How this track works
Ten practical sessions, one running company: Cartwheel, a mid-size direct-to-consumer home-and-lifestyle retailer - the same company as the sibling course on OKRs, so you are looking at one organisation's people system rather than a fresh set of examples. Across the track you will run a real ramp for a new joiner, coach instead of fix, build a growth plan someone actually opens twice, translate a career ladder into behaviour you can observe, diagnose underperformance before you intervene, run an improvement plan that is a genuine chance rather than paperwork, hold the three conversations everyone avoids, and end an employment relationship with both dignity and a defensible process. One boundary stated up front: this course does not teach goal-setting. Writing the objectives and the metrics belongs to learn OKRs with Phoebe, and we hand you there rather than repeating it.
Ramp, coach, grow, decide 6 min live
Four stages, running continuously and in parallel across your team. Most managers can name the first and the last and quietly skip the middle two, which is exactly why the last one arrives so often.
LiveThe middle two are the job3 min▶
Ramp and decide get all the attention because they have events attached - a start date, a difficult meeting. Coach and grow have no events, only weeks. That asymmetry is why they get dropped, and dropping them is what produces the events.
- Ramp is the cheapest place to fix anything, because expectations have not set yet and nobody is defending a position.
- Coach is where actual capability changes. It is weekly, unglamorous, and the single behaviour Google's Project Oxygen research put first among what great managers do.
- Grow is where you keep the people you want - scope, skills, and a route that is visible to them.
- Decide is a plan or a parting. It should be rare, and it should never be the first time the person hears the concern.
The manager with a "performance problem" every quarter. A Cartwheel team lead raised three formal concerns in a year while a peer with a similar team raised none. The difference was not the people they inherited. One held weekly coaching conversations and said awkward things early; the other saved feedback for review season, by which point the only tool left was the formal one. Stage 4 volume is a diagnostic about stages 1 to 3.
The same problem, ten times the cost 5 min live
One issue - say, work arriving late and needing rework - costs wildly different amounts depending on when you address it. The intervention gets slower, more formal, more expensive and less likely to work the longer you wait. This is the argument for saying the awkward thing in week three.
LiveFour causes, four different interventions4 min▶
Before any intervention, name the cause. Almost every wasted month in performance management comes from applying the intervention for one cause to a problem that has another. You will build this out properly in b6 and use it in the lab in b7:
- Skill - they have not been taught part of the job. The intervention is instruction, then practice, then review. Warnings do nothing to a skill gap.
- Will - they can and have stopped choosing to. The intervention is a conversation about the real reason, before anything formal.
- Fit - the person and the role are genuinely mismatched. The intervention is a different role, here or elsewhere. Coaching harder does not change a mismatch.
- Context - the role was rewritten, the team lost people, the system changed, or the manager is the problem. The intervention is fixing the context. An improvement plan aimed at a context cause is the most expensive mistake in this course.
Note that only one of those four is what a formal plan is actually designed for.
Self-studyWhy "I did not want to demoralise them" backfires2 min read▶
The most common reason managers delay is kindness: the concern feels small, the person is trying, and raising it seems disproportionate. Three things happen while you wait.
- The person loses the cheap chance to fix it. They would have preferred to know in week three, and they will say so later.
- Your own account hardens. Unspoken concerns accumulate into a story about the person rather than a list of behaviours, and stories are much harder to discuss fairly.
- The team already knows. Colleagues absorbing the rework notice that nothing is being said, and read it as either blindness or favouritism.
The kind version is not silence, it is the small early conversation - which is also, conveniently, the cheap one.
Map a Cartwheel team onto the lifecycle ★ 12 min · everyone tries
Five people on Cartwheel's merchandising team. Place each one at the stage they are actually in, then name the stage the manager is actually giving them. The gap between those two columns is the work.
List every person you manage and mark which stage they are genuinely in: ramping, being coached, growing, or in a decide situation.
Now mark what they are actually getting from you - and be honest, because "coaching" that is really a status check counts as nothing.
Circle every mismatch. Someone growing who gets no growth conversation is a resignation forming; someone in a decide situation who is getting vague coaching is a problem hardening.
Pick the single most expensive mismatch and decide what changes this week. One is enough.
| Person | Actually in | Actually getting | The cost of the gap |
|---|---|---|---|
| New buyer, week 5 | Ramp | Ramp - clear bar, weekly check | None. This is working |
| Analyst, 2 years, strong | Grow | Coaching only, no growth talk | Highest flight risk on the team |
| Analyst, forecasting late | Coach - skill gap | Nothing said yet, month 4 | Sliding toward a formal process that instruction would have prevented |
| Senior buyer, disengaged since reorg | Coach - will question | Avoidance | The conversation gets harder every week |
| Coordinator, role rewritten | Decide - context cause | An improvement plan | Wrong intervention entirely. The role changed, not the person |
Rewind three cases to their cheap moment ★ 8 min · do your own
Take three situations that became formal and trace each backwards to the earliest point where a small conversation would have settled it. This is how you learn to recognise the moment while you are in it.
For each case, write when the concern became formal, and when you or someone else first noticed something.
Measure the gap between those two dates. That gap is the leverage you gave away.
Write the sentence you would have said at the earlier date. Keep it to behaviour and effect, no labels.
Ask what made the earlier conversation feel unavailable. Usually it is not knowing what to say, which is precisely what b3 and b8 fix.
The eleven-month gap. Cartwheel's coordinator case went formal in month twelve. Asked when the first sign appeared, three separate people said month one - right after the reorg rewrote the role and the team lost two people. Nobody said anything, because everyone assumed it was temporary and nobody wanted to add to the pressure. The eventual plan then measured a person against a job that had quietly changed underneath them. The lesson is not "act faster on people". It is act faster on signals, and diagnose what the signal is actually about before deciding whose problem it is.
Try it yourself - this week ◐ 30-45 min total
- Map every person you manage onto the four stages, twice: the stage they are in, and the stage they are getting. Circle the mismatches.
- Take the largest mismatch and do one thing about it this week. Not a plan - one conversation.
- For anyone you would describe as underperforming, write which of the four causes you believe it is, and what evidence you have for that rather than the other three.
- Find one concern you are currently sitting on. Write the two-sentence early version you could say on Friday.
- Check whether your team has goals worth coaching against at all. If not, that is the OKR course's b2, not this one.
Sources covered (cited)
Taught from the management-craft canon. This session covers:
Three questions before you go 🎯 ◐ 90 seconds
1 · Which two lifecycle stages carry most of a manager's leverage?
Ramp and decide get attention because they have dates attached. Coach and grow have only weeks, so they get dropped - and dropping them is what produces the decide events in the first place.
2 · An analyst has never been taught the forecasting work the role now needs. What is the intervention?
Name the cause before choosing the intervention. A warning does nothing to a skill gap; teaching does. Will, skill, fit and context each need a different response, and only one of the four is what a formal plan is designed for.
3 · A coordinator's role was rewritten after a reorg and the team lost two people. Their manager opens an improvement plan. What is wrong?
You cannot improvement-plan your way out of a problem that was never the person. When the cause is the role, the resourcing or the manager, the fix is the context. The lab in b7 shows exactly how far a perfect plan gets you on a context case, and it is not far.