Investor Due Diligence Checklist Template
Three documents and three sheets that count coverage on the files in the room rather than on the rows of your own checklist.
Free download · No account needed
Diligence Plan
[Target] — Data Room Review Plan
Room open [date] to [date]. Coverage is reported on two denominators, and the plan is scoped against reading capacity rather than against the length of the request list.
What complete means here
Two numbers, both defined before the room opens. Line coverage is request lines ticked over lines asked. Document coverage is files with a named reviewer and a review date over files in the room. They diverge, and the gap is itself the finding.
Workstreams, owners and load
| Workstream | Owner | Files | Request lines | Read hours |
|---|---|---|---|---|
| Commercial | — | — | — | — |
| Financial | — | — | — | — |
| Legal and corporate | — | — | — | — |
| Technology | — | — | — | — |
| People | — | — | — | — |
Reading capacity, stated rather than assumed
Hours available against hours of reading sitting in the room. Where the second exceeds the first, the plan names which workstream goes unread instead of discovering it in the third week.
Escalation
What severity reaches the deal lead the same day, and what waits for the findings report.
Four hundred and twelve files in the room, against a request list of 128 lines. Three withdrawn, 125 scored, 111 ticked, so the dashboard reads 89 per cent complete. Count the room instead and 313 files carry a named reviewer and a date, which is 76 per cent. Seventy-one files satisfy no request line at all, so no owner was ever assigned to them. Ten of the deal's twelve findings came off those seventy-one, including both critical ones.
Every checklist in circulation computes its percentage from its own rows. A line ticks when anything lands under it, and one row can stand for many files: 2.7 per scored line here, 3.8 in the commercial workstream. So the tick records that somebody opened the folder. Sixteen of the 111 ticked lines received part of what they asked for, two years of three, one contractor review of seven, which is how 30 open requests report as 14.
The join nobody runs is the reverse one: which line does this file answer, and if none, whose job was it. PCAOB AS 1215 already requires audit documentation to record who performed the work and who reviewed it, file by file. Diligence borrowed the checklist and dropped that granularity. Under Rule 502(b) a buyer may obtain any additional information the issuer can produce without unreasonable effort, so an ask naming a file beats one naming a category.
What is in the pack
Document Register
One row per file rather than per request line, carrying the reviewer and the review date on the file, plus the line each file answers or NONE where no line does
Outstanding Request Tracker
Regenerated from the register, separating a line that received nothing from a line that received part of what it asked for and therefore already ticked
Finding Register
A source file ID on every row and a column recording how each finding surfaced, which after a few deals shows where a firm's findings actually come from
Diligence Plan
Defines complete as full delivery plus a named reader, and states reading capacity per person per day so a late tranche can be measured on the day it lands
Findings Report
Leads with both coverage numbers and the ceiling the request list imposes on the second one, then the workstream where the two diverge most
Information Request List
Asks for the missing part in the seller's own vocabulary, so FY2023 audited accounts rather than audited accounts when two of three years are already in
The chase sweep
Runs twice a week against the live room for a request being chased that the room already answers, and for a tranche bigger than the reading time left
How it works
- 1
Register the room
One row per file, with the upload date taken from the room rather than from today. A file index or folder export is enough to start.
- 2
Join it both ways
Forward for each request line, reverse for each file. A file that answers no line gets NONE, never the nearest plausible line to tidy the sheet.
- 3
Read the room first
Before any chaser goes out, look for the undelivered documents among the files nobody owns. The ask then becomes a confirmation rather than a demand.
- 4
Report both numbers
Ticked lines and read files, side by side, never averaged, with the ceiling stated and the divergence broken out by workstream.
Frequently asked questions
How is this different from the checklist we already use?
One number instead of two. Yours divides ticked rows by total rows, which is a statement about your list. This also divides read files by files in the room, which is a statement about the room. Screening inbound before a room exists is separate, and the deal screening pack does that.
What actually ends up as a file that answers no request line?
It clusters, and the clusters are predictable. Amendments where the line asked for contracts. Side letters where it asked for standard terms. Drafts and restatements where it asked for finals. Notes and logs where it asked for a register. Each cluster is a request line somebody should have written.
Our request list is inherited and far too long. Does it need rewriting first?
No, and rewriting it first is the usual reason this never gets done. Organise it into workstreams, count the lines, and start the register. The unmapped file count tells you which lines are missing far more reliably than reading the list through would.
Does this replace the data room provider's index and Q and A?
No, it joins to them. The provider knows what is in the room and who opened what. It does not know which request line a file answers, whether that line is fully satisfied, or which files answer nothing. The register is that join and nothing else holds it.
Is reading four hundred documents in five weeks realistic?
The plan states a rate per person per day and the arithmetic follows from it. What is not realistic is the last tranche: 107 files landed with four working days left against capacity of about 54. Reading one document closely is its own job, which pitch deck analysis covers.
Why keep the withdrawn request lines instead of deleting them?
So the totals still reconcile. Here 125 scored lines plus three withdrawn equals the 128 issued, and each withdrawal carries the basis and the date. A withdrawn line that has been deleted is indistinguishable from one that was quietly answered, months later when somebody asks.
Does it decide anything about the deal?
No. It records what is in the room, who read it, and what is missing, and does not value the target. Turning statements into an earnings figure is separate work, done by the add-back review; reading whether recurring revenue survives its own termination and notice clauses is another; the investment committee memo is where all three land.
Find out how many files in your room belong to nobody
Send the file index and your request list, however rough. The first pass returns both coverage numbers, the ceiling, and every file that answers no line.
Review my data room