Back to Blog.

The High Cost of Low Trust Starts on the Sheet

Every builder knows the moment. You open a set, flip to the door schedule, and the tag on the plan is not there. Or the callout points to a detail that never made it into the package. It is a small thing. But something shifts when you see it. You stop reading the set the way you read it five minutes ago. Now you are checking. Now you are double checking. Now you are wondering what else is wrong.

That shift has a price. And the industry has started to measure it.

Trust is not soft. It shows up in the numbers.

In 2019, Autodesk partnered with FMI to survey more than 2,500 construction professionals around the world for a report called Trust Matters: The High Cost of Low Trust [1][2]. They asked people to rate trust inside their own organization and between collaborators on projects.

Nearly everyone (93%) reported at least a basic level of trust within their organization. But only 37% said their organization had very high trust [2]. Put the other way, 63% of respondents said their organizations fell short of very high trust [3].

The gap between "fine" and "very high" turned out to matter:

  • Very high trust firms were half as likely to run into the problems that show up when data is not shared fast enough. 18% said it rarely or never happens [2].
  • They were twice as confident about meeting project schedules: 43% compared to 21% at "above average" trust firms [4].
  • They were more than twice as likely to say collaboration is central to how they work: 43% versus 19% [3].
  • 57% reported doing more than 80% of their business with repeat clients, compared to 42% of "above average" firms [3].

The report also frames the backdrop plainly. In the US alone, missed schedules cost the industry about $165 billion, roughly 17% of total multifamily, nonresidential building, and non-building construction spending in 2019 [2]. FMI and Autodesk noted that the role trust plays in that waste is measured far less often [2].

When trust is low, people verify everything. They send the extra RFI. They hold the extra meeting. They build in a little contingency, just in case. They keep their own notes so that, if it turns into a dispute later, they have a record. None of that is irrational. It is what careful people do when they are not sure the information in front of them holds together. It is also slow, and it adds up.

A lot of distrust starts with a set that disagrees with itself

Here is something I have believed for a long time. A big share of distrust on a project does not start with people. It starts with paper.

When the sheet index lists a sheet that is not in the set, or a keynote on the plan does not match the legend, or the panel on the floor plan does not line up with the schedule, the drawing set is contradicting itself. Nobody did anything malicious. Several firms, working in different software, on a compressed schedule, handed off a set before every discipline finished checking against the others. That is just how drawings get made.

But the person on the receiving end does not experience it as a process problem. They experience it as "I can't rely on this." And once one sheet lets you down, every sheet after it gets read with suspicion. The architect gets the RFI. The sub prices the uncertainty. The GC carries the risk. Trust erodes one small mismatch at a time.

Why MarkedUp exists: Harry

The idea for MarkedUp did not start in a boardroom. It started with a single, painful error.

Christmas of 2023, we were back in Australia visiting family. My brother-in-law, Harry, was interning at a major tier one builder. One year into his university degree, he had been handed the doors and windows scope.

He did what he had been taught. He ordered the windows off the schedule, using the same tools I used back in 2008: Excel and Bluebeam. Nobody checked his work. The schedule and the plans did not agree, and the door tags were missing from the plan area. Nobody caught it until it was a $150,000 mistake.

The company waved it off. Harry did not.

He had been working 15 hour days, going to university, and doing projects on weekends. Honor roll student. Showed up. Worked hard. Really cared. And the industry did not set him up to succeed. That mistake is the reason he left construction for good, and he has moved on to a different career.

That is a massive loss. I genuinely believe Harry was on a path to run a major builder one day, because he cared enough to want to do things better.

What hit me hardest was not the mistake itself. Anyone who has reviewed drawings by hand has missed something. It was that fifteen years of software had not changed the basic reality. Teams were still relying on highlighters, cross-references, and human fatigue to protect their margins. If Harry had had something to tell him the door tags were missing from the plan, that day would have gone very differently.

That is the scar. I do not want another young engineer to walk away from this industry because a set did not agree with itself and nobody had the hours to check every reference.

A quiet first set of eyes

So we built MarkedUp to do the grueling consistency checks that people should not have to perform sheet by sheet.

MarkedUp is a quiet first set of eyes. Before a set is bid, built, or shared, it checks whether the drawings agree with themselves: sheet index to sheet number, callout to detail, door to schedule to plan, keynote to legend, and MEP equipment and panels to plan. It stays silent when the set holds and speaks up when something does not line up.

It is not another place to go. Findings come back as a marked-up PDF that drops into the tools teams already use, like Bluebeam, Procore, or Adobe. Sets can come in from Trimble Connect, Trimble ProjectSight, Autodesk Forma, Google Drive, or SharePoint.

The goal is not to replace the engineer or the reviewer. It is to give them a tireless check on the boring, cross-sheet relationships, so their judgment goes to the decisions that actually need a person.

The hope: a set you can trust

Here is the hopeful part. Trust is not fixed. The FMI and Autodesk research suggests that businesses and project teams can measure trust and actively improve it, and that firms can take proactive steps to build stronger relationships and more certain outcomes [2].

I think one of the most practical steps is also one of the least glamorous: send a set that agrees with itself. When the architect knows the references hold before the set goes out, they issue with confidence. When the GC receives a set that has already been checked, they can spend their energy on scope and sequence instead of hunting for missing sheets. When the sub opens the package and the door tags match the schedule, they price the work, not the doubt.

Every one of those moments is small. Together, they are what trust is made of.

That is what Harry's story taught me, and it is why we come to work. If any of this sounds familiar, I would genuinely like to hear your version of it. Most of the time it starts with a conversation, then we run a check on one of your real projects so you can see what comes back. You can find us at markedup.io.

Christopher Mouflard

Founder, MarkedUp

Sources

1. Autodesk Construction Cloud, "FMI: The High Cost of Low Trust." https://construction.autodesk.com/resources/guides/fmi-the-high-cost-of-low-trust/

2. FMI, "Trust Matters: The High Cost of Low Trust." https://fmicorp.com/reports/trust-matters-the-high-cost-of-low-trust

3. Autodesk / PR Newswire, "New Study from FMI and Autodesk Finds Construction Organizations with the Highest Levels of Trust Perform Twice as Well on Crucial Business Metrics" (March 10, 2020). https://www.prnewswire.com/news-releases/new-study-from-fmi-and-autodesk-finds-construction-organizations-with-the-highest-levels-of-trust-perform-twice-as-well-on-crucial-business-metrics-301019637.html

4. FMI and Autodesk, Trust Matters: The High Cost of Low Trust (full report PDF). https://fmicorp.com/uploads/media/Trust_Matters-The_High_Cost_of_Poor_Trust.pdf