Where we are
Three sessions of getting the room right. Tonight is about the ninety days afterwards, in which almost nobody is thinking about your session and it either changed something or it did not.
★ What you walk out with tonight. A transfer plan with three actions in it that are not training, the two-line manager email drafted, and a number to answer the next "can we fit more in".
LiveWhat this board is, and what it is not2 min▶
Unlike the objective linter in a1, this one is a model rather than a measurement. It is worth being exact about that before you use its numbers in a meeting.
- The shape is well-supported. Unpracticed material decays quickly. Retrieval and spacing beat re-exposure. What happens after the session matters more than what happens inside it. None of those are controversial.
- The digits are calibrated for teaching. They are chosen so the trades are visible and the ordering is right. They are not a measurement of your programme, and the widget says so on its own footer.
- Use it for the ordering, not the decimals. "The manager conversation is worth more than anything I can do in the room" is the durable claim. "It is worth exactly fourteen points" is not.
The room is where the least of it is decided
6 min liveLiveThe manager email, which is the whole thing4 min▶
The largest lever on tonight's board is that somebody's manager asks about the new behaviour once, in the first week. Not a review, not a check-in process - one conversation that signals this is expected.
It works because it changes the incentive row from b1's five causes, which no course can touch. And it costs you an email:
"Your team is in Thursday's data-contracts session. The one thing that makes it stick: ask them in the first week what they classified and what they decided. Ninety seconds, once. That is the whole ask."
- Send it before the session, not after. After, it is a follow-up nobody has time for. Before, it is a briefing.
- Ask for one specific ninety-second thing. "Support the learning" is not actionable. "Ask them what they classified" is.
- Tell the room you have done it. "Your managers have been asked to check in this week." That sentence alone changes what people expect of themselves.
The transfer board
★ 22 min · everyone buildsSix design and delivery choices, one anti-lever, and three meters: the percentage still able to perform unaided at day 1, day 30 and day 90. The dashed line is always the one-off session, so you can see what each choice bought.
82% · 24% · 11%. A one-off session, well delivered, nothing around it. Day one looks excellent, which is the number everybody sees and the reason this design survives.
86% · 35% · 21%. They retrieve it rather than watch it. Day 90 roughly doubles, and the room feels harder - which is exactly why sponsors ask for this to be removed.
83% · 45% · 33%. Spread over three sittings. Day one goes down by three. Same total minutes, no false peak, and day 90 up another twelve. This is the trade people find hardest to accept.
90% · 51% · 37%. Worked examples, then faded. Helps day one substantially and day 90 modestly - it is a learning lever more than a retention one.
92% · 59% · 46%. A one-page job aid they keep. Nine points at day 90 for the cheapest artefact in the whole course, because it is the only thing still present in ninety days.
92% · 70% · 60%. Their manager asks in week one. Fourteen points, more than any classroom technique on the board, and it is a two-line email.
92% · 78% · 69%. They do it for real within five days. Scheduling the first real use is design work, and it usually needs somebody who is not you to agree to it.
Start with the one-off session and read day 90. Eleven percent. That is a good session, delivered well, with nothing around it - and it is the shape of most corporate training.
Switch on the manager lever alone. Day 90 goes from 11 to 25. One lever, no teaching, a two-line email. Sit with that for a moment before doing anything else.
Now climb the whole ladder in order and watch the dashed line stay where it is. The gap between the two curves is what design bought.
Stop at the spacing rung and look at day one. It drops by three. This is the honest cost and it is the reason spacing is hard to sell: designers get judged on day one, which is the number that does not matter.
Get to 69 at day 90. Everything on, anti-lever off. Note that three of the six levers are things that happen outside your session.
Now switch on "fit more content into the same session". Day one goes UP by three and day 90 goes DOWN by nine. This is the most common request a course designer receives, and it improves the only number the sponsor is looking at.
Try the anti-lever on its own, from the one-off session. 85 / 12 / 2 - the best day-one figure on the entire board and the worst day-90 figure. Everything that makes a session look good in the room is on that trade.
Draft your manager email now. Two lines, one specific ninety-second ask, sent before the session. It is the highest-value thing in this course and it takes four minutes.
The sentence this board gives you. Not "that would reduce learning". Instead: "Happy to add it - it takes about nine points off what people can still do in three months. Shall we drop something instead, or accept that?" A trade, with a number, offered rather than refused.
Three things that are not training
5 min liveLiveThe transfer plan, which fits on half a page4 min▶
| When | What | Who owns it |
|---|---|---|
| Before | The two-line manager email | You |
| In the room | Retrieval, worked examples, the job aid | You |
| Day 1 to 5 | They classify a real change on their own service | Their lead schedules it |
| Week 1 | The manager asks the ninety-second question | The manager |
| Day 7 and day 30 | Two questions by email | You, written and scheduled now |
Five rows. Two are yours, three belong to somebody else, and asking for those three is a conversation you have to actually have. That conversation is worth more than another pass over your materials, and it is the part that almost never happens.
Put this half page in the b6 business case. "Here is what I will do and here are the three things I need from you" is a much stronger proposal than a course outline, and it also sets up what you will and will not be accountable for.
Before session b5
25 min- Send one manager email for a real session. Two lines, one specific ask, before the session.
- Write your five-row transfer plan and mark who owns each row.
- Ask one team lead to schedule a first real use within five days. Note whether they say yes - and if not, why not, because that answer is usually a b1 finding.
- Write and schedule the day 7 and day 30 emails for your next session.
- Bring the day-90 number to b5, because b5 is about measuring whether any of this actually happened rather than modelling it.
Three questions
Clear all three and this session gets stamped in your passport.
1 · The single largest lever on the board is worth fourteen points at day 90. What is it?
It is not a teaching technique at all, it costs a two-line email, and it works by changing the incentive - which is exactly the cause from b1 that no course can touch. It also means part of your job is asking somebody else for something.
2 · Spacing the same content over three sittings makes day-one performance go down by three points. Why is that worth knowing?
A single block builds a fluency peak that is not real. Spacing refuses that peak, which is why the day-one number dips and the day-90 number roughly doubles. Knowing the cost in advance is what lets you defend the choice rather than being surprised by the feedback.
3 · A sponsor asks you to fit more content into the same session. What does the board say, and what should you say?
It is the most common request a designer receives and it improves the only number the sponsor is looking at, which is why it keeps winning. A refusal loses the relationship; agreeing loses the outcome. A trade with a number on it lets them make the decision knowingly, which is all you actually wanted.
What this session teaches, and where it came from
The board is a teaching model, not a measurement of your programme - stated on the widget footer and in full, with every constant, in the course's source map. Use it for the ordering, not for the decimals.
Delivery b4 cheat sheet · pin this
You have a model that says what should happen. Session b5 is about finding out what actually did: the four evaluation levels with their honest critique, why the one everybody measures predicts the least, and how to pick the single behaviour that already has a data trail rather than inventing a measurement programme nobody will run.