Pricing a job without knowing what your market will bear is one of the most common ways contractors leave money on the table or lose bids they should have won. The information exists. It's scattered across review sites, permit databases, industry forums, and competitor websites, and pulling it together manually takes hours. Here's how to do it in under 30 minutes using AI.

You don't need to call competitors or guess at market rates. Between public permit data, review platforms, online forums, and competitor websites, there's enough pricing signal in your local market to build a working benchmark for almost any project type. This issue walks through a four-step AI workflow that aggregates those signals, filters out the noise, and gives you a defensible number to price against. The whole process takes under 30 minutes and can be repeated any time your market shifts.

A general contractor I knowin a mid-size market had been pricing kitchen remodels at roughly the same rate for two years. He assumed he was competitive because he was winning about half his bids, which felt normal.

A client he'd worked with before mentioned casually that another contractor had come in $40,000 higher on a similar project the previous month and still gotten the job. The homeowners chose them because the proposal felt more thorough and the contractor seemed more confident in the number.

He hadn't been losing on price. He'd been underpricing by enough that clients occasionally wondered if something was missing from his proposal.

He ran the competitor research workflow in this issue, adjusted his pricing on the next three bids, and won two of them at the higher number. The third client chose someone cheaper, which told him the third client wasn't his client.

Step 1: Pull public permit data for your project type and market.

Most municipalities publish permit records that include project type, address, and declared value. That declared value is often the contracted price or close to it, and it's publicly available. Search your county or city's permit portal for the project type you're pricing and filter for the last 12 months.

Once you have a list, paste it into Claude with this prompt:

"Here is a list of permit records for [project type] in [city] over the past 12 months. Each record includes the declared project value. Identify the range, the median, and any patterns by neighborhood or project size. Flag any outliers that seem too high or too low to be reliable data points."

That gives you a baseline anchored in real transactions, not opinion.

Step 2: Mine review platforms for pricing language.

Homeowners who leave detailed reviews often mention what they paid or how the price compared to other quotes they received. Houzz, Google, Yelp, and Angi are all worth checking for your top three competitors.

Use this Claude prompt after copying a batch of reviews:

"Here are reviews for [competitor name] in [city]. Identify any mentions of price, cost, quotes received from other contractors, or comparisons to market rate. Summarize the pricing signals and note whether reviewers describe the contractor as expensive, competitive, or a good value relative to others they considered."

Run this for two or three competitors and you'll start to see where they're positioning and how clients are responding to their pricing.

Step 3: Check competitor websites for scope and pricing signals.

Competitor websites often reveal pricing indirectly through the project types they feature, the markets they target, and the language they use around value. Some contractors publish starting prices or price ranges. Most don't, but their portfolio and positioning tell you something about where they're playing.

Use this Claude prompt after pulling content from a competitor's site:

"Here is the website content for [competitor name]. Based on the projects featured, the language used, and any pricing information mentioned, what market segment are they targeting? Are they positioning as premium, mid-market, or value-priced? What does this suggest about their likely price range for [project type]?"

Step 4: Cross-reference with industry forums and local Facebook groups.

Homeowner renovation forums and local Facebook groups are where real pricing conversations happen. Searches like "[project type] cost [city] 2025" in Google often surface threads where homeowners share actual quotes they received, which is some of the most direct competitive pricing data available.

Paste relevant thread content into Claude and use this prompt:

"Here are posts from a homeowner forum discussing [project type] costs in [city]. Identify the price ranges mentioned, the number of quotes homeowners typically received, and any patterns in what they considered expensive versus reasonable. Summarize the pricing landscape as a homeowner in this market would experience it."

Combine the output from all four steps and you have a pricing picture built from real market data rather than gut feel.

1. Permit data is the most reliable signal you have. Reviews and forums contain opinion. Permit records contain declared transaction values. They're not perfect, allowances and change orders can shift the final number, but they're the closest thing to ground truth available without calling anyone. Start there before you look anywhere else.

2. Run this every quarter, not just when you lose a bid. Pricing signals shift with material costs, labor markets, and local demand. A benchmark you built 18 months ago may be meaningfully different from what your market will bear today. Quarterly updates take less than 30 minutes once the prompts are saved and the sources are familiar.

3. Focus on the median, not the outliers. Every market has contractors who are significantly cheaper than everyone else and contractors who are significantly more expensive. Neither extreme is useful for calibrating your own pricing. The median of a reasonably large sample is where the market is actually operating, and that's the number worth understanding.

4. Pricing signals tell you what the market accepts, not what you should charge. The goal of this research isn't to match the market, it's to understand it well enough to make a deliberate choice about where to position relative to it. Some contractors should price above the median and compete on quality and process. Some should price at the median and compete on reliability. Knowing the benchmark is what makes that a strategic decision rather than a guess.

5. Save your prompts and sources in a Claude Project. Once you've run this workflow once, save the prompts and your source list in a Claude Project so the next quarterly update takes half the time. The workflow compounds in value as your dataset grows and your prompts get refined to your specific market and trade.

Beyond the Bid Circle is a peer community for builders scaling with systems and AI—the same builders who've escaped the 60-hour grind. Swap war stories, shortcuts, and real strategies with people who get it.