Insights
Construction tech stack poll results
How were these polls run?
These are LinkedIn polls posted to Tyler Campbell’s audience, and anyone who saw a poll could choose to vote. That makes them a self-selected sample of people who follow a construction AI founder on LinkedIn. They are a temperature check from people in and around construction, not a random survey of the industry.
Here’s the method, stated plainly so you can judge the numbers yourself:
| WHO | Tyler Campbell's LinkedIn audience. Self-selected. Roles and companies are not verified |
|---|---|
| WHEN | Q1 2026 |
| HOW | Native LinkedIn polls, one question each, with fixed answer options |
| SAMPLE | Shown separately for every poll. Each poll stands on its own |
| ROUNDING | Percentages as LinkedIn reported them, so a poll may total 99% or 101% |
Two rules we hold to. We never add the sample sizes together, because the same people very likely voted in more than one poll. And we never present these as industry research. A small sample with an honest method is still worth reading. A small sample dressed up as a study isn’t.
Poll 1: How optimized is your tech stack?
In a poll of Tyler Campbell’s LinkedIn audience (n=46), 66% described their company’s tech stack as “patchwork” or “chaos.” Only 7% said fully optimized.
| Answer | Share |
|---|---|
| Fully optimized | 7% |
| Somewhat optimized | 28% |
| Patchwork or chaos, combined | 66% |
| Chaos alone | 20% |
Our read: patchwork is what happens when a company grows one tool at a time. Each purchase made sense when it was made. Nobody planned how they’d fit together, so the same number gets typed into three systems and somebody keeps a spreadsheet to reconcile them. AI on top of that inherits every gap.
Poll 2: Who owns technology decisions?
In a poll of Tyler Campbell’s LinkedIn audience (n=72), 24% said a dedicated tech leader owns technology decisions at their company. 47% said “the leadership team,” and 11% said nobody does.
| Answer | Share |
|---|---|
| A dedicated tech leader | 24% |
| "The leadership team" | 47% |
| Project teams | 18% |
| Nobody. It's reactive | 11% |
Our read: shared ownership is the most common answer, and it’s the one that worries us most. A decision owned by a group tends to wait for a meeting. That’s the Ownership Gap Model, and it’s one of the main reasons AI work stalls at a couple of people. More on that in why construction AI pilots don’t scale.
Poll 3: What’s the biggest barrier to fixing it?
In a poll of Tyler Campbell’s LinkedIn audience (n=65), 37% said the biggest barrier to optimizing their tech stack is no time to fix it. Another 31% said it’s hard to prove the ROI. Together, 68% picked one of those two answers.
| Answer | Share |
|---|---|
| No time to fix it | 37% |
| Hard to prove ROI | 31% |
| Too many tools | 20% |
| No clear owner | 12% |
Our read: “no time” and “can’t prove it” are the same problem from two sides. Without a number, fixing the stack never outranks the work in front of you. With a number, it gets a slot on the calendar. Only 20% blamed the tools themselves.
Poll 4: Would leadership pay to fix it?
In a poll of Tyler Campbell’s LinkedIn audience (n=44), 14% said yes, leadership would pay someone to streamline their tech, and 61% said “maybe, if ROI is clear.” That’s 75% at yes or maybe.
| Answer | Share |
|---|---|
| Yes | 14% |
| Maybe, if ROI is clear | 61% |
| Doubtful | 11% |
| No | 14% |
Our read: the appetite is there, and it comes with a condition. The 61% aren’t saying no. They’re saying show me the math. That’s a fair ask, and it’s why every engagement we run starts with a Workflow Sprint that shows what the fix is worth in hours before the full build. How to measure AI ROI in construction walks through the formula.
Poll 5: What matters most in how work gets done?
Asked what sits above everything else in how work gets done, 50% of respondents in a poll of Tyler Campbell’s LinkedIn audience (n=86) picked the people. 23% picked systems and tools.
| Answer | Share |
|---|---|
| The people | 50% |
| Systems and tools | 23% |
| The work to be done | 14% |
| It depends | 13% |
Our read: the same audience that calls its tech a patchwork puts people first by a wide margin. We agree with them. Tools don’t adopt themselves. That’s why training the team is part of how we help, not an add-on.
Latest poll
The most recent polls on this page closed in Q1 2026.
When new results come in, they’re added here with their own sample size and date. Older polls stay up, unchanged.
Changelog
- October 2026: page published with five polls from Q1 2026.
How to cite these numbers
Cite each poll on its own, with its sample size and quarter. For example: “Poll of Tyler Campbell’s LinkedIn audience, n=46, Q1 2026.” Please link back to this page so readers can see the method. Please don’t combine the sample sizes, and please don’t describe the results as an industry survey. They’re an honest read of one audience, and that’s how they should travel.
Sources
- Original polls posted on Tyler Campbell’s LinkedIn, Q1 2026.