Renewals are the only operational risk that arrives with a dollar amount pre-attached. A tool bought in a busy month by a person who has since left, on a card nobody watches, auto-renews at list price for a seat count that was true eighteen months ago. The contract had a 30-day cancellation window; the invoice arrives 45 days after it closed. Nothing about that is a technology problem — it is a calendar problem. The fix is one table: every vendor, the real renewal date, the notice window that actually cancels it, and a named person who decides at 60 days — not on invoice day.
Full template with worked example: Vendor Renewal Calendar Template — HIVE80lab Ops Notes. The contracts behind the rows get their terms checked in the free Vendor Contract Checklist.
1. The renewal table, one page
One line per product, not per vendor (vendors sell bundles where you use one of four components — those are three dead lines and one live one). Six columns:
- Vendor + product — kills are per-product; you can't offboard half a vendor.
- Annual cost — list price at last renewal plus the seat count you paid for. You negotiate with last year's number in hand.
- Renewal date — the date the contract renews, which is often not the invoice date. Check the contract, then the billing portal, then reconcile.
- Notice window — days of written notice to cancel without penalty, precomputed as the calendar date that implies (renewal date minus notice days). "30-day notice" discovered on day 12 is a full-price year.
- Decision owner — one name who decides renew / downgrade / kill. Renewals without owners are renewals by default — the vendor's favorite kind.
- Usage evidence — monthly active users, last login of the least-active seat, whether the workflow it supports still exists.
2. The 90/60/30 pipeline
- T−90 — Evidence. Seats vs. actual users, logins in the last 90 days, whether the original problem still exists. No negotiation here; just honesty about whether anyone would notice if the tool vanished.
- T−60 — Decision. Renew, downgrade, or kill — written down in one line. This is also the negotiation window: vendors discount hardest when cancellation is still cheap for you. The same conversation at T−5 has zero leverage.
- T−30 — Execute. Signed, downgraded, or cancelled and the cancellation confirmed in writing — save the confirmation. Anything killed starts the offboarding & data-deletion clock.
- T−0 — Invoice audit. Same price? Same seats? Same term? Silent price increases and seat creep are caught here, in the one month they can still be disputed.
3. The auto-renew kill list
Some renewals get re-decided every single year, by policy: month-to-month tools that quietly became annual; "trials" that converted two restructurings ago; the security product now duplicated by what the platform vendor bundles free; the marketing tool whose champion left; anything billed to a personal card (kill this quarter — it is also how a departed employee holds a key to your data). The kill list is not a punishment. It states that renewal is a decision, not a default. The tool that survives its kill-list review three years running is the one you actually needed.
4. Seat trueing before you sign
Seat contracts drift upward in only one direction. Before any seat-based renewal: provisioned seats vs. active users in the last 30 days — and reconcile the gap. The gap is departed staff nobody reclaimed (also an access-review finding), "just in case" seats nobody can justify, and accounts nobody reclaimed. Downgrade to the true number at the renewal line — vendors rarely pro-rate mid-term but almost always accept a lower seat count at renewal. A twelve-person firm doing this once usually finds 10–20% of SaaS spend was paying for former employees.
5. Four metrics that tell you the calendar is real
- Renewals caught with ≥30 days notice — below 90% means new purchases aren't entering the table.
- Spend cut at renewals — dollars downgraded or killed per quarter. Zero usually means the evidence step is skipped, not that every tool is perfectly sized.
- Orphans killed — zero-login products terminated instead of renewed. The cumulative number is the cleanest proof the table is real.
- Invoices that matched the table — mismatches this month are disputes you can still win; next quarter they are donations.
Worked example: $14,100/yr of drift found in one afternoon
A twelve-person firm ran this after an auto-renewal doubled their project-management tool's price at a seat count of 22 (they employed twelve, had provisioned 31). The table took two hours: 23 lines, 17 real. A $2,400/yr analytics tool whose champion left 14 months earlier — killed at T−60 with no usage in the window. Two overlapping password tools from different busy seasons — one killed. A security product duplicated by platform-bundled features — downgraded at renewal. Nine seats cut across three tools to match actual logins. Net on the first cycle: $14,100/yr cut, every remaining contract has an owner and a computed last-safe-cancellation date, and the quarterly 30-minute review has caught two more orphans since. Their tools didn't get worse — they stopped paying for the ones nobody would have noticed losing.
The sprawl that puts vendors on this table in the first place is measured by the free SaaS Sprawl Audit Checklist; who to call at 2 a.m. is the Vendor Incident Contacts one-pager. If you want the whole loop — the audit that finds the gaps, the plan that closes them — run by an outside pair, that is the Small-Team Ops Audit ($149, five-day turnaround, and it reads your vendor spend as part of the audit). When a vendor outage is the scenario that worries you most, the Custom Incident Runbook ($249, delivered in 48h) turns it into a procedure. The Ops Starter Kit ($14) covers the incident half; Vol. 2 ($27) covers the on-call rotation that inherits it. 20% off any paid kit with code HIVE20 at checkout.
Top comments (0)