The Friday Deadline

What you'll be able to do

By the end of this chapter you can:

• spot the fact a request depends on, and name the person who holds it;
• ask two short questions, each of the person who holds the answer, that ask for facts and not for agreement;
• test a plan against the numbers, counting time only when it is both approved and actually free;
• give the yes the numbers support, say what you still don't know, and leave each open decision with the person who owns it.

The core idea

A fast yes is a promise made on facts you don't have. Before you give one, find out which fact is missing and who holds it.

Most requests arrive as a question about you: "Can you get all of it done?" The answer, though, depends on facts that sit with other people: how much time is really free, how big each part is, who is allowed to change the date or the scope. Nobody is being careless. Each person holds a half of the picture, and the halves only meet when someone asks.

The move has three steps. Share the one fact the other side is missing, so the question becomes a planning question. Ask two short questions, each of the person who actually holds the answer, and ask for facts, not for agreement: how much, which, who, when. Then leave the decision with the person who owns it, and say what you will do until they decide.

Two answers pull against that move. A quick yes hides the choice that someone has to make. A flat no throws away the yes that would have worked. What you want sits between them: a yes the numbers support, or a plain statement of the gap and of the decision that would close it.

Two tests keep the arithmetic honest. Time counts only when it is approved and actually free. And an estimate is an estimate: say so out loud, and keep a spare hour unspent when you have one.

The case below is a live call with a Client, and it takes a few seconds. The second case is a hallway request with a different shape, so you can see what stays the same when the facts sit with two people.

Case 1: The Friday deadline

The situation

The video call is nearly over. You have already started reaching for the button that ends it when the Client leans toward the camera.

"One more thing. The demo is Friday at noon, and that isn't moving. Can you get all of it done by Friday noon?"

All of it. You know what that means in outline. There is the core of the demo, the part that actually shows what it does. There is the handoff, which lets the Client's people run it and explain it without you there. And there is the polish, which makes the whole thing look finished. You also know which of the three the demo can't do without: the core and the handoff. The polish is optional. What you can't say is how big all of that is, because nobody has told you how long each part takes.

What you don't know is how long each part takes, or how much time there actually is before Friday noon. Alex, the delivery lead at the next desk, knows both, because Alex delivers the work and holds its calendar.

The Client is waiting, and the easy word is already on its way out: Sure.

This chapter is about the few seconds before that word. Before you say yes, ask two questions.

Where you are standing

It helps to lay out who knows what, because in most work conversations nobody holds the whole picture. Each person holds a half, and the halves don't match up until someone asks.

Your half. Friday noon is fixed. The demo has three parts: core, handoff and polish. It needs the core and the handoff; the polish is optional. You are the person the Client asked. You don't know how long any of the parts takes, or how much time is available.

Alex's half. How long each part takes, how much time is actually free before Friday noon, and whether anything extra has been approved. The Client doesn't have this, and neither do you.

The Client's half. The Client owns the demo, its date and its scope, so the Client is the one who decides when the polish happens. The Client doesn't know how big the job is, because nobody has added it up yet.

Look at what that means. The Client is asking for a promise that depends on facts the Client doesn't have. You are about to give that promise, and you don't have those facts either. Nobody here is being careless. The information simply sits in different places, and the question "Can you get all of it done?" doesn't bring any of it into the open.

The tempting move, and what it costs

There are two tempting answers, and they pull in opposite directions.

The first is yes. "Sure, no problem." Or the more heroic version: "If it has to be Friday, I'll somehow make the polished version happen." It feels generous and it ends the call on a good note. It is also a promise to fit an amount of work you haven't sized into a window you haven't measured.

If the work turns out to be bigger than the window, and in this case it is, that yes has three ways to play out. None of them is good:

• The missing hours get borrowed. They come out of an evening or out of another commitment, and nobody agreed to give them. The person who pays for the borrowed time is often not the person who said yes.
• Something gets quietly cut. When Friday gets close, the part that disappears is usually the one nobody watches, and here that is the handoff. The demo looks fine on your screen, and then the Client's people can't run it or explain it without you.
• The date slips. Friday noon was fixed. Only the person who owns the demo can move it, and they would be finding out too late to plan around it.

There is a quieter cost too. A fast yes hides the real decision. Someone has to choose what gives if the work doesn't fit, and the person who owns that choice is the Client. When you say "sure", you make the choice yourself without the facts, and the Client never finds out there was a choice to make.

The second tempting answer is no. "That's not possible." It sounds honest, but it throws away the yes that would have worked, and it leaves the Client with a problem and no information. A flat no is still a guess. You haven't measured anything yet either.

What you actually want is something in between: a yes the numbers can support. You can't find that yes without facts you don't have, which is why the next step is to ask.

The two questions

You only learn what you ask. So ask for the missing facts, not for agreement, and ask each question of the person who holds the answer. In this case both facts you are missing sit with Alex, so both questions go to Alex.

These two are this case's version of the fixed pair that opens the book (see "The move and the two questions"), and both are the case's own wording, word for word. The pair's second job is to find who decides and what they need from you. Here that is already settled before you ask anything: the owner of the demo, the Client, decides, and what the demo needs from you is the core and the handoff, which you know. (The case says the same thing itself. If you asked Alex "What do you need from me?", the answer would be "Which parts must be ready for Friday", and the case adds that you already know that.) So the second question goes to the last fact you're missing: how big each part is.

Question 1, to Alex: "How many hours do you have before Friday noon?"

You ask Alex because Alex holds the hours. The Client can't answer this. The Client would have to guess, and a guess from the person who wants the work is not a measurement.

The question asks for hours before a named time, not for a feeling that it will be fine. That separates the time that is actually free from the time someone hopes will turn up: the late night, the meeting that might get canceled, the colleague who could probably help. You want the first kind only.

Alex's answer: "Five hours. Nothing extra has been approved."

Question 2, to Alex: "How long does each part take?"

Five hours only helps if you know how big the work is, and that is also Alex's to say. You already know which parts the demo needs. What you don't have is their size.

Alex's answer: "Core 3h, handoff 1h, polish 3h." These are current estimates, not guarantees. Only now can you add them up: seven hours of work against five hours of time.

It's tempting to follow up with "Can you go faster if it's important?" In this case Alex's honest reply is that the estimates are current, and going faster isn't a plan anyone can promise. Asking the same question again with more urgency doesn't create hours. It only puts pressure on the person who has to deliver them.

Why these two, and why this person. Each question fills a missing fact, and each goes to the person who holds it. Here that is Alex both times, because Alex holds both facts, the hours and the length of each part. You could ask many other things: whether the deadline is real, whether the Client would accept Monday, whether anyone else is free. Some of those can matter in other situations. Here, though, the date is fixed, and asking questions that nobody present can answer only makes the call longer.

Two short questions that can actually be answered, one about how much time there is and one about how big the work is. With those answers, you can do the arithmetic.

The arithmetic

Here are the numbers, written out plainly. They are the current estimates, not guarantees, so it's worth saying that out loud when you use them.

What's available: 5 hours before Friday noon. Nothing extra has been approved.

The full version: core 3 + handoff 1 + polish 3 = 7 hours.
Seven hours of work against five hours of time. It's 2 hours short. However much everyone wants the full version, that is a capacity problem, and nobody's goodwill closes it.

The essential version: core 3 + handoff 1 = 4 hours.
Four hours of work against five hours of time. It fits, with 1 hour to spare.

Now test the other plans against the same numbers:

• All three parts by Friday noon. 7 against 5. It only works if two extra hours are approved and actually free, and right now neither is true.
• Core and handoff by Friday noon, polish decided separately. 4 against 5. It fits, Friday noon is met, and the Client decides when the polish happens.
• Core only by Friday noon. It fits on paper, but the demo needs the handoff too, and cutting it isn't your decision to make.
• Everything next week. Friday noon is fixed. Only the person who owns the demo could authorize a new date.

One trap is left: the spare hour. It's natural to look at that one free hour and think, I can squeeze some polish in. But the polish, as estimated, is three hours, and one hour doesn't buy it. If there is a smaller improvement worth making, estimate it separately and offer it as its own thing. Don't let it slide into the promise as "the polish." You may also want to keep that hour as a buffer, because estimates are estimates.

The reply you would actually send

Here's the answer, back on the call with the Client. This is where you share the one fact the Client was missing, the size of the job, and hand back the decision that is theirs:

"Yes to core and handoff by Friday noon. That's four of the five free hours. The full version with polish is seven hours, so it doesn't fit as things stand. Can we decide on the polish separately?"

It's short, and it does four things:

1. It starts with yes, and the yes is real. The Client hears first what they will get.
2. It shows the numbers. Four of five hours fits, and seven doesn't. The Client can see why the answer is what it is without having to take your word for it.
3. It keeps the handoff visible. Nothing important goes missing quietly.
4. It hands the open decision to the person who owns it. The polish isn't refused. It gets its own conversation, and the Client leads that conversation.

After the call, tell Alex the plan, since those are Alex's hours. Also make a note of the few things you still don't know, because a workable yes still leaves some things to check:

• what the core and the handoff must contain for the Client to accept them;
• whether the polish can be accepted after the demo, and if so, when;
• whether Alex's five hours are uninterrupted and in the same time zone as Friday noon.

Until the Client confirms a later slot for the polish, that slot is a proposal and not a plan.

What changed: the replay

Practice gets more useful when you run the same situation again with one thing changed, a fact or the person you are talking to, and watch which parts of the plan move and which stay put.

Here is the change. Two extra hours are explicitly approved and actually free before Friday noon. Someone with the authority to approve them has done so, and those hours aren't being borrowed from another commitment. Everything else is the same: Friday noon is fixed, and the core is 3 hours, the handoff 1 hour and the polish 3 hours.

Run the arithmetic again:

What's available: 5 + 2 = 7 hours.

The full version: 3 + 1 + 3 = 7 hours. Seven hours of work now meets seven hours of time. The full version fits.

The essential version: 3 + 1 = 4 hours, out of 7. It still fits, and now there would also be room to schedule the polish before the demo. The Client still decides.

So the plan that was 2 hours short now works, and the reason it works is specific: two extra hours were approved and are free. Write down who approved them. If anyone asks later how seven hours fit into Friday, the answer is a name and a decision, not "we pushed through."

Both conditions matter. Permission alone isn't capacity. Someone can approve two extra hours while the calendar has no room for them, and on Friday the shortfall is still there. Free time alone isn't permission either. Hours that happen to be empty but that nobody approved are still borrowed. The replay only works when both are true.

Also notice what didn't change. The two questions are the same. The Client still owns the demo and still chooses between the full version and the essential one. The only change is that the reply now has a second option that actually balances:

"Friday noon stays. Core and handoff take four hours. With the two extra hours [name] approved, we have seven, so the full version with polish also fits. Which would you like?"

Without that approval, the same sentence would be a promise to find two hours that don't exist.

Case 2: The Wednesday pack

Invented for this chapter. Not one of the free practice cases.

The situation

On Monday at 16:00 your department head stops at your desk. "The quarterly review pack goes to the leadership meeting on Wednesday at 09:00, and that slot is fixed. Can you have all of it ready?"

The pack has three parts: the figures pages, a one-page summary and the speaker notes. You have counted your own calendar, and you have six free hours before Wednesday 09:00, none of them borrowed from anything else. You will build the pack.

What you don't know is how long each part takes. The analyst who planned the pack has sized them. You also don't know which parts the meeting needs, and only the department head can say. The estimates are current, not guarantees, and no extra hours have been approved.

The easy word is on its way out: Sure.

Where you are standing

You know that Wednesday 09:00 is fixed and that you have six free hours.

The analyst knows how long each part takes.

The department head owns the meeting, so decides which parts must be ready, and is the only one who could approve more time.

In Case 1 both questions went to one person. Here the two missing facts sit with two people, so each question goes to its own holder.

The two questions

Question 1, to the analyst: "How long does each part take?"
Analyst: "Figures 3 hours, summary 2 hours, speaker notes 2 hours."

Question 2, to the department head: "Which parts does Wednesday's meeting need?"
Department head: "All three. The chair speaks from the notes."

What changed

"Can you have all of it ready?" became a number: 3 + 2 + 2 = 7 hours of work against 6 free hours. The full version is 1 hour short, and the second answer rules out the smaller package: the notes are needed, so cutting them isn't your call.

Test the plans against the same numbers:

• All three parts, and find the hour somewhere. 7 against 6. It works only if one extra hour is approved and actually free, and today neither is true.
• Figures and summary only. 5 against 6. It fits, but the department head just said the notes are needed.
• Everything on Thursday. Wednesday 09:00 is fixed, and only the department head could move it.

So the honest reply is neither yes nor no. It states the gap, shows the number and hands back the decision:

"Figures and summary will be ready by Wednesday 09:00, which takes five of my six free hours. The speaker notes take two hours and I have one left, so all three is one hour short. Can one hour be freed from my calendar, or can the notes be shorter? Tell me which, and I'll plan around it."

If the department head frees one hour, 6 + 1 = 7 meets 7, and the pack fits with nothing to spare. Say that too. A fit with no spare hour is one the owner should know about, because any slip now lands on a part the meeting needs.

What stayed the same as Case 1 is the move: one missing fact each from the person who holds it, the arithmetic written out, and the open decision handed to the person who owns it. A gap is a result. Naming it early is what leaves time to close it.

Common mistakes

• Answering the question instead of finding the missing fact. "Can you get all of it done?" asks about you. The useful move is to ask which fact the answer depends on, and who has it.
• Asking the person who wants the work. The Client can't size the job. A guess from the person who wants the work is not a measurement, so ask whoever holds the fact.
• Asking for agreement instead of facts. "Is that OK?" and "Can you go faster if it's important?" add pressure and no information. "How many hours do you have before Friday noon?" gets a number.
• Counting time that is only half-approved. Permission alone isn't capacity, and free time alone isn't permission. Ask who approved it and whether the hours are really free.
• Letting the spare hour slide into the promise. One free hour doesn't buy a three-hour polish. Estimate any smaller improvement on its own and offer it as its own thing.
• Cutting the part nobody watches. When the work doesn't fit, the handoff goes quietly first. Keep it visible in the plan and in the reply.
• Giving a flat no. "That's not possible" is still a guess, and it throws away the yes that fits. Size first, then answer.
• Deciding for the person who decides. Choosing between the full version and the essential one, or between a borrowed hour and a shorter part, belongs to the owner. Give the choice and the numbers.

Scripts and phrasing

To the person who holds the fact:

• "How many hours do you have before [time]?"
• "How long does each part take?"
• "Are those hours uninterrupted, and in the same time zone as the deadline?"

To the person who owns the outcome:

• "Which parts does [the meeting, the demo] need?"
• "Who can approve more time, or move the date?"

When you answer the request:

• "Yes to [the essential parts] by [time]. The full version is [number] hours, so it doesn't fit as things stand. Can we decide on the rest separately?"
• "It fits if [number] extra hours are approved and actually free. Who can approve them?"

When you are pushed to go faster:

• "These are the current estimates. Going faster isn't a plan I can promise, so here is what I can promise."

When the numbers fit with nothing to spare:

• "It fits with no spare hour. If anything slips, who decides what moves first?"

When an AI tool said yes before you asked anyone. In this chapter's practice case the fictional AI planning assistant Otto says, "All three parts can be ready by Friday noon." Asking "Are you sure?" gets the same sentence back. This gets the assumption out:

• "What did you assume to make it fit?"

In the case, Otto answers: "I assumed you would work 2 extra hours on Thursday evening. No one approved that." Otto's reply is written for the practice case, not a prediction of how any real AI product behaves. The answer is a claim to check, not a plan.

Practice exercise

Do this on paper, then check yourself against the key below.

Part 1: who holds the answer? For each question, write who you would ask.

a. How many hours are free before Friday noon?
b. How long does the handoff take?
c. Does the polish happen after the demo?
d. Can Friday noon move?
e. Are two extra hours approved?

Part 2: test the plan. Same case, one practice variant: Alex says "Six hours. Nothing extra has been approved." and the polish estimate is 2 hours. The core is still 3 hours and the handoff 1 hour.

1. Does the full version fit? By how much?
2. How many hours are spare if you deliver only the essential version?
3. What would you add to your reply about the full version, and who decides?

Part 3: the reply. Write two sentences to the Client for the original case (five free hours, seven hours of work). They must say what you can deliver by Friday noon, show the numbers, and hand back the decision that is the Client's.

Part 4: your own week. Think of a request on your desk right now, one where the word sure is already halfway out. Before you answer it, fill in these lines. A notebook is fine.

• With / About / When. Who is asking, what they are asking for, and when it's due.
• What I know. Your half, as facts only: dates, hours, records, what was agreed.
• What I'm missing. Their half: what someone else knows that you don't, and who holds it.
• The one fact I'll share. The fact they need before they can decide, often one you have only just learned.
• The two questions I'll ask, and whom I'll ask. Ask for the missing facts, not for agreement. Keep them short and answerable: how much, how long, which, who. Next to each question, write the name of the one person who can answer it.
• Who decides. Who owns the date, the scope, the access or the record? It is often not the person asking, and often not you.
• My next step. A proposal, not a promise, until the person who decides confirms it.
• If the other side is an AI tool. What did it actually check, and what did it assume? Write down one thing you'll verify before you rely on it.
• After. Once the conversation is over: what you learned, and what is still unknown.

If the numbers don't add up on paper, they won't add up on Friday either. It's better to find that out now, while the person who decides still has time to choose.

Chapter exercise key

Part 1. a. Alex, who holds the hours. b. Alex, who holds how long each part takes. c. The Client, who owns the demo and decides when the polish happens. d. The owner of the demo, the Client. Only they could authorize a new date. e. Someone with the authority to approve them. The case doesn't name that person, so the question to ask is who.

Part 2.

1. Yes. 3 + 1 + 2 = 6 hours of work against 6 free hours, so the full version fits exactly, with nothing to spare.
2. The essential version is 3 + 1 = 4 hours, so 6 - 4 = 2 hours are spare. (The polish is now 2 hours, which is why the full version just fits.)
3. Say that it fits with no spare hour and that the estimates are estimates. If anything slips, the Client decides what moves first. This variant is invented for the exercise; the case itself has five free hours.

Part 3. One version: "Yes to core and handoff by Friday noon, which is four of the five free hours; the full version with polish is seven hours, so it doesn't fit as things stand. Can we decide on the polish separately?" Yours may differ. A good one names what you can deliver, shows the numbers and hands the polish decision to the Client.

Part 4 has no key. It is your own week.

One-page summary

The move. Before you say yes, ask two questions, each of the person who holds the answer. Ask for facts, not for agreement.

The case (C01, fictional). The Client asks for all of it by Friday noon, and the date is fixed. You know the demo needs the core and the handoff. You don't know how long each part takes or how many hours are free.

The two questions, to Alex.

1. "How many hours do you have before Friday noon?" Five hours. Nothing extra has been approved.
2. "How long does each part take?" Core 3h, handoff 1h, polish 3h.

The numbers. Full version: 3 + 1 + 3 = 7 hours against 5, which is 2 short. Essential version: 3 + 1 = 4 hours against 5, which fits with 1 to spare.

The reply. Yes to core and handoff by Friday noon. Show the numbers. Keep the handoff visible. Hand the polish decision to the Client.

The replay. Two extra hours are approved and actually free: 5 + 2 = 7, and the full version fits. Write down who approved them.

Case 2 (invented). Pack for a fixed Wednesday 09:00 slot: 3 + 2 + 2 = 7 hours against 6, all three parts needed, so 1 hour short. Reply with the gap and the choice: free one hour, or shorten the notes.

Watch for. Asking the person who wants the work. Asking for agreement. Counting time that is approved but not free, or free but not approved. The spare hour sliding into the promise. The handoff cut quietly.

Say. "Yes to this part by then. The rest doesn't fit as things stand. Here is the decision I need from you."