Playground acquisition diligence

The buyer sent over 400 requests in a spreadsheet and stood up the VDR. I built an app to track them internally. Each request had an owner, a status, and a due date, and the app synced with the data room every night.

I also connected Claude to the tracker. It could read a request, draft an answer, and attach a file. I checked it before anything was posted.

The app had a page for the merger agreement too. Each clause had a risk level and a suggested response, and I sent our comments to counsel from there.

The buyer's consultants' written questions had their own queue. We drafted, reviewed, and signed off on every answer in one place.

I sent the deal team an email every night at 9pm. It had the day's progress, what was overdue, and a leaderboard. Claude wrote a short “gossip column” from what changed in the tracker that day.

Each day gave us a "happiness meter" from the perspective of the buyer's diligence lead (an estimate). It went down when items were overdue or reopened. I changed his LinkedIn picture to have twelve different facial expressions based on the ranking. Don't worry, we sent it to him after the deal, and he loved it.

Alongside the deal I was asked a simple question: if this doesn't go through, how long is our runway and what should we cut?

The only true way the team tracked burn was looking at the bank account balance each month. There was no granular deep dive.

I pulled the raw data from each source and tied it to the books. Then I built a dashboard for burn and runway on top of it.

It has a cash forecast that I tested against past months. It also has a cut board that shows when each cut would actually hit the bank.

For the buyer: Excel

The buyer's first ask was a financial model and a forecast. I built one in Excel from scratch, since nothing existed since the Series A docs. Revenue starts from the number of active schools and builds up to payments and SaaS. Headcount and opex come after that.

Every line ties back to the general ledger. A script builds the whole workbook, and it checks every tie-out before I send a new version.

For the team: a dashboard

The internal team did not speak Excel. They kept asking for fewer spreadsheet-style views, so I built them a dashboard on top of the same model.

It runs the same formulas in the browser. Anyone on the team (with Google Auth access) could move a lever and see what happened to cash and runway without opening a workbook.