Practical Task
Compare contractor estimates without getting lost in the wording
Normalize two or more estimates into the same categories so missing scope, allowances, and big price differences are easier to spot.
Checking this request…
What this helps you do
Homeowners or renters comparing quotes for the same project.
What you need
- Two or more written estimates
- Your non-negotiables for the project
Synthetic example values
Normalize the scope before you compare the totals.“Not stated” is not the same as “not included.” AI can expose the gap; the written estimate or contractor has to resolve it.
Do it step by step
- 1
Remove personal contact details you do not want to share.
- 2
Upload or paste the estimates into the same conversation.
- 3
Ask AI to build a side-by-side table using the same scope categories for every contractor.
- 4
Have it flag missing items, vague allowances, exclusions, warranty differences, and places where one quote may not be comparable to another.
- 5
Use the output to create questions for each contractor. Do not treat the AI's interpretation as a substitute for the actual contract.
Use this in your AI assistant
Starter instructions
Copy the starter below, then paste it into ChatGPT, Copilot, Gemini, Claude, or whichever capable AI assistant you use. If this Task calls for photos or files, attach them in that AI assistant too — not on this page.
Compare these estimates for the same project. Normalize them into the same scope categories and show: included work, excluded work, allowances, materials, schedule, warranty, payment terms, and total price. Flag anything that is missing, vague, or not directly comparable. Then give me a short question list for each contractor. Do not guess where the estimates are silent.
Helpful checkWhat a good result looks like
A common-language comparison and a sharper list of questions before you sign anything.
Tips and watch-outs
- Ask it to distinguish 'not mentioned' from 'not included.' Those are not the same thing.
- For expensive or legally important work, verify the final scope and terms yourself or with a qualified professional.
OptionalGo deeper when you want to
Make every important line traceable to the actual quote
A normalized table is useful only if you can get back to the source. For each important scope item, ask the assistant to show whether it is included, excluded, not mentioned, or unclear and identify the estimate language that led to that classification.
That turns the comparison into a review aid instead of a second document you have to trust blindly.
Better comparison rule
For every row, use one of four statuses: included, excluded, not mentioned, or unclear. Add a short source note from the estimate that supports the status. Never convert silence into an exclusion.
Stress-test the price difference before ranking anyone
Two totals can look far apart because the quotes are carrying different assumptions. Ask what could plausibly move the final price: allowances, demolition surprises, permit responsibility, disposal, finish materials, change-order language, payment milestones, or work that one contractor simply did not mention.
The goal is not to have AI estimate the final bill. It is to expose which unknowns are large enough to ask about before the comparison feels fair.
- Separate fixed price from allowances and optional work.
- Call out owner-supplied materials or work by others.
- Flag vague phrases such as 'as needed' when they could materially change cost.
- Keep unknowns as unknowns instead of inventing market prices.
Build the decision matrix only after contractors answer
A weighted score can be useful, but only after the factual comparison is stable. First send the question list, collect the answers, and update the normalized scope. Then add your priorities such as schedule confidence, materials, warranty, communication, payment structure, or change-order risk.
Keep the score as a way to make your own tradeoffs visible. It should not become an AI-generated declaration that one contractor is objectively best.
Part of a larger Workflow
See where this Task fits
Use this Task on its own, or return to the full Workflow for the other pieces and their order.
Useful next
Only if it fits what you’re doing.
One practical next step and one explanation that may help. Ignore either if you already have what you need.