How to quote deadlines you actually hit — and what to do when a meteor lands on you or a colleague

Another producer joined the project.

It's the kind of line your eye slides right past: joined, fine, there were plenty of people on that project already. Except the producer didn't come alone. He came with new data from the client. Data almost never arrives at the start, when there's room for it. It arrives near the end, when there isn't.

We were building interactive displays for the Ministry of Agriculture. Not the hardware — everything inside it: graphics, animation, multimedia, the actual program a person at a trade show swipes with a finger and something happens in response. New data meant the program's templates had to be redrawn. Not a number tweaked in the corner. Redrawn, and then some.

Ten days until the show.

The math was simple and grim. Drawing is half the job: whatever gets drawn then has to be programmed, and there's nothing to program until the art exists. So ten days weren't ten days of graphics. They were some days of graphics, and everything else got whatever was left. And no matter how you shifted things around, "some days" came out smaller than what the job needed.

Absolute fucking disaster.

Where the two came from

By then I already had a rule. Simple, works for anyone who owns a deadline — take it. But first, where it came from: without that, you won't see what actually broke.

The rule: multiply by two.

It looks so simple you'd walk right past it. I would have too, if I hadn't seen where the two comes from.

There's a whole shelf of polite phrasings. "Build in a reasonable buffer." "Add thirty percent for risk." Very comfortable places to do nothing. You just take the stated timeline and multiply it by two. For me the two is roughly like pi: a constant, not open for discussion.

I didn't find the two in projects. I found it at Burger King.

I worked there as training director for the Northwest region: an enormous number of people move through you, a stream of them. On top of that I was seriously into psychology at the time — crowd, social, managerial. That volume is where the two showed up. When you see the same mistake in one person, it's their character. When you see it in everyone, it's a constant.

Then came a completely different life — the Mil helicopter plant, where a project involves more than one person and more than one department. That's where I started applying the two to deadlines. The people doing the work have their bravado: I'll get it done faster. Management has its own: it needs to be faster. You stand exactly between two fires, and both pull the same direction, both of them completely sincere — nobody here has bad intentions. I argued with everyone and I always multiplied by two. And I always shipped on time.

There are lazy exceptions, of course. Those you multiply by four.

How the math actually works

Here's the mechanics. I hand over a spec or a brief, get timelines back from the team, and multiply by two on the spot. Inside that schedule I set two more deadlines besides the final one. If the project is big and complicated — I cut it into pieces and stick intermediate deadlines inside each piece.

It all rests on one thing: worked on it isn't finished it. Someone honestly worked for two weeks, and that's the plain truth. It has nothing to do with whether it'll be done in two weeks.

How to sell this upstairs

Whether to show that schedule to management is a question with no general answer — you have to look at the specific people and their psychology. With reasonable ones you can talk straight: you show them "worked on it isn't finished it," and they usually agree pretty easily. With unreasonable ones you have to feed them whatever nonsense sticks, as long as it buys you the second timeline and the budget.

And feeding them nonsense isn't self-serving — there's nothing in it for you. The goal is one thing: ship on time and ship it well. The upshot is that those same unreasonable people win in front of their own bosses, because the project landed on schedule. They just never figure out what it cost.

Better to fight once for real numbers than to blow every one of them later.

Where the two stops saving you

Back to those ten days.

Here's the unpleasant part. Everything I described above was in place on that project — and it didn't help. The easy explanation suggests itself: you buffered badly, then. That explanation is wrong, and here's why.

The two was in that project. I built it in the way I always do, and the intermediate deadlines were sitting in the schedule. The two is just about something else.

The two covers friction. Someone gets sick, someone overestimates themselves, an integration goes sideways, a designer misses an edge case. That's not force majeure. That's normal work, and there's a buffer for it. A plan is a hypothesis, not an order, and the two exists precisely so the hypothesis is honest.

What happened here is that a new person entered the project and brought new inputs from the client. That's not a delay inside the plan. That's a different plan, served under the same deadline.

Moving the date wasn't possible, and here's why.

Everything I described above works under one condition: there's a person on the other side you can negotiate with. Show them the schedule, explain "worked on it isn't finished it," win the second timeline and the budget. Government work is double-edged exactly because that condition doesn't hold. Sometimes it's genuinely interesting work — and simultaneously awful, because you're always the one at fault. Their screwup is your problem. Your screwup is also your problem. You either swallow it and do the work quietly, or you don't do it at all. And since you're always the one at fault, they can legally not pay you for a missed date.

So there was exactly one way out: make it.

Ten days at sixteen hours. Every day. In the mornings I was throwing up bile from the lack of sleep. My body refused to call that normal before I did.

We made it. If we'd missed the date, we simply wouldn't have been paid.

Though "made it" is only act one. Then comes payment, and something else begins: they methodically screw you over every defect, including their own, purely to knock the price down. And once the client finally lets go, the antitrust service or the security service shows up and runs a financial audit.

Some relationship.

What to do when someone next to you is panicking

What I kept from that story isn't a method or a life hack. It's a scale.

The thing is, the scale isn't useful to me. What do I need it for — I've already been there. It's useful in a different situation: someone next to you is in a bad place, you want to help, and all you've got in your hands is the urge to say something nice.

When someone near me is losing it because the client changed a feature, added another one, cut a third, or backed out entirely — I look at it calmly. Not because I'm wise. There's just a mark inside me, and that person is a very long way from it.

And here's the dumbest thing you can do: tell them "yeah, well, mine was worse." You haven't walked in their sandals. Your scale is yours, their panic is real, and there's nothing here to compare.

So I do it differently.

First I agree. I say it straight: yeah, this situation is genuinely fucked. Not "come on, it's nothing," not "it could be worse." Once the disaster is acknowledged, the person stops defending themselves.

And then I ask them to imagine they've been fired.

Picture where you'd go. Your favorite café, sitting down with a coffee. Or the park, your usual spot. So you're sitting there, and you've been fired. What do you feel? Is it the end of the world — or the start of something? Maybe it's a gift: someone finally made the decision for you to leave a job that had been grinding you down. Or a reason to move to another company for more money.

Usually by this point the person sees it themselves. Then all that's left is to say: look, even in the most fucked scenario there's a pile of upside for you, and it isn't scary at all.

And for the future — two things. Multiply everything by two. And keep in mind that one day something will happen that you can't influence: running into Angelina Jolie in Moscow is unbelievable, but the odds, while approaching zero, are never equal to it.

The two covers almost everything. Nothing covers you from Jolie.

I still multiply by two, and it still works. I just know now where the two ends. And if you ever hit that edge — it won't be because you counted badly.