meant to stay.

• Document auto-renew terms in plain language on the order form and the customer portal
• Send notice emails at T-minus 60, 30, and 7, with the invoice preview attached to the T-minus 7 note
• Require explicit opt-in for any uplift above the published threshold; default to flat if the customer does not respond
• Build a seven-day pre-invoice reconciliation hold so finance can suppress any at-risk or disputed renewals

The renewal window is the single highest-leverage expansion conversation of the year. The customer is already engaged, the business case is on the table, and procurement is briefed. Build the expansion motion into the Value Review stage so it happens on every renewal, not just the ones a rep remembers to pitch. At T-minus 90 the system should prompt the CSM to pull the usage report, flag any account that has crossed its license tier threshold, surface any adjacent product the account has not adopted, and route a lightweight expansion brief to the account executive. The brief includes current ARR, usage against limits, three expansion plays scored by fit, and a suggested uplift range anchored to published list pricing. The AE then owns the expansion conversation, the CSM owns the renewal conversation, and the two merge into a single commercial offer at T-minus 60. Teams that run this consistently see net retention gains of ten to fifteen points without changing their pricing, because expansion is finally happening on purpose instead of by accident.

• Auto-generate an expansion brief at T-minus 90, pulled from usage, product adoption, and account-fit signals
• Route the brief to the AE while the CSM owns the health conversation; the two reconcile before T-minus 60
• Score every expansion play against three criteria: fit, usage signal strength, and economic buyer appetite
• Track expansion win rate separately from renewal win rate so you can tune the motion over time

Health scores and automation only matter if the humans who own the number actually look at the data. Run a weekly renewal review with the segment leader, every CSM, and the account executive counterpart. Walk every renewal that is inside the T-minus 120 window and every at-risk account outside the window. For each one, ask three questions: what is the current health trend, what is the specific commitment the customer has made or needs to make this week, and what is the probability we land it at or above current ARR. Record the probability. Track delta from actuals. If you discover the team is systematically under-forecasting at ninety days and over-forecasting at thirty, you have a calibration problem, and calibration problems are fixable with coaching. The review is not a status meeting. It is a working session where next steps get assigned, saves get planned, and expansion asks get finalized. Hold it every week, same time, same agenda.

When a customer gives notice, the first forty-eight hours decide the outcome. Build the save motion ahead of time so nobody is improvising at the moment of maximum stress. The save kit contains four assets. First, a diagnostic script that walks a CSM through five questions in under fifteen minutes to understand whether the issue is product, price, people, or priority. Second, a tiered concession ladder with pre-approved pricing, term, and feature-unlock options the CSM can offer without escalation. Third, an executive sponsor outreach template so the account executive and a VP can co-sign a save plan within twenty-four hours. Fourth, a post-save commitment plan with a sixty-day check-in built in so a save does not just defer the churn by one cycle. Track save rate as a separate metric from gross retention. A save rate under thirty percent means the diagnostic is wrong or the concession ladder is too stiff. Above sixty percent and you are probably giving away margin that was not required.

• Write the diagnostic script with five questions covering product fit, pricing, stakeholder change, and strategic priority
• Pre-approve a concession ladder with CFO sign-off so CSMs can act without a two-day escalation delay
• Build an executive sponsor template library so VP outreach is twenty minutes, not two days
• Attach a sixty-day commitment plan to every save so you track whether the save held

Once the renewal process is live, four metrics belong in front of leadership every quarter. Gross revenue retention: how much of starting ARR you kept, before any expansion. Net revenue retention: how much starting ARR you kept, after expansion. Logo retention: what percentage of customer logos renewed, which exposes mix shifts the dollar metrics can hide. And renewal win rate by segment, so you see whether your mid-market retention is masking enterprise erosion or the reverse. Compare each metric to the prior four quarters and against your published target. Then dig into the drivers. If gross retention slipped, is the leak in a specific segment, a specific product line, or a specific cohort tenure band? If expansion dropped, did a product gap emerge, or did an AE change absorb attention away from the motion? Treat the renewal process like a product. Measure it, iterate on it, retire the parts that stopped working. The system that worked a year ago will not keep working forever, and renewal processes decay quietly, so the quarterly review is the circuit breaker that catches drift before it becomes damage.

• Report gross retention, net retention, logo retention, and renewal win rate every quarter at the leadership table
• Segment every metric by product, cohort, and ARR band so trends are visible, not hidden in the average
• Compare actual outcomes to the probabilities the team committed to; the gap is your true calibration score
• Retire health signals and stage criteria that stopped predicting outcomes; the system is a living artifact