The day a key leader hands in their notice, most organisations discover the truth about their succession plan: it was a slide, not a pipeline. The name in the box has the title but not the readiness, because nobody invested in getting them there. Career development and succession planning are the same discipline viewed from two ends, one grows people over time, the other makes sure that growth has produced someone who can step up when a seat falls empty.

The quick version

  • Career development is the ongoing work of growing a person's capability, mostly through the right job experiences, not training courses. The classic shorthand is 70-20-10: roughly 70% from challenging assignments, 20% from people, 10% from formal learning.
  • Succession planning is making sure you have ready or near-ready people for the roles that would hurt most to lose, not just the CEO, but every position whose sudden vacancy would stall the business.
  • The common tool for reading talent is the 9-box grid: performance now on one axis, future potential on the other. It is a conversation-starter, not a verdict.
  • The trap is "names in boxes", a tidy chart that gives false comfort. A succession plan is only real if the named successors are actually being developed toward the role.

The idea in depth: people grow on the job, not in the classroom

If you want someone ready for a bigger role in two years, the most important thing you can do is give them harder work now, not enrol them in a course. That instinct has an evidence base. Researchers Morgan McCall, Michael Lombardo and Robert Eichinger, working at the Center for Creative Leadership, studied how successful executives actually developed and found that the dominant source of growth was demanding, stretch experiences on the job, stalled projects, turnarounds, first-time people leadership, scope they had not handled before. The finding was distilled into the 70-20-10 guideline: about 70% of development comes from challenging assignments, 20% from developmental relationships (managers, mentors, feedback), and 10% from formal coursework.

The practical consequence: treat job assignments as your primary development tool, and be deliberate about them. In your next development conversation, drop "what training do you want?" and ask instead, "what is the next experience you have never had, and how do we engineer it into your real work this quarter?" A first stint owning a budget. Leading a cross-functional project. Fixing something broken. Presenting to the board. Those are the curriculum. Training supports them; it does not replace them.

One honest limitation. The original 70-20-10 numbers came largely from executives self-reporting how they believed they had learned, a method with known weaknesses around memory and bias, and the precise ratio was never meant to be a law. Treat it as a corrective to the over-investment in classroom training, not an exact recipe. What has actually held up across decades of CCL's "Lessons of Experience" research is the defensible claim underneath: experience is the main teacher of leaders. The tidy percentages are the part to hold loosely.

Reading talent without fooling yourself: the 9-box grid

Once you are developing people deliberately, you need a shared language for who is ready for what. The most common one is the 9-box grid, a simple matrix plotting current performance against future potential, giving nine cells from "low performance, low potential" up to the top-right "high performer, high potential" usually flagged as a future leader. Its lineage is older than HR: the grid descends from a strategy tool McKinsey built for General Electric in the 1970s to compare business units, later adapted to compare people (the GE–McKinsey matrix is the strategy ancestor).

flowchart TD
  subgraph Potential["↑ Potential"]
  end
  A(["High potential /
lower performance
invest, unblock"]) --- B(["High potential /
solid performance
stretch now"])
  B --- C(["Star
successor track,
retain hard"])
  D(["Enigma
diagnose the gap"]) --- E(["Core / growing
develop steadily"])
  E --- F(["High performer
grow or reward in role"])
  A --- D
  B --- E
  C --- F
The 9-box, read as a development map rather than a ranking, each cell implies a different move. Leaders Loop

Use the grid as a conversation, then, not a scoreboard. Its value lives in a room of managers calibrating together, "you rated her high potential; on what evidence?", which surfaces hidden talent and exposes the manager who hoards or under-rates their people. A well-placed name in the top-right is a signal to accelerate development and guard against losing them. A "high performance, low potential" read is a prompt to reward someone in role rather than push them into a job they will not thrive in.

The catch. "Potential" is the slipperiest word in talent management, and the 9-box quietly invites bias to wear a number. Ratings drift toward people who resemble the rater, who are visible, or who are simply well-liked; once someone is labelled "low potential," the label can become self-fulfilling as opportunities stop flowing to them. The grid is only as honest as the calibration behind it. Run it with evidence and challenge, revisit placements regularly, and never show an employee a box as a verdict, it is a snapshot of a judgement, and judgements are what should be argued about.

Why the bench is usually empty, and how to fill it

Here is the uncomfortable part. Most organisations are worse at succession than they think. In a widely cited survey of senior HR and board members run by Stanford's Rock Center for Corporate Governance with Heidrick & Struggles, the picture was stark: many companies grooming a CEO successor still felt unprepared, and a striking share reported having zero ready internal candidates for the top job at all (David Larcker and colleagues, "CEO Succession Planning Lags Badly," Stanford GSB). The same gap shows up below the C-suite, in the director and senior-manager roles that quietly run the place.

Ram Charan put the diagnosis bluntly in "Ending the CEO Succession Crisis" (Harvard Business Review, 2005): succession fails because companies treat it as an event, a scramble triggered by a departure, rather than a continuous discipline of developing leaders for years beforehand. Charan, Stephen Drotter and James Noel turned that into a framework in The Leadership Pipeline: leadership is not one skill but a series of distinct transitions, managing yourself, then others, then managers, then a function, then a business. Each demands you let go of what made you successful at the last level. The person brilliant at the work often makes a poor manager of the work, because the job has genuinely changed.

A succession plan is only as real as the development happening behind the names.

Which is why succession has to be a quarterly habit, not an annual panic, and anchored on transitions rather than titles. Identify your critical roles, the handful of positions whose sudden loss would actually hurt, and for each, name who could step in now, who could be ready in a year, and who is a longer-term bet. Then, crucially, give each named successor the experience that closes the gap to the next transition: a first-line manager who might run a function needs to lead a manager before the seat opens, not after. The chart is the easy part; the development behind it is the plan.

flowchart LR
  A(["Identify critical roles
(loss would hurt)"]) --> B(["Name successors:
ready-now / 1-yr / longer"])
  B --> C(["Find each one's next
transition (Pipeline)"])
  C --> D(["Engineer the stretch
experience now (70-20-10)"])
  D --> E(["Review every quarter,
re-calibrate (9-box)"])
  E --> B
Succession as a loop, not a slide: name, develop, review, repeat. Leaders Loop

A worked example

Take a regional operations team, the leader, Priya, runs logistics for a distributor. (Illustrative people throughout; a teaching example, not a real organisation.) Her deputy, Sam, is the obvious successor: top performer, knows the operation cold, the name everyone would write in the box. On paper the plan is "done."

Run Sam through the lens above and the comfort evaporates. On the 9-box, Sam is a clear high performer, but honest calibration reveals every example of Sam's strength is individual excellence: solving hard problems personally, being the expert. There is almost no evidence of the next transition, getting results through other managers rather than doing the work. That is the Leadership Pipeline gap, and no training course closes it.

So the development becomes concrete and 70-20-10-shaped. Over the next two quarters, Priya hands Sam a real stretch assignment: lead the team-lead group through a depot consolidation, owning the outcome but barred from doing the hands-on work that is Sam's comfort zone. A monthly coaching conversation (the "20") focuses on one thing, delegating instead of rescuing, and a short course on performance conversations (the "10") supports it. Twelve months on, Sam has either grown into a credible ready-now successor with evidence of leading through others, or the experiment has surfaced, safely and early, that Sam is a star individual contributor better rewarded in role, and the real successor is someone else. Either outcome is a win: you learned the truth before the seat fell empty, not after.

Frequently asked questions

Isn't succession planning only for the CEO and the board?

No, that framing is exactly why most plans are thin. The discipline applies to any critical role: a position whose sudden vacancy would stall a project, lose a key relationship, or leave irreplaceable knowledge walking out the door. In many organisations the most dangerous gap is not the CEO (who gets board attention) but a handful of senior specialists and middle managers nobody has a backup for. Start by listing the roles you would panic about losing.

What's the difference between a high performer and a high-potential person?

Performance is about results in the current job; potential is about the capacity to succeed in a bigger or different one. They often overlap but not always, plenty of brilliant specialists have low appetite or aptitude for leading others, and pushing them upward fails everyone. The Leadership Pipeline framing helps here: potential is really the readiness to make the next transition, which demands letting go of the skills that made you a star. Develop high potentials toward leadership; reward high performers in role.

We're too small for formal succession planning. What do we do?

Small teams need it more, not less, because losing one person is a larger share of the whole. You do not need software or a nine-box workshop. You need a single honest conversation, repeated each quarter: for each role that would hurt to lose, who could cover it, and what one experience would make them readier? Cross-training and deliberate stretch assignments are succession planning, they just do not call themselves that.

Should we always promote from within?

Not always, but a healthy internal pipeline is worth building because external hires into senior roles carry real risk, they take longer to deliver, cost more, and fail more often than internal promotions, partly because they lack the context and relationships an insider already has. The point of succession planning is to give yourself the choice: a strong internal candidate means you hire externally because the outside option is genuinely better, not because your bench was empty. An external hire can also bring needed fresh thinking; the goal is a real decision, not a forced one.

How do I develop someone when I can't promote them yet?

Growth does not require a new title. The 70-20-10 logic says most development is experiential, so widen the current role: hand over a project you would normally own, have them deputise for you in a forum, give them a cross-functional problem, let them mentor a junior. These stretch the same muscles a promotion would, keep an ambitious person engaged while they wait, and give you evidence of readiness before you bet a real role on it.

Related in the Toolkit

Career development sits in the middle of the talent lifecycle, it depends on the calibrated read of who is ready (performance & potential), and it draws on the range of leadership approaches a growing manager has to learn (leadership styles & models) as they move through each pipeline transition.

Where to go next