What Contractors Are Actually Using AI For in 2026 (And What's a Waste of Time)
Every trade publication and LinkedIn post wants to tell you AI is going to transform construction. Most of it reads like it was written by someone who has never been on a job site.
This post is different. We work with contractors in New Hampshire and across New England. We see what they're actually doing with AI tools, what's saving them real time and money, and what they tried once and gave up on.
Here's the honest breakdown.
The Big Picture First
AI in construction is not going to replace your estimators, your foremen, or your ability to look at a job and know what it's going to take. Not in 2026, probably not in 2030.
What AI is very good at is handling the administrative and communication overhead that eats up hours every week -- writing, organizing, following up, researching, drafting. That's where the ROI is for most contractors right now.
If you're a solo operator or running a small crew, you're probably doing all of that yourself in the evenings. AI makes that work go faster and come out better.
What's Actually Working
1. Quote and Estimate Letters
This is the single most common use case we see among contractors who are actively using AI.
You've done the site visit, you've run your numbers in your estimating software. Now you have to write a proposal that explains what you're doing and why it costs what it costs. That takes time. If you hate writing, it takes even more time, and the final product shows it.
AI can turn your rough notes into a professional, detailed proposal letter in two minutes.
How it works: You type or voice-memo your notes after a walkthrough -- "Deck replacement, 400 sq ft, Trex composite, custom railing, footings to 48 inches, remove existing structure, customer wants to keep existing stairs if possible" -- and ask ChatGPT or Claude to turn that into a formal proposal letter addressed to the customer. You get a clean, professional document that explains the scope, materials, and process. Edit it, add your pricing table, and send.
The result: Contractors we work with are saving 30-60 minutes per estimate on writing alone. If you're doing 10 estimates a month, that's up to 10 hours back.
One note: AI should write the letter, not set your prices. Your numbers come from your experience, your costs, and your market. AI just formats and communicates them clearly.
2. Subcontractor Coordination Messages
You're managing multiple subs on a project. You need to communicate schedule changes, material updates, site access instructions, and punch list items. Writing these messages clearly takes time, and vague messages lead to expensive misunderstandings.
AI handles this well. You tell it what changed and who needs to know, and it drafts a clear, professional message.
Example prompt: "Write a message to my electrical subcontractor letting him know the rough-in inspection is pushed from Thursday to Monday due to the framing delay. Tell him we need him on site by 7am Monday and to confirm. Tone should be straightforward, not apologetic."
Done in 10 seconds. This sounds small but if you're doing this 5 times a week, it adds up.
3. Permit Research and Code Lookup
This is one that surprises people. AI tools -- particularly Claude and the newer ChatGPT models -- have gotten very good at parsing building codes, zoning regulations, and permit requirements.
You still have to verify everything with your local building department. Don't skip that. But using AI to do a first-pass research on setbacks, height restrictions, load requirements, or energy code questions saves hours of digging through documents.
Example: A deck contractor in the Lakes Region uses Claude to research local setback and railing requirements for each new municipality he works in before the permit application. What used to be an hour of website digging and phone calls is now a 10-minute AI research session plus a quick call to confirm the key details.
Caution: AI can be wrong or out of date on specific local codes. Always confirm with the authority having jurisdiction. Use AI to get oriented, not to replace verification.
4. Estimate Follow-Up Sequences
You sent a proposal. No response. Now what?
Most contractors send one follow-up email, wait a week, and if nothing comes back they assume the customer went with someone else. Sometimes that's true. Often it's not -- they just got busy, or they're waiting on a financing decision, or they haven't had time to read the proposal carefully.
A three-touch follow-up sequence keeps you in front without being a pain:
Touch 1 (3 days after proposal): "Just checking in to see if you had a chance to look over the proposal. Happy to answer any questions or schedule a call to walk through it."
Touch 2 (7 days after proposal): "Still here if you need anything. A few customers have asked about [common concern] -- wanted to mention that we handle that by [brief explanation]." This touch adds value by addressing a potential objection proactively.
Touch 3 (14 days after proposal): "I'll follow up one more time and then give you some space. If the timing isn't right for this season, no problem -- we can hold your spot for next year. Let me know either way."
AI writes all three of these in a few minutes. You personalize the name and project details and send. Tools like GoHighLevel or HubSpot can automate the sending so you don't even have to remember.
5. Job Site Safety Checklists and SOPs
Writing documentation is the thing most contractors put off forever. Safety checklists, onboarding procedures for new crew members, standard operating procedures for common tasks -- all of it important, all of it time-consuming to write from scratch.
AI dramatically shortens this. Tell it what the document needs to cover, give it some context about your company and how you operate, and it produces a detailed draft in minutes. You review, edit for your specific situation, and save it.
A framing crew in Manchester used this to build out their full safety binder -- the thing OSHA wants to see if they show up on a job site. What would have been a full day of work took about two hours with AI assistance.
What's Hype (Or Not Worth It Yet)
AI Estimating Software
There are a handful of tools claiming to automatically generate estimates from photos or drawings. The pitch is that you take a photo of a job site and AI tells you what it'll cost.
In practice, these tools are still early and not reliable enough to trust for professional estimates. They can ballpark square footage from photos, which is useful for rough order of magnitude, but the pricing databases don't reflect your local market, your supplier relationships, or your overhead structure.
Use AI to write your proposals, not to set your prices.
AI-Generated Design Visualizations for Clients
A few platforms let you take a photo of a house and generate AI visualizations of what a renovation might look like -- new siding colors, a deck added, windows replaced. Interesting concept.
But the output is still noticeably artificial. Most clients can tell. And if the visualization doesn't match what you actually build, you've created an expectation problem.
This one might be worth watching in 2027. Skip it for now.
Autonomous Job Scheduling and Dispatch
There are tools that promise to automatically schedule jobs, optimize routing, and dispatch crew members. The technology exists but implementation for a small contractor is currently complex and buggy.
The ROI is real for medium-sized companies with 10+ trucks and crews. For a solo operator or small crew, you'll spend more time configuring the tool than you'll save running it.
AI-Powered Bid Analysis
Some platforms analyze historical bid data to help you price more competitively. These are interesting for larger GCs bidding commercial work. For a small residential contractor in NH, you don't have the data volume or the bid complexity to make this worthwhile.
A Realistic Weekly AI Workflow for a Small Contractor
Here's what a contractor using AI effectively might actually do in a week:
Monday: Voice memo notes from Friday's site visits into ChatGPT, get proposal drafts for two customers. Edit and send.
Tuesday: Follow-up messages for three outstanding proposals, written with AI. 15 minutes total.
Wednesday: Quick AI research on permit requirements for a job in a new town. 20 minutes, then a 5-minute call to the building department to confirm.
Thursday: Write a schedule change message to two subs using AI. Post an AI-written update to your Google Business Profile.
Friday: Respond to this week's Google reviews with AI-assisted responses. Maybe generate next week's social media post about a completed project.
Total time spent: about 2 hours. Time saved versus doing all of this manually: probably 6-8 hours. That's the trade.
Where to Start If You Haven't Yet
If you're a contractor who hasn't used AI tools at all, start here:
- Get a ChatGPT account (free tier works for this) or Claude.ai
- Use it for your next proposal letter. Dump your notes in and ask it to write a professional proposal. Edit it. See if it's better than what you'd write yourself in the same time.
- If it is, build the habit. Use it for every proposal, every follow-up message, every sub coordination note.
That's it. Start with writing. Once you see the time savings, you'll naturally start asking "what else can this do?"
The Bottom Line
AI isn't going to change what you do on the job site. It's going to change how much time you spend on the business side of the job. Proposals, follow-up, communication, documentation, research -- these are the places where an hour with AI does what used to take an afternoon.
For NH and New England contractors who are running lean, that's meaningful.
Want to See This in Person?
We run workshops specifically for contractors and trades business owners where we work through these setups live. You'll leave with working tools, not slides.
See dates and register at owneroperated.ai/workshops
Or get in touch directly: owneroperated.ai/book