The first version of this article said construction technology spending had reached "$18 billion" a year according to "McKinsey & Company's latest construction technology survey," that it had grown 24% a year, and that it equaled 1.5% of industry revenue. It then broke the $18 billion into eleven categories with spend, growth and adoption rates, and gave an "average ROI" for each tool type, down to "3.2x" for project management software and "0.6x" for robotics.
We could not find any of that in a McKinsey publication or anywhere else. The category table, the ROI multiples, the "42% of investments fail" figure, the comparisons with manufacturing and financial services, and the FAQ answers built on them have all been removed. There is no reliable public measure of total U.S. construction technology spending, and an invented one is worse than none.
What remains is a smaller set of numbers that can be checked, and a practical way to decide what to buy.
What the Surveys and Filings Show
AI adoption: the AGC/Sage outlook
The most current contractor-level data on technology come from the 2026 Construction Hiring and Business Outlook by the Associated General Contractors of America and Sage. AGC surveyed members from November 4 to December 15, 2025, and got 951 responses from 49 states and the District of Columbia. About half of the responding firms did $50 million or less of work in 2025, and 59% had fewer than 100 employees.
On artificial intelligence, the survey found:
- 61% of firms said they use AI or plan to increase their investment in it, up from 44% in the previous year's survey.
- 45% use AI for office and administrative functions.
- 23% use it for estimating.
- 20% use it for design or preconstruction.
- 16% use it for recruitment, training or other HR functions.
The pattern is worth noticing. The most common uses are back-office ones, where the tools are general-purpose and the risk of a bad output is low. Field and production uses are not in the top of the list.
A public construction software company
Procore is a publicly traded company focused on construction management software, so its quarterly results give a dated view of what contractors and owners are paying for. In its second quarter 2026 results, released July 29, 2026, Procore reported:
- Quarterly revenue of $375 million, up 16% from a year earlier.
- 2,871 customers contributing more than $100,000 of annual recurring revenue as of June 30, 2026, up 14%.
- Full-year 2026 revenue guidance of $1,510 million to $1,514 million.
That is one company, not the market. But mid-teens growth at a company of that size is a reasonable sign that spending on project management software is still growing, without anyone having to guess at a total.
What's not measured
No federal statistical agency publishes construction industry spending on software or digital tools as a separate number that we could find. Private "market size" estimates vary widely depending on whether they count global or U.S. spending, include hardware, include design software sold to architects and engineers, or count only construction-specific products. If you see a single headline figure, check what it includes before you use it.
Where the Money Typically Goes
Without sourced spend or ROI figures, here is how the main categories compare in practical terms. These are descriptions, not measurements.
Project management and document control. Drawings, RFIs, submittals, daily logs and change events in one shared system. This is the category Procore's revenue reflects. The value comes from fewer lost documents and faster turnaround on RFIs and submittals, which you can measure on your own jobs.
Estimating and takeoff. On-screen takeoff from PDFs or models. Prices are published for some tools: Bluebeam lists its plans at $260, $330, $440 and $590 per user per year, billed annually, on its pricing page (checked October 18, 2026; the top tier is marked as an introductory price). AGC's survey suggests estimating is also the second most common place contractors are trying AI.
Accounting and job costing. Construction-specific accounting with job cost, WIP schedules, retainage, AIA-style progress billing and certified payroll. Generic small-business accounting software handles some of this with workarounds. Our article on construction profit margins covers why job-level cost tracking matters.
Time tracking and field reporting. Mobile time entry, crew logs and photo documentation. Useful on prevailing wage work, where certified payroll depends on accurate hours by classification.
BIM and coordination. 3D models for clash detection and MEP coordination. Often required by the owner or GC on larger projects, in which case the decision is less about ROI and more about whether you can do the work at all.
Drones and reality capture. Aerial survey, progress photos and stockpile volumes. Anyone flying a drone for commercial work needs an FAA remote pilot certificate with a small UAS rating, or must be directly supervised by someone who has one, under 14 CFR 107.12. Part 107 also caps altitude at 400 feet above ground level, except within 400 feet of a structure, and requires 3 statute miles of visibility (14 CFR 107.51). Many contractors hire out drone work rather than staff it.
Robotics, AR/VR and wearables. Layout robots, rebar-tying machines, headset-based model overlays and sensor wearables. Adoption is still limited and results depend heavily on the task. The original article's "5-8 years away" prediction was opinion, not data.
How to Decide What to Buy
The useful parts of the original article were about buying discipline, and that holds up without the invented numbers.
Start with a cost you can already see
Pick a problem that is costing money on current jobs: RFIs that sit for weeks, takeoffs that take too long to bid everything you want, rework from missed coordination, payroll errors on prevailing wage jobs. Write down what it costs now, in hours or dollars, from your own records. If you cannot put a number on the problem, you will not be able to tell whether the tool fixed it.
Price the whole implementation
The license is the visible cost. Training, setup, data migration, integration with accounting, and the slower weeks while crews and PMs learn the system are real costs too. Ask the vendor for references from firms your size and ask those firms what the rollout actually took.
Pilot on one job or one team
Run the tool on one project or with one estimator first. Measure the same things you wrote down at the start. If the pilot does not show a difference, do not roll it out company-wide on the hope that scale will fix it.
Give it an owner
Every tool needs someone inside the company whose job includes making it work, keeping templates current and training new hires. Without that person, software tends to get paid for and not used.
Watch for overlap
Most contractors end up with several systems that do not share data, and the same information gets typed into estimating, project management and accounting. Before adding a new tool, check whether it connects to what you already use, and whether an existing vendor already offers the feature.
Be careful with AI outputs
The AGC survey shows contractors moving quickly on AI. For estimating, scheduling and contract review, treat AI output as a draft that a qualified person checks. A wrong quantity or a missed contract clause costs the same whether a person or a model made the mistake.
A Starting Stack for a Small Contractor
Without inventing prices, here is a reasonable order of adoption for a firm under about 50 people:
- Construction accounting with job costing, because every other decision depends on knowing which jobs make money.
- Digital takeoff, because it directly affects how many bids you can turn and how accurate they are.
- Mobile time tracking tied to payroll, especially if you do public or prevailing wage work.
- A shared project management and document system once you run several jobs at once or work for GCs who require one.
- Specialized tools such as BIM, drones or reality capture when a project or owner calls for them, often hired out first.
Get current quotes directly from vendors. Many construction software companies do not publish list prices, and the ones that do change them.
Frequently Asked Questions
How much does the construction industry spend on technology?
There is no reliable public total. The "$18 billion" figure in the original version of this article, attributed to McKinsey, could not be found in any McKinsey publication and has been removed. Private market estimates vary widely depending on scope. For a dated, checkable signal, Procore reported Q2 2026 revenue of $375 million, up 16% year over year.
How many contractors use AI?
In the AGC/Sage 2026 Construction Hiring and Business Outlook (951 respondents, surveyed November to December 2025), 61% of firms said they use AI or plan to increase their AI investment, up from 44% a year earlier. Forty-five percent use it for office and administrative work, 23% for estimating and 20% for design or preconstruction.
What is the return on construction software?
We do not know of a credible industry-wide ROI figure for any category, and we removed the multiples this article used to carry. Measure it on your own jobs: record the cost of the problem before you buy, run a pilot, and compare.
Do I need a license to fly a drone on my jobsite?
For commercial use, the person flying needs an FAA remote pilot certificate with a small UAS rating, or must fly under the direct supervision of a certificated remote pilot who can take control, under 14 CFR 107.12. Part 107 also limits altitude to 400 feet above ground level except near structures.
What does estimating software cost?
It varies by product. As one published example, Bluebeam lists plans at $260 to $590 per user per year, billed annually, as of October 2026. Many other vendors quote prices only on request.



