Where we are
Session 1's reader block was built out of assumptions. Some of them are wrong, and the wrong ones are why three careful updates produced nothing. Tonight is ten minutes of asking that would have saved all three.
★ What you walk out with tonight. Two questions you can ask before drafting anything, a read on why your last message got silence, and at least one assumption about your reader that turns out to be wrong.
LiveThe failure this session prevents2 min▶
Daybreak's data lead has been writing updates about why the redesign is late. Careful, honest, detailed explanations of the payments integration.
The founder does not care why it is late. She cared about that in week one. What she needs now is to know what she tells the board on the 30th, and she has been waiting for somebody to give her a version of the answer she can say out loud.
Three updates. All accurate. All about the wrong thing. Ten minutes of asking, three weeks earlier, would have found that - and this is the ordinary shape of the failure, not an unusual one.
Two questions, asked before you write anything
7 min liveNot a discovery process. Two sentences, sent in a chat message or asked in a corridor, and they change what you write more than any amount of editing will.
LiveHow to ask them without it being strange4 min▶
Both questions sound blunt written down. In practice they are ordinary, as long as you attach them to something concrete:
- "I am about to write up the redesign status. What would be most useful for you to have from it?" That is question one, and it invites the decision rather than demanding it.
- "Is there anything about this you are worried about that I should get ahead of?" That is question two. "Get ahead of" is the phrase that makes it safe - it offers help rather than requesting confession.
- Ask before you draft, not after. Asking after produces polite agreement with what you already wrote, because the person can see you have done the work and they do not want to waste it.
If you can only ask one, ask the second. "What would you do with that" often gets a vague answer because people have not thought about it. "What are you worried about" gets a specific one almost every time, because they have absolutely thought about that.
"I wanted to give you an update on the redesign - so the payments integration turned out to be much more complex than we scoped, and what happened was..."
You are now four minutes into an explanation while the founder waits for the part that affects her. She will be polite. She will also leave without what she needed, and you will both feel the conversation went fine.
"Before I write the redesign update - what would be most useful for you to have from it? And is there anything about it you are worried about that I should get ahead of?"
"Honestly, I need something I can say to the board on the 30th. And I do not want to look like I do not know what is going on in my own company." Ninety seconds, and the entire message you were about to write is now the wrong message.
Listening to understand, and reading the silence
6 min liveMost people listen by preparing their answer during the other person's sentence. That is not a character flaw, it is what happens when you are anxious about your turn - and it means you hear the topic and miss the thing.
LiveThree moves that make you a better listener today4 min▶
- Do not prepare your answer while they talk. You will lose the last third of what they say, which is where people put the thing they were unsure about saying. Accept a two-second pause before you reply - it reads as considered, not slow.
- Reflect back before you respond. "So the thing that matters is the 30th, and the explanation is not the point." One sentence, and it does two jobs: it proves you heard, and it gives them a cheap chance to correct you before you go and act on a misunderstanding.
- Ask one more question than feels natural. The second answer is almost always more specific than the first. "Anything else?" is the entire technique.
"So what you need is a board line, not an explanation - and the thing to avoid is you hearing about a change in the room." If that is wrong, you find out in four seconds. If it is right, you have just been given permission to delete most of your draft.
Self-studyWhat a non-reply is telling you4 min▶
Silence is data. It is just ambiguous data, and there are only about four things it usually means:
| Silence probably means | The tell | What to change |
|---|---|---|
| There was nothing to answer | No ask, or an ask with no date | Session 1. Add a verb and a deadline |
| It was too long to start | Sent at 500 words to a phone reader | Session 4. Cut to the ask |
| Wrong person | They cannot authorise the thing | Ask who can, directly |
| The decision already happened | Silence plus a change you were not told about | This is a relationship problem, not a writing one |
The fourth is the uncomfortable one and it is worth naming: if decisions in your area are being made without you and you find out afterwards, no amount of clearer writing fixes it. That is a conversation, and session 6 is where it lives.
Ask, hear, and delete most of your draft
★ 22 min · everyone buildsYou will do this for real between sessions. Tonight you rehearse it on Daybreak, then write the two questions for your own reader in words you would actually use.
Predict. Write one line: what you think they need and what you think they are worried about. Two sentences. Sealed, so you can check yourself afterwards.
Ask. The two questions, attached to something concrete. "I am about to write up X, what would be most useful?" and "Anything about it you are worried about that I should get ahead of?"
Reflect. Say it back in one sentence and let them correct it. This is the cheapest error-check available anywhere in this course.
One more. "Anything else?" The second answer is more specific than the first, almost every time.
Rewrite. Now go back to your session 1 ask and change it. If it did not change at all, you either asked badly or you already knew - and the first is more likely than the second.
Write your two predictions. What your reader needs, what they are worried about. Sealed. This is the same discipline as the pre-registered guess in learn-product-design-with-phoebe, and for the same reason.
Write the two questions in your own words. Not mine. If "what are you worried about" is too direct for your relationship with this person, find the version that is not - "anything I should get ahead of" works in almost every culture and hierarchy.
Now do the Daybreak exercise. Read the founder's ninety seconds above and write down the four things in it that change the message. Then check them against the pins.
Notice what you would have replied to. Most people answer the payments question, because it is the one they have an answer ready for. Answering the thing you are prepared for is the single most common listening failure and it feels like competence.
Rewrite the Daybreak ask using what she said. The old one was about scope and date. The new one has to give her a board line and stop her being surprised again, which is a different sentence entirely.
Write your reflect-back sentence. "So what you need is X, and the thing to avoid is Y." Practise saying it out loud once. It feels unnatural for about a week and then it becomes how you end every briefing.
Diagnose one real silence. Take a message of yours that got no reply and put it against the four-row table. Be honest about which row it is. If it is the fourth, that is session 6.
If you take one thing from tonight. Ask before you draft. After you have written it, you are no longer asking a question - you are asking for approval, and you will get it, and it will not help you.
Before session 3
20 min plus one conversation- Write the two predictions, then actually ask. One real person, before your next real message. Ten minutes.
- Mark what you got wrong. Explicitly. If nothing was wrong, check whether you asked after drafting rather than before.
- Use the reflect-back once. "So what you need is X." Note what happens to the rest of the conversation.
- Ask "anything else?" once and write down the second answer separately from the first. Compare how specific they are.
- Diagnose one silence against the four rows. Bring your answer to session 3, because the fix for two of the four rows is structural and session 3 is structure.
Three questions
Clear all three and session 2 gets stamped in your passport.
1 · Why ask "what would you do with that?" before drafting rather than after?
People are kind. Shown a finished draft, they will not tell you it answers the wrong question, because that costs you a week and costs them an awkward conversation. Asked beforehand, the same person tells you plainly. The timing is the whole technique.
2 · The founder says "I do not really mind that it slipped, things slip." What should you do with that?
That sentence is a gift: it deletes 80 percent of three previous updates. Treating a clear statement as politeness is how people end up writing the same ignored message repeatedly. If she wants the detail later she will ask, and now you will know she means it.
3 · You send a clear ask with a date, to the right person, and get silence - and then find the decision was made without you. Which row is that?
The first three rows of the silence table are fixed by writing better. The fourth is not, and pretending otherwise leads people to endlessly polish messages while being routed around. That is a conversation to have, and session 6 is where it is taught.
What this session teaches, and where it came from
The Daybreak conversation is written for teaching. Full note in the course's source map.
Session 2 cheat sheet · pin this
You now know what she needs and what she fears. Session 3 is where it goes on the page in an order: the ask at the top, three skeletons worth memorising, and the buried-lede failure with an actual number attached - because "your ask is in sentence 4 of 4" is a much harder thing to argue with than "this feels a bit long".