Process
Data room
The data room is the repository a credit team underwrites from. Standing it up is the moment a raise becomes real, and its condition does more to set the timetable and the diligence bill than anything else a borrower controls.
Also called virtual data room · VDR · lender data room · diligence room
A credit team does not read a room front to back. It goes straight to the things that decide whether there is a deal.
| Content | When it is opened |
|---|---|
| Constitutional documents | 3–24 on the scale |
| Contracts and leases | 28–52 on the scale |
| The financial model | 52–78 on the scale |
| Accounts, with MI reconciled to them | Everything hangs on these |
The order in which a credit team typically opens a data room. Illustrative of practice; no reading order is published and rooms are read differently by different teams.
What it is
The repository a credit team underwrites from, and the thing to understand first is that it is not a document.
The information memorandum is the argument. The model is the projection. The data room is the evidence that both are true, and a lender reads it in that spirit: not to be persuaded, but to check.
The standard to build to is a room a credit team can underwrite from without chasing. That is a higher bar than completeness, because a room can contain everything and still require a dozen emails to make sense of.
When it gets stood up
At the start of the process, and that moment is what makes a raise real.
A timetable is measured from a signed engagement letter, because that is the day the work can begin in earnest: the information request goes out, the data room is stood up, and the timetable becomes real.
Everything before that, choosing an adviser, agreeing scope, assembling a first pass of the numbers, is real time too, and it is the part wholly inside a borrower's control. So a business with a fixed deadline should treat the preparation as part of the runway rather than as a prelude to it.
The practical version: start assembling the room before you need it. Nothing in it is secret and almost all of it exists already, so the work is collation and reconciliation rather than creation.
Rooms rarely fail for missing documents. They fail because two documents in them disagree and nobody has explained why.
| Failure | What it costs |
|---|---|
| No index or structure | 4–26 on the scale |
| Stale management information | 30–55 on the scale |
| Drip-feeding material on request | 50–74 on the scale |
| Figures that do not reconcile | Turns credit into audit |
How rooms fail, ranked by how much damage each does to a credit process. Illustrative of practice rather than measured data.
What goes in it
Enough for a credit team to answer its own questions, organised so it can find them.
The financial core comes first: three years of statutory accounts, current management information reconciled to them, the model, and an earnings bridge explaining every adjustment between the reported figure and the one the facility is sized on.
Then the commercial evidence: the top customers with revenue and tenure, the contract position on each, the order book or pipeline with a note on how firm it is, and the supplier position where it matters.
Then the balance sheet detail: the debtor ledger aged, stock with a note on its basis, capital expenditure split between maintenance and growth, and the existing debt with its terms and maturities.
Then the corporate and legal layer: structure chart, constitutional documents, material contracts, property and leases, and any litigation or contingent liability. That last layer is read late but its absence is noticed early.
What does not belong in it
Anything you would not want read closely, and anything that raises a question you have not answered.
A room is not a document dump. Uploading everything on the theory that completeness is safety produces a room in which the important material is buried and the reader forms an impression of disorganisation before reaching anything good.
Board minutes are the common example. They frequently contain candid discussion of problems, options considered and risks named, none of which is improper and all of which reads badly without context. If they are being provided, know what is in them.
Draft documents and superseded versions are another. A room containing three versions of the same model with no indication which governs invites the reader to find the differences, and they will.
The test is simple: for each item, could a reader form a wrong impression from it without asking you? If so, either explain it in the room or do not include it.
Where rooms fail
Not usually on missing documents. They fail on figures that do not reconcile.
When the model, the management information and the statutory accounts disagree and nothing explains the difference, a credit assessment becomes an audit. The reader stops evaluating the business and starts establishing which number is right, and every subsequent figure is treated with the same suspicion.
Drip-feeding is the next most damaging. Material arriving one request at a time over several weeks reads as either disorganisation or reluctance, and a credit team cannot tell which. Both readings are bad.
Stale management information is third: accounts that are months old raise a fair question about how closely the business is run, quite separately from what they show.
A missing index is the mildest failure and still worth avoiding, because a reader who cannot find things assumes there is less there than there is.
Preparation is the cheapest thing a borrower can buy. It moves the timetable, the diligence bill and the terms in the same direction.
| Effect | How much it changes |
|---|---|
| First impression | 5–28 on the scale |
| The diligence bill | 35–60 on the scale |
| The timetable | 55–80 on the scale |
| Credibility of the forecast | The one that sets terms |
What a well-prepared room changes, ranked by how much. The timetable effect is charted on /library/sponsor and the diligence-cost effect on /library/quality-of-earnings; both are prose here rather than repeated as figures.
Why it changes the price
Because preparation moves the diligence bill, the timetable and the credibility of the forecast, and all three feed the terms.
On the bill: a clean, well-prepared borrower with current numbers keeps the diligence cost down, and a messy data room invites the lender to dig, and you pay for the digging. That is a direct, invoiced consequence of room quality.
On the timetable: a well-run deal with a pre-packed data room can close in about eight weeks, where a messy one can drag past five months. That gap is largely room condition rather than credit quality.
And on credibility, which matters most and is least visible. A room whose historic numbers reconcile makes the forecast worth reading. A room whose historic numbers do not means the forecast is discounted before anyone examines its assumptions, and a discounted forecast produces a smaller facility.
What the room does during the process
Actively, with one person owning it, and with a log of what was asked and answered.
Questions arrive continuously once several lenders are reading. Answering them into the room rather than by email means the next lender to ask the same question finds the answer already there, which compounds across a competitive process.
A question log earns its keep twice. It shows what the market is worried about, which is useful in itself, and it prevents the same question being answered two different ways to two lenders.
Access needs controlling deliberately. Lenders reading the same room at the same time is normal and fine; what is not fine is losing track of who has seen what, particularly where the room contains customer names or commercially sensitive contract terms.
And keep it current. A room that was accurate at launch and is not refreshed with the next month's management information ages visibly during a twelve-week process.
Acquisitions need two
Two data rooms, not one, and acquirers routinely prepare only the target's.
An acquisition facility is sized on the combined group, so the acquirer's own position has to be documented to the same standard as the target's. A buyer who has run careful diligence on the target and neglected their own numbers loses time at exactly the point where the transaction timetable is least forgiving.
The two rooms answer separate questions. The target's room supports the price and the earnings being acquired. The acquirer's room supports the capacity to carry the resulting debt, which is what determines whether the facility exists.
Both have to exist before lenders are approached. On a deal-led timetable, where someone else's exchange date sets the pace, there is no time to assemble the second one afterwards.
Common questions
What is a data room in a debt raise?
The repository a credit team underwrites from. The information memorandum is the argument and the model is the projection; the room is the evidence that both are true. The standard to build to is a room a lender can underwrite from without chasing you.
What should go in it?
Three years of statutory accounts with current management information reconciled to them, the model, and an earnings bridge. Then the commercial evidence: top customers with revenue, tenure and contract position, and the order book. Then balance-sheet detail and existing debt terms. Then the corporate and legal layer, which is read late but noticed if absent.
How many documents should a data room have?
There is no published number and a count is the wrong measure. A room can contain everything and still fail if the figures do not reconcile or the reader cannot find things. The standard is being underwritable without chasing, not a document count.
What should I keep out?
Anything that raises a question you have not answered. Board minutes are the common trap, since they candidly record problems and options in a way that reads badly without context. Superseded model versions are another. For each item ask whether a reader could form a wrong impression without asking you.
What is the most damaging mistake?
Figures that do not reconcile between the model, the management information and the statutory accounts, with nothing explaining the difference. It turns a credit assessment into an audit, and every subsequent number is then read with the same suspicion.
Does a good data room really change the terms?
Yes, through three channels. It keeps the diligence bill down, since a messy room invites the lender to dig and you pay for the digging. It shortens the timetable, since a pre-packed room is much of the difference between closing in about eight weeks and dragging past five months. And it makes the forecast credible, which is what sets the facility size.
When should I start preparing it?
Before you need it. The clock on a raise starts when the engagement letter is signed and the room is stood up, and everything before that is inside your control. Almost nothing in a room needs creating; the work is collation and reconciliation, which can be done in advance.
Do I need two rooms for an acquisition?
Effectively yes. The facility is sized on the combined group, so your own position must be documented to the same standard as the target's. Acquirers routinely prepare the target's room carefully and neglect their own, then lose time on a timetable that somebody else controls.
The full treatment sits in the guide: how long does a debt raise take.
This page explains a term as it is used in the UK lower-mid-market. It is general information, not advice on any particular facility. Terms vary between lenders and between deals, and the drafting in your own agreement governs.