The Slowest Kid on the Hike Runs Your Cash Flow
What a 1984 novel about a factory explains about your closeout, your bonding capacity, and why hiring another project manager will not fix it
There is a conference room in Mesa where I have watched this happen more than once.
Wednesday morning OAC. Coffee that has been sitting on the burner since seven. The owner’s rep at the head of the table with a laptop. The architect on the screen. The superintendent came straight from the field with dust still on his boots, already thinking about the job he starts Monday in Casa Grande.
They go around the table. Schedule. Weather. Two RFIs still open. Then somebody says the number.
Ninety-eight percent.
The room nods. It’s a good number. It was also the number three weeks ago, and the number two weeks before that, and it is going to be the number again next Wednesday. Everybody in the room knows it. Nobody says it out loud, because saying it out loud would require somebody to explain why, and the explanation is boring and involves four PDFs and a vendor list.
The superintendent’s crews are gone. The project manager has three new jobs. The estimator is heads-down on a bid due Friday. Every person in that room is working hard, and that number has not moved in twenty-one days.
That is the part that should bother you. Not the delay. The effort. All of that effort, and the job sits at ninety-eight.
If you have been in construction in Arizona for more than about fifteen minutes, you have sat in that room.
I want to tell you why the number does not move, and I am going to get there by way of a novel about a factory.
The Slowest Kid on the Hike
In 1984, an Israeli physicist named Eliyahu Goldratt published a business book disguised as a novel. It is called The Goal. It follows a plant manager named Alex Rogo who has ninety days to fix his factory before corporate shuts it down.
The scene everybody remembers has nothing to do with the factory.
Rogo takes his son’s scout troop on a hike and cannot figure out why the line keeps stretching out. The fast kids pull ahead. The gaps open. He calls for everybody to close up. They close up, and a few minutes later the gaps are right back.
The reason is a boy named Herbie. Herbie is the slowest kid in the troop, and he is carrying an overloaded pack.
Rogo tries the obvious fix first. He pushes the front of the line harder. All that does is stretch the line further, because the boys ahead of Herbie can walk as fast as they want and the boys behind him cannot pass. The troop has not gained a single step. It is just more spread out, with the same kid setting the pace from the back.
So he moves Herbie to the front. Now nobody can get ahead of him, the line cannot stretch, and the troop’s speed is Herbie’s speed. Visible. Measurable. The only number that matters.
Then they open Herbie’s pack and start handing the weight to the fast kids. Herbie speeds up. The whole troop speeds up with him.
Goldratt called this the Theory of Constraints, and the useful part fits in one sentence. Every system moves at the speed of its single slowest step, and effort spent anywhere else is not improvement. It only looks like improvement.
Your Closeout Is a Hike
Here is the reframe, and it is the whole piece.
“Your firm does not move at the speed of your best people. It moves at the speed of your constraint. And on a lot of Arizona jobs, that constraint is the punch list.”
Last week I walked through a real closeout with a well-run Southern Arizona general contractor, using real documents instead of a sanitized version. Forty-seven items. Four different design firms. Not one item on any of the four lists said which subcontractor was responsible.
Two subs on that job both listed framing and drywall as their trades. Two more both listed plumbing and HVAC. The vendor directory read like a puzzle, with companies appearing as three-letter codes that told you nothing about what they build.
And exactly one person in the entire system could look at any given item and know, from memory and from having been there, whose problem it actually was.
That is Herbie. Not the person. The step.
Everything downstream of that step is stacked up behind it. Final inspection. Owner sign-off. Final pay application. Retainage release. Your bank balance. Your bonding capacity. The next job you are trying to convince a surety you can carry.
One person’s afternoon is standing in front of all of it.
Everybody Is Yelling at the Front of the Line
Here is what I have watched over twenty years of selling technology to Arizona builders. When a firm decides to get serious about improvement, they almost never start at the constraint. They start where the demo looked best.
Estimating software, so takeoffs go faster. Drones and progress photography, so the owner can see the steel go up. A dashboard so the OAC has better slides. A second monitor for the PM. A new project management platform because the last one did not stick.
Every one of those can be a good purchase. None of them is Herbie.
They are the fast kids up front. Speed them up and the line stretches. You win the bid faster, mobilize faster, frame faster, and then arrive at the same closeout with the same forty-seven items and the same one person who knows which sub owns the gate.
You did not make the hike faster. You made the gaps bigger.
And the reason this is so easy to miss is that nobody in the story is doing anything wrong. The PM is not lazy. The superintendent is not disengaged. The software is not bad software. Everybody is walking as fast as they can walk.
The tech is not the bottleneck. It hasn’t been for a while.
What the Hike Actually Costs
Now for the part that turns a metaphor into a number.
Goldratt’s rule is that an hour lost at the constraint is an hour lost for the entire system, and in construction that translates directly into cash that is earned, owed, and unavailable.
- Contractors carry receivables an average of roughly 57 days, per the Construction Financial Management Association’s Financial Benchmarker. That figure has barely moved in a decade. It was 58.7 days the year before.
- Slow payment cost the U.S. construction industry an estimated $280 billion in 2024, roughly 14 percent added to total construction spending, according to Rabbet’s construction payments research. That is not a rounding error. That is the industry lending money to itself at scale.
- Retainage on most contracts runs 5 to 10 percent of contract value, and on the average commercial job that number is close to the entire profit you booked when you signed.
- Sureties size bonding programs at roughly 10 times adjusted working capital, with aggregate programs commonly running 10 to 20 times, per CFMA and surety industry underwriting convention.
- And here is the one most builders have never had explained to them. Underwriters strip receivables aged past 90 days out of that working capital calculation before they apply the multiplier. Retainage broken out separately on your aging report is often carved out as an exception. A final billing hung up behind an unresolved punch item usually is not. On paper, that is just old money.
Read those last two together, because that is where the metaphor stops being cute.
A punch list that drags a final billing past ninety days does not just cost you the cash. It shrinks the number your surety multiplies. And the multiplier is ten.
THE HERBIE MATH
Pull your last three closed jobs and fill in three blanks.
- Days from substantial completion to final payment: ________
- Dollars still outstanding past 90 days on those jobs: $________
- Multiply that second number by ten.
That third number is roughly the bonding capacity sitting inside your punch list.
Every $50,000 stuck past ninety days is somewhere around $500,000 of work you cannot bond. Not work you lost. Work you were never able to go get.
I have never once seen that math on a whiteboard in a job trailer. I have seen a lot of whiteboards with twenty-seven other problems on them.
You Do Not Fix Herbie by Hiring Another Herbie
The instinct, once a firm finally identifies the constraint, is to throw a person at it. Hire another project manager. Assign a closeout coordinator. Put somebody on it full time. Sometimes that is the right call. Usually it is expensive and slow, and it misses what actually happened on the hike.
They did not swap Herbie out for a faster kid. They took weight out of his pack.
So what is in the pack on a punch list?
It is not the walking. Nobody minds walking the job. The weight is the sorting. Reading forty-seven ambiguous items, cross-referencing a vendor directory full of three-letter codes, deciding which of your two drywall subs owns each item, catching the ones where a sequence is buried inside the sentence, and then writing eleven separate emails to eleven separate subs.
That is hours of pure translation work, and it is sitting on the busiest person in your company at the worst moment of the project.
Which is the same argument I have been making all year. Your data is the lever, not the tool. The vendor directory, the scope assignments, the completed list from the last job, the knowledge in one person’s head about which company builds gates. That is the pack weight. It is also, conveniently, exactly the kind of thing a machine is good at carrying.
What We Are Doing About It on September 10
On Thursday, September 10, at 11 AM Arizona, we are running Session 3 of our AI in Construction series, sponsored through the Arizona Builders Alliance. The session is called The Punch List Agent, and it is the finale of Series One.
Nate Kappler of Kappcon Construction is going to be in the room with his real documents. The team building it will put together a Claude-based agent live, on screen, that reads an actual punch list, takes the project vendor directory as context, assigns each item to the responsible trade, flags the items where two subs could both legitimately own it, and drafts the outreach.
Then we check its work against the completed list showing who actually did each item.
Including the items it gets wrong.
That kind of yes takes real courage from a builder, and I do not take it for granted.
I want to be honest about scope, because I am not selling anybody a silver bullet on a live call. The agent assigns and drafts. It does not push into your project management platform, it does not chase subs, and it does not read the red circles a landscape architect drew on a plan snippet. Those are real problems, they are solvable, and they are not what we are building on the tenth.
We are taking weight out of one pack. That is the entire ambition, and it is enough.
THE PUNCH LIST AGENT- SESSION 3 OF THE AI IN CONSTRUCTION SERIES
When: Thursday, September 10, 2026 at 11 AM Arizona
Featuring: Nate Kappler, President, Kappcon Construction
Live Build: A Claude-based Punch List Agent on real Kappcon documents
Hosted by: Dimensions and Smartbuild, sponsored through the Arizona Builders Alliance
Cost: Free. 60 minutes. Recording available to all registrants.
Register Here: AI In Construction – The Punch List Agent
The Move
This one is small on purpose, and you can do it before lunch tomorrow.
Open the vendor list for the job you are closest to finishing. Look at it the way a stranger would look at it, which is to say the way anybody other than the one person who has been on that site would look at it.
Now add one column. Call it Scope In Plain English. Fill it in. Gates and fencing. Aluminum storefront. Framing, not drywall. Drywall, not framing.
That’s it. That is the whole move.
It takes about twenty minutes, it costs nothing, and it does three things. It gets the assignment knowledge out of one head and onto a document, which means the constraint is no longer a single person’s availability. It makes the ambiguities visible before the punch list lands instead of during. And it is the exact context file any agent you eventually build will need in order to be useful.
You do not have to buy anything to start moving weight out of the pack. You just have to write down what is in it.
That conference room in Mesa is going to fill up again on Wednesday. Same coffee. Same laptop. Same superintendent with dust on his boots, thinking about Casa Grande.Somebody is going to say ninety-eight percent, and the room is going to nod.
The number is not stuck because anybody in that room is slow. It is stuck because one step in the system is carrying a pack nobody has ever bothered to open, and every other improvement anybody makes just stretches the line out behind it.
Find your Herbie. Then take some weight out of his pack.
The hike does not go faster any other way.
For over 20 years, Dimensions has been the trusted IT partner for Arizona’s architecture, engineering, and construction industry. We help AEC firms communicate better, collaborate smarter, and actually use the technology they’ve invested in. Because in construction, the tools only work if your team does.
IT Built For Builders.
