Succession Planning: The Process, the 9-Box Talent Review, and a Template That Gets Used (2026)

A company I will not name had three critical roles on its succession plan, each with a named successor, each marked ready now. Then the VP of Engineering resigned. The successor had accepted another offer six weeks earlier, because nobody had ever mentioned to her that she was on the plan. The second name had never run a budget or written a board paper. The third was real. One of three, and they only found out which at the worst possible moment. That is the normal state of succession planning, and it is not caused by bad intentions. It is caused by confusing a list of names with a plan.

What is succession planning?

Succession planning is the process of identifying the roles an organisation cannot afford to leave empty, working out who could fill them and when, and closing the distance between those two things deliberately. The load-bearing word is process. A list of names against a list of jobs is not a succession plan, and the gap between the two is the subject of the next section.

A working definition, in one sentence: a succession plan is a list of gaps, each with an owner and a date. If your plan contains names but no gaps, it is documentation of an opinion.

Succession planning vs replacement planning

Most organisations sincerely believe they are doing the first and are demonstrably doing the second. The distinction is not academic, because only one of the two builds any capability.

The ten-second test. Pick any name on your plan and ask what has changed for that person since the last review. Not what has changed about the plan, what has changed for them: a new responsibility, a budget, an exposure, a conversation. If the honest answer is nothing, you have a replacement list.

Replacement planning is not worthless. It is genuinely useful for emergency cover, and knowing who would hold the wheel for six weeks is worth having. It simply does not create successors, and organisations routinely mistake having the document for having the bench.

Which roles belong on the plan?

Not the org chart. Ask two questions of every role: what breaks if this person leaves next month, and how long would it take to restore. The list that falls out looks strikingly different from a seniority ranking.

• The obvious senior roles. Genuinely critical, and also the ones with the most external options, so they are usually the least urgent to worry about.
• The single-point-of-failure specialist. The only person who has closed a year-end. The engineer who wrote the billing system. The one name on a regulatory sign-off. These are the real key person risks, and they are dangerous precisely because nobody is watching them.
• The relationship holder. The account lead who personally holds your three largest clients. Their departure takes revenue with it, not just capacity.
• The role your strategy is about to create. If the plan implies a function you do not yet have, its first leader is a succession question today, not a recruiting question in eighteen months. This is where succession and workforce planning meet.

The 9-box grid and the talent review

The 9-box grid is a three-by-three matrix plotting performance against potential, and the 9-box talent review is the calibrated meeting where managers place their people on it and defend the placement. Used well, it is the most useful hour in the whole succession process. Used badly, it is a labelling exercise that leaves everyone slightly worse informed.

Three boxes are misread almost everywhere, and getting them right changes what you do with real people.

Rough diamond (high potential, low performance). The instinct is a performance conversation. The more common truth is the wrong seat: a strong person in work that does not use them, or under a manager who is not developing them. Move them before you manage them out, and if the move does not work you have learned something either way.

Trusted specialist (low potential, high performance). The single most damaging misreading in the grid. Low potential in this framework means limited appetite or capacity for a broader role, not low value. These people are frequently your deepest key person risks and among your most valuable employees. Retain them hard, pay them properly, and stop trying to promote them into management they never asked for.

Star (high, high). Obvious in the room and often neglected afterwards, because stars are assumed to be fine. They are the ones with the most external options. A star with no named next step is a resignation with a delay on it.

Running the meeting.Four rules do most of the work. Define potential in writing beforehand, because otherwise it silently becomes "how much I like this person". Require evidence, not adjectives: a placement is defended with something that happened, in the way a good interview scorecard forces evidence at the hiring stage. Let peers challenge placements, because calibration across managers is the entire point and a manager rating their own team in isolation is not calibration. And end every placement with a named action and an owner, or the box means nothing.

The succession planning process, step by step

Seven steps. Steps 1 and 6 are the ones organisations skip, and skipping either one converts the whole exercise back into a slide.

Readiness ratings and bench strength

Readiness bands drift upward every single review unless they are defined in writing beforehand. Left undefined, ready now quietly comes to mean well regarded, and the plan stops describing reality about eighteen months before anyone notices.

Bench strength is the share of critical roles with at least one credible internal successor, and it is the one number worth taking to a board. Three rules keep it honest.

• Express it as a fraction of roles, not a count of people. One strong candidate named against four roles is not four successors, it is a concentration risk pretending to be coverage.
• Report the trend, not the snapshot. Ready-now coverage moving from 3 of 12 to 6 of 12 over two years is evidence the process works. A single snapshot proves only that a spreadsheet got filled in.
• Publish the "no cover" count deliberately. It is the most valuable line on the sheet. It converts into a budget request today rather than a crisis in nine months, and hiding it to make the plan look complete is the single worst thing you can do with a succession process.

Pair bench strength with one outcome measure, usually the share of senior vacancies filled internally, which sits naturally alongside your other recruitment metrics and is the number that proves the plan produced something.

A succession planning template that gets used

One row per role, nine columns. Templates fail by being too elaborate, so the test for adding a column is whether a missing value in it would change a decision.

The first six columns are what everyone builds. GAP, ACTION and BY WHENare the three that almost nobody adds, and they are the entire difference between a plan and a list. They convert the document into work that appears in somebody's quarter, which is the only reason any process survives contact with a busy organisation.

Keep the emergency layer separate and short. For each critical role, one line: who holds the wheel from day one if this person is gone tomorrow, what they are authorised to decide, and where the credentials and the documentation live. That is a different question from who should eventually get the job, and merging the two is how organisations end up with neither.

How to tell someone without promising them the job

This is where most succession processes quietly die. The fear is understandable: name someone as a successor and you have created an entitlement, and if the role goes elsewhere you have manufactured a resignation.

The risk in the other direction is larger and much better evidenced by everyone's own exit data. Strong people leave because nothing visible is happening to them, and they leave without ever knowing they were on a slide. Silence does not protect you from the downside, it just relocates it somewhere you will not see it coming.

The wording that resolves the tension separates development from promise. Four sentences:

• "You are someone we are developing for bigger roles." A statement about investment, not about a vacancy.
• "Here are the two gaps I think stand between you and that." Specific, evidenced, and honest. This is also the part that makes it credible.
• "Here is the assignment that closes the first one, starting next quarter." The proof that the sentence before it was real.
• "I am not promising a specific job, because the timing depends on things neither of us controls." Said out loud, early, it protects both of you and nobody has ever resented hearing it.

Two things to avoid. Do not name a date for a role that depends on someone else resigning, because you are promising something you cannot deliver and creating an interest in a colleague's departure. And do not let it be a single conversation: an annual mention with nothing in between reads as flattery, which is why the development action and the quarterly check matter more than the announcement.

When the answer is an external hire

A succession plan that never concludes "we need to hire" is not being run honestly. Roles with no credible internal successor are normal, particularly in small companies and in functions you are building for the first time, and identifying them early is the point rather than a failure of the process.

What changes is that you are now hiring on your own timetable instead of somebody else's. That buys you:

• A budget line before the vacancy. The difference between a planned search and an emergency one is mostly whether finance heard about it first.
• A warm pipeline instead of a cold start. Twelve months of unhurried conversations, which is exactly what a talent pipeline is for.
• A screen built against the role as it will be,using the three-year requirements you wrote in step 2 rather than the outgoing incumbent's job description. Our guide to screening criteria covers how to turn those into something you can actually score against.

And when you do hire externally into a role that had internal candidates, tell them before they read it in an announcement, with the reason. Handled badly this is one of the most reliable ways to lose two people instead of filling one seat, which is a theme our guide to internal recruitment covers in more detail.

Succession planning for small businesses

Same logic, smaller scope, and considerably more acute risk, because a company of thirty people has nowhere to absorb the loss of the one person who knows how something works.

Start with three to five roles. Accept early that for several of them there is no internal successor, and treat that as the useful output it is: it tells you whether the mitigation is a hire, a documented handover, a cross-trained second person, or an outside service contract. Then prioritise documentation over development, because in a small company most key person risk is knowledge living in one head, and writing down the year-end close or the client history is faster and cheaper than growing a successor.

The one-page version that genuinely works for a small business: for each of your five most critical roles, name who covers it from day one, list the three things only the incumbent currently knows, and set a date to write each of those three down. That is not glamorous and it removes more risk per hour spent than any framework on this page.

The mistakes that hollow out a succession plan

• Limiting it to the executive team. Misses the specialists who are the real key person risks.
• Never telling the successors. The most common cause of losing the person you were counting on.
• Undefined readiness bands."Ready now" drifts into "well regarded" within two cycles.
• Confusing high performance with high potential. Promotes excellent specialists into management they did not want and are not suited to, and you lose the specialist as well.
• Closing gaps with training courses. Vocabulary, not capability. Assignments with real consequences are what develop people.
• Planning against today's job description. You will develop somebody perfectly for a role that will not exist in that form.
• One candidate named against four roles. Reads as coverage on the slide, is concentration in reality.
• Annual-only review. Eleven months stale at the moment you need it.
• Hiding the roles with no cover. To make the plan look complete, which converts a manageable gap into an emergency.
• No action between reviews. The one that subsumes all the others. At that point you are doing replacement planning and calling it succession.

Succession is a retention programme wearing a different name

The second-order effect is the one that justifies the time. A visible, functioning succession process tells everybody watching that internal progression is real here, and that is a stronger retention signal than most things a company spends money on. The reverse is equally true: filling senior roles externally while a succession plan sits in a folder teaches your strongest people exactly what the plan is worth.

It also produces information you needed regardless of whether anyone ever gets promoted. Companies that run honest talent reviews routinely discover that a whole function has no bench, that two of their best people are bored and about to leave, or that a critical process has depended on one person for six years. That is the same information you would otherwise buy expensively through an exit interview after it is too late to act, and it is closely tied to the tenure patterns in your employee turnover rate: exits after three years are usually about progression, which is precisely what succession planning is supposed to make visible.

The whole discipline reduces to one question you can put to any succession plan in front of you: what work does this document create between now and the next review? If the answer is none, the names on it are a forecast rather than a plan, and forecasts do not develop anybody. Start with the roles that would genuinely hurt, be honest about the ones with no cover, and when that honesty turns into an external search, paste the three-year requirements and your applicant batch into Rankid's bulk resume screening so the shortlist is built against the role you are actually filling.