Log every addendum against the estimate and confirm each one is priced before bid day with AI

Short answer

Use Claude or ChatGPT, then Copilot in Excel: The AI splits each addendum into one row per item with the sheets and spec sections it touches (NOT STATED instead of guesses), then Copilot checks the log against the estimate's cost codes before submission and lists unpriced, pending or unlogged items and gaps in the addendum sequence. Addenda tend to land close to bid day, the important changes are buried in PDFs, and the usual failure is not missing an addendum but receiving it and never getting it into the number. Keep one addenda log per bid, have Claude or ChatGPT break each addendum into one row per item with the sheets and spec sections it touches, overlay-compare every reissued sheet for unclouded changes, and before you submit have Copilot in Excel list every row that has no estimate line or initials against it. Pricing and scope decisions stay with the estimator. A person still checks: every row against the addendum and revised sheets, an overlay of every reissued sheet, and every pricing, scope and bid-form decision.

AI tie-in

Tool: Claude or ChatGPT, then Copilot in Excel

The AI splits each addendum into one row per item with the sheets and spec sections it touches (NOT STATED instead of guesses), then Copilot checks the log against the estimate's cost codes before submission and lists unpriced, pending or unlogged items and gaps in the addendum sequence.

Prompt 1

Read this addendum [upload PDF]. List every item as its own row with these columns: addendum number, item number, plain-language summary of the change (one or two sentences, in your own words), sheet numbers referenced, spec section numbers referenced, whether it adds, deletes or revises a sheet, and whether it changes the bid form, bid date, alternates or unit prices (YES, NO or NOT STATED). Copy sheet and section numbers exactly as printed. Do not estimate cost, decide scope or say whether an item affects price.

Prompt 2

Here is my addenda log [paste] and the list of cost codes and line items in my estimate [paste]. List: every log row with no estimate reference or no initials; rows with status pending or cost impact unknown; log rows that cite a cost code not in my estimate; any gap in the addendum numbers; and every row that changes the bid form, bid date, alternates or unit prices. Use only these two tables and do not judge pricing or scope.

Human check: Every row against the addendum and revised sheets, an overlay of every reissued sheet, and every pricing, scope and bid-form decision.

The problem

Log every addendum against the estimate and confirm each one is priced before bid day with AI. This is a common, time-consuming task for a estimator.

What this workflow does

Addenda tend to land close to bid day, the important changes are buried in PDFs, and the usual failure is not missing an addendum but receiving it and never getting it into the number. Keep one addenda log per bid, have Claude or ChatGPT break each addendum into one row per item with the sheets and spec sections it touches, overlay-compare every reissued sheet for unclouded changes, and before you submit have Copilot in Excel list every row that has no estimate line or initials against it. Pricing and scope decisions stay with the estimator.

How it works: AI drafts or analyzes. A person operates the tools, checks the work, and takes action.

Your checkpoint: The AI splits addenda into rows and finds unlogged items; it does not price, interpret scope or confirm bid compliance. Check every row against the addendum and the revised sheets, and treat unclouded changes as something only an overlay and a person will catch.

Before you start

Access to the named tool; confirm license, version and cost with the vendor.
A small, non-confidential sample to try the workflow on first.
A qualified person to review the output before it is used.

Tools and access

Follow the steps

Start the log the day the bid set comes out, not when the first addendum arrives. Use three tabs: an addenda log (addendum number, issue date, date received, item number, plain-language summary, sheets and spec sections touched, cost impact yes, no or unknown, trades affected, estimate line or cost code where it was priced, initials, status), a sheet register (sheet number, current revision, which addendum issued it) and a questions tab for clarifications you still need.
When an addendum arrives, upload the PDF to Claude or ChatGPT and have it list every item as its own row: what changed in plain words, the sheet and section numbers it references, and whether it changes the bid form, bid date, alternates, unit prices or the sheet list. Reference numbers are copied exactly and anything unclear is NOT STATED. It does not price anything or decide scope.
Run an overlay or compare on every reissued sheet against the previous revision (for example with the compare tools in Bluebeam Revu), because plenty of real changes are never clouded. Add each unclouded change you find as its own row.
Work the log: for each row the estimator decides the cost impact and records where it landed in the estimate, or writes no cost with initials. Send trades a short summary of the rows that affect them rather than the whole addendum.
Before submitting, paste the log and the estimate's cost-code list into Copilot in Excel and have it list rows with no estimate reference or initials, rows still pending or unknown, cost codes in the log that do not exist in the estimate, gaps in the addendum number sequence, and items that change the bid form.
Read the log top to bottom, acknowledge every addendum on the bid form the way the instructions to bidders require, and save a frozen copy of the log with the bid.

Try a small example

Replace the bracketed fields with your own project details. No tested sample run is published for this workflow yet.

Sample input

Fill in the bracketed [fields] in the text below with your project data.

Starter prompt

Read this addendum [upload PDF]. List every item as its own row with these columns: addendum number, item number, plain-language summary of the change (one or two sentences, in your own words), sheet numbers referenced, spec section numbers referenced, whether it adds, deletes or revises a sheet, and whether it changes the bid form, bid date, alternates or unit prices (YES, NO or NOT STATED). Copy sheet and section numbers exactly as printed. Do not estimate cost, decide scope or say whether an item affects price.

Expected output

Not published yet. Treat any output as a draft and check it against the source documents.

Check the result

Every value, citation or quantity traces back to a source document.
Missing information is marked as missing, not filled in.

Limits and recovery

The AI splits addenda into rows and finds unlogged items; it does not price, interpret scope or confirm bid compliance. Check every row against the addendum and the revised sheets, and treat unclouded changes as something only an overlay and a person will catch.
Check your organization’s data policy before sharing project files with third-party tools.

Stop or switch when: The source cannot be verified, permissions are missing, or the result contains unsupported facts.

Frequently asked questions

How does AI help with this?

Claude or ChatGPT, then Copilot in Excel: The AI splits each addendum into one row per item with the sheets and spec sections it touches (NOT STATED instead of guesses), then Copilot checks the log against the estimate's cost codes before submission and lists unpriced, pending or unlogged items and gaps in the addendum sequence. A person still checks: Every row against the addendum and revised sheets, an overlay of every reissued sheet, and every pricing, scope and bid-form decision.

How do I log every addendum against the estimate and confirm each one is priced before bid day with AI?

Start the log the day the bid set comes out, not when the first addendum arrives. Use three tabs: an addenda log (addendum number, issue date, date received, item number, plain-language summary, sheets and spec sections touched, cost impact yes, no or unknown, trades affected, estimate line or cost code where it was priced, initials, status), a sheet register (sheet number, current revision, which addendum issued it) and a questions tab for clarifications you still need. When an addendum arrives, upload the PDF to Claude or ChatGPT and have it list every item as its own row: what changed in plain words, the sheet and section numbers it references, and whether it changes the bid form, bid date, alternates, unit prices or the sheet list. Reference numbers are copied exactly and anything unclear is NOT STATED. It does not price anything or decide scope. Run an overlay or compare on every reissued sheet against the previous revision (for example with the compare tools in Bluebeam Revu), because plenty of real changes are never clouded. Add each unclouded change you find as its own row. The full step list is above.

Does this use an AI agent or regular software?

It is labeled AI assists: AI drafts or analyzes. A person operates the tools, checks the work, and takes action.

What tools do I need?

Claude, ChatGPT, Bluebeam Revu, Microsoft Excel, Microsoft Copilot. Confirm license, version and cost with the vendor, and try it on a small non-confidential sample first.

Has this workflow been tested?

Workflow guide · not independently tested. Written from vendor documentation and practitioner use. Labels apply to the steps described here, not every feature of a product. Links were checked when the page was written.

Evidence and maintenance

Written from vendor documentation and practitioner use. Labels apply to the steps described here, not every feature of a product. Links were checked when the page was written.