A construction firm's bid pipeline lived in 22 spreadsheet tabs, and none of its project stages had been checked.
The Client and the Problem
Our client is a construction firm that bids on public water, wastewater and school projects. Its business development team tracked every opportunity in its own pipeline sheets: 2,498 rows across 22 tabs.
That setup had three problems:
- The same project often sat on more than one tab, so the real number of opportunities was unclear
- Each project’s stage was whatever someone had last typed in, not a phase anyone had checked
- The pipeline only held projects the team already knew about
A bid team’s time goes to the projects it believes are real and close to bidding. When the list can’t tell you which ones those are, good projects get missed and dead ones get chased.
1. Merge the Sheets Into One Database
We pulled every tab into one database and combined duplicates into a single record per project. The 2,498 rows came down to 1,134 unique projects.
One list of projects instead of 22 tabs
2. Qualify by Market and Size
We kept the projects in the markets the client targets: water and wastewater in two Southeast states, and schools in northern New England. Each had to be $5M or more, or have no known value.
207 qualified projects from the client’s sheets
3. Confirm Each Project’s Stage
A project counted as confirmed only when a dated public page stated its phase, such as “design at 30%” or “bids due.” Schedules and budget lines did not count. That check also showed which projects were already decided: 30 of the 73 confirmed projects had been awarded, were under construction or had closed, so they were marked do not pursue.
73 projects with a confirmed stage, 6 of them at bids open or bids expected
4. Research Projects the Team Didn’t Have
We searched state school building-aid lists, school district pages and local press for school projects over $10M, and linked each one to the page it came from. We found 24 new projects. Eleven were outside the client’s current markets or already awarded. The other 13 qualified, worth an estimated $613M, all still at planning stage.
13 new qualified school projects, bringing the qualified total to 220
5. Put the Pipeline in Claude
The 220 qualified projects went into a private connector in Claude, so the team can ask for a list in plain words, such as “Which water projects over $50M expect bids in the next six months?” Every answer shows its source and the date it was last checked. A person reviews every change, and any source not rechecked within 30 days drops out.
The team can pull its own lists without waiting on a report
The Results
The client’s pipeline before the work, compared with the database at the time of our report:
| Measure | Before | After |
|---|---|---|
| Pipeline | 2,498 rows across 22 spreadsheet tabs | 1,158 unique projects in one database: 1,134 from the client's sheets plus 24 from our research |
| Project stage | As last entered in the sheet | 73 confirmed on a dated public page; 30 marked do not pursue |
| New opportunities | Only projects the team already tracked | 13 new qualified school projects, an estimated $613M, all in planning |
| Finding a list | Filtering spreadsheets | Asking Claude, with a source and check date on every answer |
Project values are the client’s estimates unless a public source states one. Projects in planning can still change in scope and funding.
Key Takeaway
A longer lead list doesn’t help a bid team. Knowing which projects are real, still open and close to bidding does, and so does finding the projects nobody on the team was tracking yet. That is the work behind our data intelligence service.
Want to know which opportunities your team is missing?