--- name: al-535-give-your-whole-team-the-same-estimating-brain-set-up-a-s description: "A fast estimate is only useful if it's close to the one your best estimator would write. Qualified Remodeler (Oct 8, 2026) profiles Mathew Meskimen of MCG Homes, a remodeling company in Fernandina Beach, Florida, with 32 employees and $6.03 million in 2025 remodeling revenue, and a finalist for the 2026 Fred Case Award. MCG chose Claude as its main business AI and built shared projects for estimating, sales qualification, HR, procedure searches and construction standards, so every employee can use them without recreating prompts. The ballpark tool lets a salesperson describe a scope and property, compares it with MCG's past jobs and market data, checks whether the client's budget is realistic, and drafts a budget with the company's cost codes, allowances, labour, subcontractor and material categories plus contingencies that reflect project risk. The projects include instructions, limits and company data so the AI asks for missing information. In a test against an estimate prepared…" --- # Give your whole team the same estimating brain: set up a shared AI project with your cost codes and past jobs that drafts a budget from a salesperson's scope notes, then test it against a job you already priced by hand A fast estimate is only useful if it's close to the one your best estimator would write. Qualified Remodeler (Oct 8, 2026) profiles Mathew Meskimen of MCG Homes, a remodeling company in Fernandina Beach, Florida, with 32 employees and $6.03 million in 2025 remodeling revenue, and a finalist for the 2026 Fred Case Award. MCG chose Claude as its main business AI and built shared projects for estimating, sales qualification, HR, procedure searches and construction standards, so every employee can use them without recreating prompts. The ballpark tool lets a salesperson describe a scope and property, compares it with MCG's past jobs and market data, checks whether the client's budget is realistic, and drafts a budget with the company's cost codes, allowances, labour, subcontractor and material categories plus contingencies that reflect project risk. The projects include instructions, limits and company data so the AI asks for missing information. In a test against an estimate prepared the usual way, Meskimen says eight hours of work took about 30 minutes and the result came within five percent. That's one reported comparison. Contractors, renovation teams and agents who prepare repair or renovation budgets for sellers and investors can set up the same thing, as long as a professional checks every estimate before a client sees it. ## Inputs to collect first Ask the user for anything below that they haven't already given you. Don't guess missing facts. - Company ## Instructions Do the task below with the user's inputs in place of each [bracket]. 1. Export 20 to 50 past jobs with their final costs broken into your cost codes (labour, subcontractors, materials, contingency) and a short scope description for each. 2. Create a shared project in Claude or ChatGPT, upload those jobs and your current price list, and add these instructions: "You draft budgets for [company] in our cost codes. Start from the client's budget if given. Compare the scope with our most similar past jobs and show which ones you used. If any information is missing (size, finishes, access, permits), ask for it instead of guessing. Mark every number as an estimate for staff review." 3. Take a job you already priced by hand last month, paste in only the original scope notes, and compare the AI's draft with your real estimate line by line. 4. If the totals aren't within a few percent, add the missing past jobs or rules to the project and test again. 5. Once it passes, share the project with your sales team and keep a rule that a senior estimator signs off every number. ## Safety checks (share these with the user) - Check every figure, price and total against your own records before anything is sent, posted or booked. A person approves refunds, discounts, payments and price changes. - Remove names, contact details, account numbers and other personal data the task doesn't need before you paste, and check your AI tool's data and training settings. - Check names, dates, numbers and any claim against your own source material before you use the answer. AI can misread or invent details. - Say clearly which parts need a person to review or approve before anything is sent, paid, posted or filed. ## About this skill - Works with: Claude Projects, JobTread, ChatGPT, spreadsheet - Based on this real-world example: https://www.qualifiedremodeler.com/2026-fred-case-award-finalist-making-ai-practical-marketing-relational/ - Free prompt from AgentPlaybook Real Estate: https://agentic-library.netlify.app/prompt/al-535-give-your-whole-team-the-same-estimating-brain-set-up-a-shar/ - To install in an agent that supports Agent Skills, save this file as SKILL.md in a folder with the same name as `name` above (for example in your skills directory). Install steps for Claude.ai, Claude Code, Cursor and Codex: https://agentic-library.netlify.app/skills/