What I Actually Look At in a Construction Operations Audit

It is not a sales call and it is not a software pitch. It is half an hour of watching how work really moves through your operation, and an honest read of where the time goes.

Brad Berlin  |  May 2026  |  Operations

← Back to all posts

When I tell a contractor I do a free operations audit, there is usually a small wariness in the room. Is this a sales call. Am I about to get pitched software. It is a fair thing to wonder, so it is worth saying plainly what an operations audit is, and what it is not.

An audit is the looking that has to happen before anyone recommends anything. You cannot fix a workflow you have not watched, and most construction operations have never been watched end to end by someone whose only job, for that half hour, is to see where the time goes. This is what that looks like.

Most operations problems hide in the workflow, not the software

Owners usually arrive with a guess already formed. The software needs replacing, or the shop is a person short. Sometimes that guess is right, and a good audit will say so. But more often the friction is not inside any one tool and not inside any one person. It lives in the seams between them, the hand-offs and the waiting and the re-entry, and seams do not show up on a software invoice or a performance review.

That is why an audit does not start with your software list or your org chart. It starts by watching how work actually moves. I have written about where estimating hours go and the gap between the tools that create them. The audit is simply how you find those things inside one specific shop.

It starts by following one real job, end to end

The first thing I do is not interview people. It is pick one recent bid and trace it from the invitation to the submitted number. Who touched it. What they did with it. What they were waiting on. Where it sat overnight.

One real bid, followed all the way through, teaches more than an hour of anyone describing their process, because the described process and the real one are never quite the same. The description is how the work is supposed to go. Following the job shows how it actually went, including the parts nobody thinks to mention because they have stopped noticing them.

Where the friction usually shows up

After enough of these, you learn where to look. A handful of places account for most of it.

Estimating. The point where an estimator stops estimating and starts couriering data: rebuilding the abstract sheet, chasing subcontractor quotes, re-keying numbers between tools that do not talk. The judgment work is what you pay an estimator for. The audit looks for how much of the week is not that.

Coordination. The small asks. The how-is-it-going message, the where-does-this-stand status check, the email forwarded so someone else can finally act on it. Each one is tiny. Together they are a real tax, and because each is tiny, nobody has ever added them up.

Disconnected systems. I do not count this as a number of software tools. Operationally it looks like the same figure typed into three places, a workbook moving around as emailed versions, and one person who is the only one who reliably knows which file is current. Old bid history that nobody can search is the same pattern, aged.

Approvals. Where work stops and waits on a yes. A bid waiting on a go or no-go. A change waiting on the owner. The audit looks for where jobs sit, and who the sign-off is.

Status visibility. Whether anyone can see where a job is without asking. When status lives only in someone's head, every status question interrupts two people instead of none.

What usually surprises the owner

Two things come up almost every time. The first is that the slow part is rarely the one thing they expected. They came in certain it was the takeoff, or one particular employee, or a tool that needed replacing. It is almost always spread across many small seams nobody had ever named.

The second is harder to say, so I will say it plainly. The owner's own desk is often one of the bottlenecks. When every meaningful approval routes through one person, that person becomes the constraint on everything downstream of them, and they are usually the last in the building to see it. That is not a criticism. It is just what an outside half hour of looking tends to surface.

What an audit is not

It is not a sales presentation. It is not a software pitch, and it does not end with a recommendation to go buy AI. It is not a phase-one discovery workshop with frameworks and a maturity score, and it does not need a slide deck.

It is half an hour of looking and asking, by someone whose only task in that window is to see where the time goes, followed by an honest read. That is the whole of it.

The audit is the diagnosis, not the sale. It is watching how work actually moves through your operation, by someone whose only job in that half hour is to see where the time leaks. It does not end with buy AI. It ends with where the friction is, and what is worth doing about it.

What happens after

The output is plain language. Here is where your hours are leaking. Here is what is worth fixing. Here is what is not worth touching, because not every rough edge is worth the disruption of changing it.

If there is a workflow worth implementing, that is a separate and properly scoped conversation, with its own number attached. If there is not, I say so. The audit's job is the diagnosis. It is not the sale, and treating it as one would ruin the only thing it is good for.

Why the looking has to come first

Every recommendation worth making rests on having watched the real work. A fix proposed without that is a guess wearing a confident voice, and construction has enough of those already.

The audit is just the looking. It is unglamorous, it is specific, and it is the step that everything useful depends on. If your operation has never had that half hour, it is worth having, whoever you have it with.


If you want that half hour, the operations audit is free, it takes about thirty minutes, and there is no pitch attached. I look at how work moves through your shop and tell you where the friction is. What you do with that is entirely up to you.

Book a Free Operations Audit
BB

Brad Berlin

Founder, Berlin Management Group  |  Phoenix, AZ

Brad is an operations executive who works with Phoenix construction and construction-adjacent businesses, implementing AI into the workflows that run estimating, bids, and project administration.

A note on sourcing: this article describes how I run an operations audit and the workflow patterns I have observed doing first-hand discovery work with Phoenix commercial construction and construction-adjacent businesses. It describes a process and general patterns, not a specific client engagement. No company, project, or figure in this article identifies a particular business.