CIM Review Checklist: What a Partner Checks

Five steps, in order, each with one name against it

By the time a CIM reaches the partner, the team has read it too many times to see it. This is the read that catches what they stopped seeing, and it runs in the order a buyer would run it.

Anything wrong in a CIM costs more later. A number that moves after release has to be explained to every bidder, and each one starts to wonder what else will move. Read it on paper, end to end, in one sitting.

1Final draft
Deal team
Ties every number to the financial model and the accounts, and clears the file of comments and hidden content.
2A clear day before release
Partner
Reads in order, as a buyer would, and marks each check below.
3After the partner's read
Seller
Confirms the facts in writing, and approves names, biographies and the forecast.
4Before release
Counsel
Supplies the notice page and confirms the version for competing buyers.
5Release day
Deal team
Sends through the data room, watermarked, and logs recipient, version and date.

Every check has a place on the page

Most of what goes wrong in a CIM is visible on the summary page, if you know where to look. Here is a sample opening page with ten checks pinned to where they apply.

Exhibit 1

Ten of the 30 checks can be run on the opening page alone

Sample CIM summary page for Project Harbour with checks pinned where they apply. £m unless stated. Year ends 30 June.

Project HarbourConfidential information memorandum

1

A £60.0m UK specialist distributor, with 20.0% adjusted EBITDA margins, for sale in full

Revenue FY26A: £60.0m
Adj. EBITDA FY26A: £12.0m4
Margin: 20.0%
Staff: 210

2

Reason for sale. The founder is retiring after 28 years and will stay for a handover.

3

Perimeter. The whole company and both sites. The main site is owned by the founder and leased to the company.

9

Market. About 18% of a £330m UK niche (Trade Report, 2026).

6

Current trading. Q1 FY27 revenue is 2.0% ahead of budget and 9.0% ahead of last year.

5

Revenue and adj. EBITDA, £m, year to 30 June, FRS 102

49.0

9.2

FY24A

54.0

10.6

FY25A

60.0

12.0

FY26A

66.0

13.5

FY27F

7

FY27F is management's forecast: 6.0 of growth from two contracted accounts and a price rise.

8

FY26A revenue by customer, £m and % of total

Customer A: 8.4 14.0%
Customer B: 6.6 11.0%
Customer C: 4.8 8.0%
Customer D: 3.6 6.0%
Customer E: 3.0 5.0%
Others: 33.6 56.0%

10

1
The opening says what is for sale
What it does, who it sells to and how big it is, before any market background.
2
The reason for sale holds up
Will the founder say the same thing in a meeting?
3
The perimeter is clear
The main site belongs to the founder, and the page says so.
4
Every headline ties out
£12.0m is the bridge in Exhibit 2, the financial section and the financial model.
5
Periods and basis are labelled
Year end, actual against forecast, and the accounting basis.
6
Current trading is in
Year to date against budget and last year, on the first page.
7
The forecast has an owner
Labelled as management's, with the driver named next to the number.
8
Customers follow one key
Customer A here is Customer A on every later page.
9
Every market figure has a source
Share and market size carry a report name and a date.
10
The project name is the only name
Footer, file name and document properties all say Project Harbour.

Note: Market share, the report name, the staff count and the trading figures are sample content, written to show what a sourced claim looks like.

Source: Sample data. Project Harbour and every company, customer and vendor shown are invented.

Adjusted EBITDA is the number the whole process prices

Every add-back is worth the multiple. At 10.0x, a £0.6m adjustment that fails in diligence takes £6.0m off the headline. So read the bridge as a buyer's accountant would: ask what document supports each line, and whether the cost really stops.

Exhibit 2

£1.8m of net adjustments takes reported EBITDA of £10.2m to £12.0m, and each line needs a document

Reported to adjusted EBITDA, Project Harbour, FY26A. £m. Year ended 30 June 2026. Axis starts at £9.0m.

10.2

Reported EBITDA

0.6

Litigation settled

0.5

Owner pay above market

0.6

Site closure

0.3

Systems migration

(0.2)

Rent to market level

12.0

Adjusted EBITDA

Adjustment	£m	Support	Reviewer's note (1)
Litigation settled	0.6	Settlement agreement, index 2.3	Genuinely one-off: the claim is closed.
Owner pay above market	0.5	Pay benchmark, index 5.3	Needs a named replacement cost, not a guess.
Site closure	0.6	Closure costs, index 2.3	A smaller closure in FY24A makes this look recurring. Explain it or drop it.
Systems migration	0.3	Invoices, index 2.3	Check nothing here is ordinary running cost.
Rent to market level	(0.2)	Lease and valuation, index 6.1	A negative add-back: the founder charges less than market rent.
Net adjustments	1.8

(1) A reviewer writes this column before a buyer's accountant does. Index refs point to the data room index.

Source: Sample data. Project Harbour and every company, customer and vendor shown are invented.

One number, one value, in every document

Bidders put the teaser, the CIM, the management presentation and the data room side by side. Build the same sheet first. When a number changes, fix it at source and rerun the whole sheet, because a fix made on one page is how the next mismatch starts.

Exhibit 3

The tie-out fails on 2 of 6 lines: fix them at source, then rerun the whole sheet

One number across four documents, Project Harbour. Units as stated on each line. As at the release draft.

Line	Teaser	CIM	Management presentation	Data room	Result
Revenue FY26A, £m	60.0	60.0	60.0	60.0	Ties
Adj. EBITDA FY26A, £m	12.0	12.0	12.0	12.0	Ties
Adj. EBITDA FY27F, £m	13.5	13.5	13.2	13.5	Does not tie
Top five customers, % of revenue	44.0%	44.0%	44.0%	44.0%	Ties
Staff at year end	210	210	210	205	Does not tie
Year end	30 June	30 June	30 June	30 June	Ties

Note: The forecast in the management presentation and the staff count in the data room are older versions. Both are invented for this example.

Source: Sample data. Project Harbour and every company, customer and vendor shown are invented.

30 checks, in the order a buyer reads

The story

A buyer decides whether to keep reading in the first few pages.

The opening says what is for sale, and why now.What the business does, who it sells to, how big it is and why the owners are selling, before any market background.
Why it mattersA buyer is deciding whether to assign people to this, and that decision is made early.
What going wrong looks likeFive pages of market background come before the company. The reader skims, and their first question shows they missed the point.
The reason for sale holds up.Succession, a partner leaving or the need for a bigger owner is fine if it is true and the seller will say the same thing in a meeting.
Why it mattersIt is the first thing a buyer tests, because it tells them what the seller knows that they do not.
What going wrong looks likeThe CIM says retirement and the founder talks about staying for growth. Now every other statement is in doubt.
The perimeter is clear.Which entities, brands, properties and people are in the deal, and which are not. Anything the shareholders own and the business uses is called out.
Why it mattersA price only means something against a defined perimeter.
What going wrong looks likeThe main site turns out to belong to the founder. It surfaces in diligence and reopens the price.
The story belongs to this company.Take the name off. If a sentence could describe a competitor, it is not doing any work.
Why it mattersA generic story earns a generic price.
What going wrong looks likeThe strengths page lists people, quality and customer focus. Any competitor could sign it.
Risks are framed, not hidden.Customer concentration, key-person dependence, a soft year. A buyer will find each one, so let them read your framing first.
Why it mattersA risk disclosed with context costs less than the same risk discovered.
What going wrong looks likeConcentration appears for the first time in the data room. The bidder reprices and asks what else was left out.
A reader abroad can follow it.Spell out local terms, tax regimes and rankings that mean nothing outside the country.
Why it mattersForeign buyers took 52% of UK-target deals in 2025, per our deal market study, so the likely reader may not know the market.
What going wrong looks likeA foreign bidder cannot place the company in its market, and asks for a second call that should not have been needed.

The numbers

Every headline ties out.Revenue, margin, EBITDA and headcount on the summary page match the financial section, the financial model and the statutory accounts.
Why it mattersBidders put the CIM against the accounts within hours of receiving it.
What going wrong looks likeRevenue on the summary differs from the financial section on a rounding basis. It is small, and it says nobody checked.
Adjusted EBITDA has a bridge.Reported to adjusted, one line per item, each with a document behind it in the data room. An item that recurs every year is not a one-off.
Why it mattersThe price is a multiple of this number, so every add-back is worth that multiple.
What going wrong looks likeA closure cost appears in two of three years. Quality of earnings strips it, and the bid falls by the multiple.
Owner costs are normalised.Owner pay at a market rate, family on the payroll, personal costs through the company, and rent on property the owner holds.
Why it mattersA buyer prices the business under new ownership, so these costs adjust in both directions.
What going wrong looks likeThe founder draws a low salary and no adjustment is made. The buyer adds a replacement cost and EBITDA falls.
Periods and basis are labelled.Year ends with dates, actual against forecast, LTM with its end date, and the accounting basis, including any change between years.
Why it mattersAn unlabelled period forces the reader to guess, and they will guess against you.
What going wrong looks likeA change of accounting basis looks like margin growth until the diligence adviser unpicks it.
Current trading is in.Year to date against budget and against last year.
Why it mattersIt is the first thing a buyer asks for, and its absence reads as an answer.
What going wrong looks likeA bidder learns the year is behind budget only after reading a forecast that never mentioned it.
The forecast has drivers and an owner.Growth by volume, price, new customers or capacity, not a percentage. It is labelled as management's forecast and management has approved it.
Why it mattersManagement will be asked to defend it in a meeting.
What going wrong looks likeManagers meet bidders and cannot explain the growth line they supposedly approved.
Next year can be walked from today.The step from current trading to the first forecast year rests on known contracts, pipeline and price. A sharp turn needs a reason on the same page.
Why it mattersThe first forecast year is the one bidders can test, and it carries the most weight.
What going wrong looks likeA flat year is followed by a sharp rise with no named driver, and the bidder's financial model resets it to flat.
Units and arithmetic hold.One unit and currency throughout, percentages add to 100, totals foot, and charts agree with the tables beside them.
Why it mattersArithmetic errors are the cheapest way to lose a reader's trust.
What going wrong looks likePercentages add to 101, and a careful reader starts checking everything else.

The claims

Every market figure has a source and a date.Size, growth and share.
Why it mattersA buyer's commercial diligence team will check them.
What going wrong looks likeAn old market report is quoted as current, and the bidder's adviser has the latest edition.
Superlatives have evidence.Leading, unique, proprietary. Prove it on the page or cut the word.
Why it mattersOne unsupported claim discounts the supported ones next to it.
What going wrong looks likeA trade buyer in the sector knows the real ranking, and stops believing the rest.
Customer statements are true as written.Long-term means a signed term, recurring means contracted, and retention is calculated on a stated basis.
Why it mattersThese statements drive how much of the revenue a buyer will pay for.
What going wrong looks likeLong-term contracts turn out to be rolling agreements with short notice periods.
Biographies are agreed.Each person has read their own line. Anyone likely to leave at the sale is not presented as the future of the business.
Why it mattersA buyer is pricing the team as well as the numbers.
What going wrong looks likeThe finance director named as key to the plan says in a management meeting that they intend to leave.
The seller has confirmed the facts.Written confirmation that the document is accurate to the seller's knowledge. A partner's read does not replace it.
Why it mattersThe adviser relies on the seller for the facts, and the confirmation records that.
What going wrong looks likeA factual error surfaces in diligence, and nobody can show who approved it.

Confidentiality

Customers and suppliers follow one key.Customer A is the same customer on every page, and the description does not give the name away.
Why it mattersAnonymisation only protects the seller if it is consistent.
What going wrong looks likeCustomer A on one page is Customer B on another, and a bidder works out the name by cross-reading.
People are not exposed.Staff appear by role and pay by band. Managers are named only where the seller has agreed.
Why it mattersStaff lists are personal data, and they are poaching lists.
What going wrong looks likeA named organisation chart reaches a competitor, which starts calling the senior engineers.
The file is clean.No tracked changes, comments, speaker notes, hidden slides, author metadata or chart workbooks embedded behind the slides.
Why it mattersWhat cannot be seen on the page can still be in the file.
What going wrong looks likeA chart's embedded workbook holds the customer list by name, and someone double-clicks it.
The project name is the only name.Check the file name, footers, image names, alt text and document properties for the company's real name.
Why it mattersOne stray property is enough to break a blind process.
What going wrong looks likeThe PDF's author field carries the seller's name, and a teaser recipient works out who it is.
Competing buyers get their own version.No customer-level pricing, margins or pipeline. Ask counsel where the line sits on this deal.
Why it mattersIf the sale fails, a competitor still holds what it read.
What going wrong looks likeA trade bidder drops out after reading customer margins, then uses them in the next tender.

Release

The notice page is current.It comes from counsel, names the right parties and covers each country the document will reach.
Why it mattersIt sets the terms on which the document is read.
What going wrong looks likeAn old notice names a previous entity, or leaves out a country the CIM is going to.
A signed NDA is on file for the exact recipient.Including any portfolio company, co-investor or adviser who will read it.
Why it mattersAn NDA binds the party that signed it, not its sister company.
What going wrong looks likeA portfolio company reads the CIM under its sponsor's NDA, which does not cover it.
It agrees with everything else.Same numbers as the teaser, the data room and the management presentation.
Why it mattersBidders compare the documents side by side.
What going wrong looks likeThe first mismatch makes a buyer doubt the rest.
Release is logged.Recipient, version and date, in the buyer tracker. Send through the data room, watermarked, with print and download set on purpose.
Why it mattersIf the document leaks, the log is how you find out where.
What going wrong looks likeSomeone emails a PDF outside the data room, and nobody can say who holds which version.
What comes next is ready.The process letter, the date for indicative offers and the route for questions are ready on the day the CIM lands. Use the Q&A log from the first question.
Why it mattersInterest is at its highest the day the CIM lands.
What going wrong looks likeFirst questions sit unanswered for a week because no route was set up.
Someone has read it cold.Ask a colleague outside the deal to read the opening. If they cannot say what the business does, it is not finished.
Why it mattersThe team cannot unsee what it knows.
What going wrong looks likeThe opening assumes knowledge only the team has.

How good teams still get this wrong

The mistake	What it costs	Do this instead
Reading on screen in tracked changes.	You review the edits rather than the document, and never read it as a buyer will.	Read a clean copy on paper, end to end, in one sitting.
The partner reads after the seller signs off.	Every comment reopens a sign-off, and the seller loses patience with the process.	Partner first, seller second, counsel third.
Fixing a number on one page.	The summary now disagrees with the financial section, and nobody knows which is right.	Keep one source for headline numbers and rerun the tie-out after every change.
One document for every buyer.	A competing buyer reads customer margins it should not see.	Decide the competitor version when the buyer list is agreed, not at release.
Treating the forecast as the adviser's work.	Management cannot defend it in a meeting, and the bidder discounts it.	Management builds and owns it. The deal team tests and presents it.
Leaving current trading to the Q&A.	Bidders assume the worst, and the first answer lands badly.	Put year to date in the CIM, and update it through the process.

CIM partner review checklist

Five steps, in order, each with one name against it

Every check has a place on the page

A £60.0m UK specialist distributor, with 20.0% adjusted EBITDA margins, for sale in full

Adjusted EBITDA is the number the whole process prices

One number, one value, in every document

30 checks, in the order a buyer reads

The story

The numbers

The claims

Confidentiality

Release

How good teams still get this wrong

Read with this

Rather hand this work over?