Why your team pads every estimate (and why they're right)

Below the Waterline · Episode 09 · Blog

There are two numbers for every job your company does. The number your team would use if the work went exactly the way it was planned, and the number they actually use. You have only ever seen the second one, and the difference between them is real money.

Why does every estimate have a cushion in it?

Think of a document that leaves your company marked final. The estimate. The scope. The schedule. The spec. The proposal. Whatever the word is in your business.

Now think about how often that document is exactly what happens.

For most companies, the honest answer is almost never. Something changes, and it changes on nearly every job.

Here's the part to really get. Everybody already knows it's going to change. Ask the person who builds that document. Ask the person who receives it. Ask the person who does the work. Every one of them will tell you without hesitating that it changes, and most of them can tell you exactly the ways it usually changes.

They all know. And the document still goes out saying final.

So somebody built a cushion into it. Not a padded invoice, nothing shady. An allowance. A little extra time, a little extra material, a slightly higher number. As Dori puts it, it is "put there deliberately by somebody genuinely great at their job, who has been burned before and does not intend to be burned again."

Three questions to ask about the cushion in your business

Go find it. It's there. Then ask three things about it.

1. Who decided how big it should be?
2. When is the last time anybody checked whether that's still the right size?
3. What, specifically, is it protecting against?

The third one is the one nobody can answer quickly. You'll get "field conditions," or "the client always adds things," or "that vendor is unreliable." Those are real answers. None of them are specific enough to do anything with.

Takeaway: the cushion is not hidden. It is unexamined, which is different, and the difference is why nobody has ever gone looking.

Listen now

Is it bad that my team pads estimates?

No. And this is the part most founders get backwards on the way to a conversation they were about to have badly.

That cushion is not a management problem. It is not a discipline problem. It is not even a mistake. Given what they know, and given what they're handed, it is the correct professional call. If your team took it out tomorrow, you would have a worse company by Friday.

Which is exactly why it has never been examined. There is nothing wrong to point at. The work is good. The jobs get done. The clients are happy. Your people are excellent.

And you are paying for something on every single job that has never once appeared on a line item.

You will not find it by looking harder, because it doesn't sit in one place. Part of it is in the estimate. Part of it is in a conversation between two of your people that you weren't in. Part of it is a decision somebody made on a Tuesday that was completely correct and that nobody wrote down.

No single person in your company sees enough of it to call it a pattern. Which means the only person positioned to see the whole thing is you, and you are the one person it never reaches.

So what is the cushion actually made of?

Your team isn't padding because they're cautious. They're padding because they're good. And the day they stop being good is the day you have a real problem.

So the cushion is not the thing to fix. The reason they need it is.

Takeaway: correct the design, never the person. A cushion built by a capable person is information about your process, not evidence about their character.

What's the difference between a trigger failure and a format failure?

A handoff is any point where important information changes hands, between functions and inside them. There are five questions to ask at any handoff: where the information comes from, where it needs to land, what makes it move, what shape it travels in, and how it breaks.

Question three is the trigger. Question four is the format. This episode is entirely about question four.

The trigger is the one most often missing. Ask a room of founders what fires a handoff in their company and somebody will answer "Janet remembers." But most often missing is not the same as most critical, and that distinction is where the money is.

The trigger decides whether the handoff happens. The format decides whether it produces the intended result.

Miss the trigger, and nothing moves. Somebody eventually notices, because there is a thing sitting still.

Miss the format, and the work moves. It arrives. And then somebody has to make up the difference.

The three currencies your company pays

Somebody makes up the difference, and they make it up in one of three ways.

Time. They come and ask. Somebody walks over, or picks up the phone, and gets what the format didn't carry.

Judgment. They don't ask. They decide it themselves, on the spot. If your team is seasoned enough and solid, they decide it well enough.

Money. Somebody prices the uncertainty in before the job ever starts, because they know the time and judgment costs are coming.

Money is the one that leaves a mark, and it leaves it somewhere you would never think to look, which is your own estimate.

You have probably seen all three of these this week without recognizing any of them. The currency of time was the person who stopped by to check one thing before they sent it. The currency of judgment was a decision somebody made on their own that turned out fine. And the currency of money was the number your estimator wrote down before either of those had happened, because they already knew both were coming.

Takeaway: a stalled handoff announces itself. A completed one never does, which is why format failure costs more and gets noticed less.

Why can't I see this in my own company?

Because your people are good at their jobs.

Format failure in a company full of average people would be loud in its own way. The trigger fires, the work arrives incomplete, somebody makes a poor call on the spot, and the failure shows up further down the line. That is why so many companies conclude they have a people problem when what they have is a systems problem.

Format failure in a company full of excellent people is silent. The work arrives incomplete relative to what the receiver has to decide, and the receiver, being good at their job, completes it.

Your people are the reason you can't see this. Not because anyone is hiding anything. Because they are absorbing it, correctly, every day.

If you have ever been proud that your team just figures things out, the pride is accurate. It is also the exact reason this is invisible in your business, and it is why the business is so dependent on specific people, particularly on you.

What does an undesigned handoff actually cost? A client story

Dori works with a company where part of the business is drilling holes in the ground. As part of scoping and pricing a job, an engineer designs where all the holes go, and that drawing is serious work. The location of those holes determines how the whole job performs and how long it takes. They are the answer at the end of a real chain of engineering reasoning.

Then the crew goes to the site.

There's a boulder where a hole is supposed to go. There's a power line nobody drew, and you can't drill that close to a pole. The slope of the ground is different.

It would be easy to say somebody messed up, and from the crew's perspective it sure feels that way. But nobody messed up. This is a company that does excellent work, and the reason the work is excellent is precisely that they know the site is going to be just enough different from the plan that they figure it out. It isn't an anomaly to them. It's a normal Tuesday. The crew is skilled, they read the conditions, they make good calls, and the job gets done right.

Here's the part that costs money.

The engineer knows all of that too. So when the engineer sits down to draw, they are not drawing for the site on the piece of paper. They are drawing for the site the crew will probably find.

Say the job needs fifteen holes if the drawing were executed exactly as drawn. Because the engineer knows things will move, they draw twenty. (These numbers are illustrative, not this client's data.)

Those extra five are not waste. They are not a mistake. That is a good engineer making the correct decision with what experience has taught them is very often incomplete information, and drawing fifteen and being wrong would cost far more than drawing twenty and being safe.

But somebody has to pay for the extra five. Either the bid goes up and you are slightly less competitive on every job you quote, or the bid holds and the margin absorbs it. Most companies do some of both, and almost nobody has ever put a number on it.

That is what an undesigned seam costs. Not a failure anybody could point at. Five extra, on every job, forever.

Takeaway: the cost of an undesigned handoff is never a line item. It is a small, permanent tax paid in competitiveness or in margin, and usually in both.

Where does the fix actually live?

Look at what would have to be true for that engineer to draw fifteen.

They'd need to know what the site is actually like before they draw it, which means somebody walks it first, with the drawing, to get the reality against the picture.

And they'd need a real working relationship with the crew, so the site lead could talk directly to an engineer in the moment they're drilling.

Neither of those is a document. And neither of them is a step anybody skipped. They were never in the process at all, for a variety of reasons, most notably because the company is staffed with people capable enough to work around their absence. The seam never broke, so nobody ever had a reason to design it.

That is almost certainly true in your business too. We have been told that the accountability chart is the solution to things falling through the cracks. It isn't. The chart answers who and what. The missing layer is the one that actually designs the handoffs, and Dori calls it the Execution Blueprint.

And if it doesn't feel like you have this problem, that's the tell. The seams you have never designed are the ones your best people are quietly covering. That is exactly why you have never had a reason to look at them.

Why doesn't anyone design these handoffs?

Here's the reason, and see if you agree.

A format that can carry uncertainty has to admit that uncertainty exists.

The drawing says twenty holes because writing "somewhere between fifteen and twenty, depending on what we find" feels like not knowing your own business. The estimate carries a cushion instead of a range, because a range looks like hedging. The scope says final because final is what a professional sends.

Certainty is what you're selling. To the client, to your team, and to yourself.

Often, that pattern signals something sitting below the waterline. It is rarely really about the document. It is about what you believe you are supposed to look like to the people counting on you.

And here is the relief in it. Your people already know. They have been pricing the uncertainty in for years, quietly and competently, on every job. Writing it down doesn't create it. It just stops making everybody pay for it privately.

The person who built the cushion

That number isn't only protecting the job. It's protecting them.

Sometime in the past, they gave a number without enough room in it. It went badly, for the project and for them. So now the number carries a little insurance. Some of that insurance is for the company, and some of it is for them, likely without their ever having thought about it that way.

So if you go and ask what that cushion is protecting against, understand what you are actually asking. You are asking somebody to hand you a risk they have been carrying privately for years.

Do that well, and that person will be grateful they don't have to carry it alone any longer.

It is also true that some of that cushion is irreducible. There is real uncertainty that no amount of Execution Architecture can design out, so this isn't about getting to zero. You are trying to find out which part is true uncertainty and which part is an undesigned handoff.

Takeaway: the conversation about the cushion is a conversation about risk somebody has been holding alone. Lead with that, not with the number.

How do you fix a handoff that keeps costing money?

Two moves, and neither one is a documentation project.

1. Let the format carry the range, not just the point

Not "the hole goes here." Instead: here is the intent, here is what's fixed, here is what can flex and how far, and here is what to do when it can't flex far enough.

A drawing that says where the design can bend is a better drawing, not a weaker one. And the person standing at that pivot point is equipped instead of alone.

2. Design the complete handoff, not just the document

Add the site walk. Define a session where the team discusses the discrepancies and learns together.

This is not extra process. This is designing a handoff, and designing a seam that won't break in execution.

Takeaway: the fix is a designed conversation plus a format that admits what everyone already knows. Neither one is a binder.

FAQ

Is padding an estimate the same as inflating a price?

No. Padding an invoice is a billing problem. A cushion in an estimate is a risk allowance, built deliberately by somebody who has watched jobs change and does not want to be caught short. One is dishonest, the other is professional judgment. Treating the second as the first is how founders damage trust with their best people.

Should I just tell my team to stop padding?

No, and it would make things worse. The cushion is the correct call given what your team knows and what they're handed. Remove the cushion without changing what they're handed and you have removed their protection while leaving the uncertainty in place. Change the handoff first, and the cushion shrinks on its own.

What's a seam?

Any point where important information changes hands. Between functions, and inside them. Operations is usually the densest seam territory in a company. A seam isn't a department boundary, it's a point of transfer, and most companies have never mapped theirs.

How is this different from just writing better SOPs?

An SOP is a document that sits beside the work. This is about the shape information travels in when it moves between two people who both have to make decisions with it. A perfect document that can't carry uncertainty still fails, and it fails silently, because the receiver completes it.

Doesn't a range look unprofessional to a client?

That instinct is exactly what the episode is about. A range that states what's fixed, what can flex, and how far reads as command of the work. A single number that changes on the job reads as a number that was never real. Certainty on the page is not the same as certainty in the outcome.

Where do I start?

Take one handoff you have been thinking about while reading this. Find the cushion attached to it, then go ask the person who built it what it protects against. Wait through the pause. The second answer is usually the real one.