Bench strength and succession

Applies to any team with at least one role the business cannot afford to lose. Produces a bench scorecard, a performance and potential map, succession plans, and a talent review outcome.

The doctrine: succession asks who can do the next job. Bench strength proves it before you need them. A résumé only proves what someone did in the last system. A succession plan is a development plan with names attached, built years before the vacancy, and potential is a claim that needs evidence.

The sequence

1. Name the critical roles and scan the horizon. A role is critical if losing its holder would put a quarter's number, a top customer, or a key capability at risk within 90 days. Include the leader's own role. Then scan 18 to 24 months ahead: retirements, flight risk, and promotions that will open seats. Read references/bench-scorecard.md.
2. Score the bench. For each critical role: how many qualified backfills, how soon each is ready, and whether the incumbent is a single point of failure. Compute the aggregate bench score.
3. Assess performance with evidence. Results against target, quality of results, and how they were produced. No placement without data.
4. Assess potential with evidence. Four signals: learning speed, aspiration, the altitude the person already operates at, and whether others get better around them. Read references/potential-and-readiness.md.
5. Map and decide readiness. Place people on the performance and potential grid as a conversation tool, then make a readiness call for each successor: ready now, ready in 12 months, ready in 24 months, or not a successor for this role.
6. Run the talent review. Independent pre-scoring, evidence on the table, bias checks, and decisions with owners. Read references/talent-review.md.
7. Write the plans. For each critical role: interim cover, short-term backfill, permanent successor and development plan, and the trigger for an external hire. Build readiness through real cross-functional work, not classroom development. When the plan calls for an outside hire, raise the evidence standard: test whether past success transfers to this operating system. Read references/succession-plans.md.
8. Review quarterly. Re-score the bench, update readiness, and measure in three tiers: actions taken, coverage and readiness moving, and outcomes when transitions happen.

What good looks like

• The expert's tell: every critical role has a named, evidence-backed successor or an explicit external-hire trigger, and the leader's own role is on the list.
• The common mistake: the high-potential list is a popularity list. Polish, visibility, and senior sponsorship stand in for evidence of learning speed and aspiration.
• The second mistake: naming successors without developing them. A name on a slide with no stretch assignment is not a bench.
• The external-hire mistake: reading title and scale as readiness. The most dangerous succession mistakes are outstanding candidates evaluated against the wrong evidence.
• The third mistake: ignoring people who are outgrowing their roles. Top performers with no credible path become someone else's bench.
• How to know it worked: when a critical role opens, a ready successor steps in, performance holds through the transition, and regrettable attrition among top performers stays low.
• Field cases: references/worked-examples.md. Score the review with references/review-rubric.md. Ship at 90 or above.

Rules

• MUST include the leader's own role in the critical-role list.
• MUST back every performance and potential placement with evidence; NEVER place people on impressions.
• MUST assess potential on learning speed, aspiration, operating altitude, and multiplier evidence; NEVER on polish, tenure, or visibility.
• MUST collect independent ratings before the talent review discussion.
• MUST name an interim cover and an external-hire trigger for every critical role without a ready-now successor.
• NEVER mark an external candidate ready now from title and scale alone. Require comparable work, operating-context fit, adaptability, and multiplier fit.
• MUST give every named successor a development action with a date.
• NEVER treat a grid placement as permanent. Recalibrate every quarter.
• NEVER use protected characteristics, or proxies for them, in any placement.
• MUST flag top performers outgrowing their roles and name their next challenge or its timing.