A Useful Risk Register for Small Teams

BUILD A RISK REGISTER A SMALL TEAM WILL ACTUALLY USE ==================================================== The register dies when it becomes a spreadsheet people fill once and never open. Four rules keep it alive: 1. One page, max 10-15 risks. If everything is a risk, nothing is. 2. Every risk has ONE named owner. No teams, no "whoever". 3. Review it inside an existing meeting (standup or weekly sync), 10 minutes max. Never create a separate "risk meeting." 4. A risk without a leading indicator is a guess, not a risk. You track signals, not vibes. THE FIELDS (and how to define them so they're not ambiguous) Risk ID + statement One sentence: "If X happens, then Y." Naming the trigger makes it actionable. Likelihood (1-5) — define the scale once and reuse it: 1 = almost never, 2 = unlikely, 3 = possible, 4 = likely, 5 = near certain Impact (1-5): 1 = negligible, 2 = minor, 3 = significant, 4 = major, 5 = catastrophic Score = Likelihood x Impact. Set a threshold (e.g. >= 12) = "active risk, mitigation required." Below threshold = watch only. This stops the register from being 15 rows of everything. Leading indicator The observable signal that fires BEFORE the risk materializes. Examples: error rate crossing 1%, load test failing, content sign-off slipping past a date, status page going amber. Each indicator needs a number or a date, not "keep an eye on it." Mitigation Concrete actions with deadlines. "Add auto-scaling by Aug 30" beats "have a scaling plan." Include the trigger that escalates to the fallback. Owner One person. They own the indicator, the mitigation, and the status update. Review cadence How often this specific risk gets looked at (daily during launch week, weekly otherwise). Status column: Open / Watch / Closed. CADENCE MECHANICS - At each review, only touch risks whose status is Open or Watch. Closed risks stay in the log as history but get zero meeting time. - Ask three questions per risk: Did the indicator move? Is the mitigation on track? Does the owner still make sense? - Close a risk only when the event passed harmlessly OR the mitigation is fully in place and verified. - If a risk has been "Open" with no movement for 3 reviews, either escalate it or close it. Stagnation is the register's death. FILLED EXAMPLE — WEB LAUNCH REGISTER (launch in 3 weeks) ======================================================== Scale: L 1-5, I 1-5, score = L x I. Active threshold: >= 12. R1 If traffic spikes 3x at launch, hosting collapses and the site is down for hours. L 3 I 4 Score 12 ACTIVE Indicator: load test fails above 2x expected; error rate > 0.5% in staging; CDN queue depth rising. Mitigation: auto-scaling enabled + tested by Aug 28; CDN in front; load test at 3x traffic Aug 29; rollback plan documented. Owner: backend dev Cadence: daily during launch week. Status: Open R2 If the payment provider has an outage, checkout fails and orders are lost. L 3 I 5 Score 15 ACTIVE Indicator: provider status page amber; failed-transaction rate > 2%; monitoring alert on checkout endpoint. Mitigation: payment fallback provider wired up by Aug 27; failed transactions queued for retry; customer comms template drafted. Owner: payments dev Cadence: daily during launch week. Status: Open R3 If copy/legal pages aren't final, we launch with placeholder text or wrong pricing. L 4 I 3 Score 12 ACTIVE Indicator: content sign-off not done by Sep 1; open content tickets > 3; pricing page still draft. Mitigation: content freeze Sep 1; named sign-off checklist per page; owner chases daily from Aug 28. Owner: product manager Cadence: weekly until freeze, daily after. Status: Open R4 If old URLs aren't redirected, we lose SEO rankings and 404 on inbound links. L 3 I 3 Score 9 WATCH Indicator: redirect map incomplete; crawl of staging shows 404s on old paths; sitemap not resubmitted. Mitigation: redirect map built by Aug 29; staging crawl checklist run; sitemap submitted day of launch. Owner: marketing/SEO Cadence: weekly. Status: Open R5 If the key on-call engineer is unavailable at launch, no one can fix a live incident. L 2 I 4 Score 8 WATCH Indicator: runbooks not written; single owner on critical tasks; no secondary contact named. Mitigation: runbooks for top 5 failure modes by Aug 30; secondary owner named per critical task; on-call rotation published. Owner: team lead Cadence: weekly. Status: Open R6 If cookie banner / privacy policy isn't compliant, we get flagged or fined post-launch. L 2 I 4 Score 8 WATCH Indicator: legal review not started by Sep 1; cookie banner untested on staging. Mitigation: banner implemented and tested Aug 29; privacy policy live on staging; legal sign-off by Sep 2. Owner: ops Cadence: weekly. Status: Open HOW TO RUN THE FIRST REVIEW (10 minutes) - Walk R1-R6 top-down by score. R2 and R1 get the most time. - For each: read the indicator aloud, owner says "moving / not moving," mitigation deadline checked. - Update status. If nothing changed on a risk for 3 reviews, escalate or close. - After launch, review daily for week 1, then weekly for a month, then close everything and archive. The register works because the meeting is short, the fields force concreteness, and every row has a single person accountable for it.