Why the order matters
Sellers usually appoint an adviser and then start preparing. That works, but it wastes the first two months of an engagement on work the seller could have done alone, and it means the adviser is building a buyer list before anyone knows what is actually being sold.
The reverse order is better. Eight weeks of internal work before the first adviser conversation changes the conversation itself, because you arrive with a defined asset rather than an intention.
Weeks one and two: settle the perimeter
Decide what is being sold and write it down. Products, customer contracts, people, intellectual property, domains, data, and the shared services the division consumes.
This is a decision, not an analysis. Where the answer is honestly unclear, choose the version that produces the cleaner asset and note the alternative. Ambiguity here propagates into everything downstream, and it is the single most common cause of re-trading later.
Keep the working group small and under a confidentiality agreement. Most leaks in the preparation phase come from uncontrolled internal circulation of documents rather than from buyers.
Weeks two to five: build the standalone numbers
Three parts, and the third is the one that matters.
Revenue the division earns from its own customers, stripped of internal transfer pricing and inter-company arrangements. Direct cost, meaning the people, infrastructure and vendors that exist solely to serve it. Then the share of group functions it consumes, costed at what a standalone company of that size would actually pay rather than at your group's allocation formula.
That third figure is where credibility is won or lost. Buyers build their own version and compare. A seller whose model is defensible line by line gets less scrutiny on everything else; a seller whose number is obviously flattering invites a full re-underwriting.
While you are in the accounts, pull the retention data. Net and gross revenue retention by cohort, and the revenue share of your top ten customers. You need to know both before anyone quotes you a multiple, because those two numbers move it more than sector does.
Weeks four to six: find the problems yourself
Run the diligence a buyer would run.
Are customer contracts assignable, and do any need consent. Are contractor IP assignments complete. Has anyone reviewed the open-source licences in the codebase. Is there a certification the group maintains that the division would have to maintain alone. Is there litigation, and is it disclosed.
None of these are difficult to check and all of them are expensive to discover in week six of a live process. Finding them now means you fix what is fixable and prepare an explanation for what is not.
If the asset is large enough to warrant it, a vendor-side quality of earnings review belongs here. Finding the adjustments yourself is materially cheaper than having a buyer find them for you.
Weeks six to eight: the people question
Identify who has to stay for the business to be worth what you think it is worth. Usually it is a small number: the division lead, one or two engineers who hold critical knowledge, and whoever owns the largest customer relationships.
Decide what retention looks like and what it costs. Have the conversations that can be had confidentially. Buyers price management continuity directly, and a departure announced mid-process is close to fatal.
What you now have
At the end of eight weeks you have a defined perimeter, standalone financials with a documented basis, a known list of diligence problems with a plan for each, retention and concentration data, and a secured team.
That is a sellable asset rather than an intention to sell, and it changes the adviser conversation from a discussion about possibilities into a discussion about price, timing and which acquirers to approach.
The decision this work also informs
The other thing eight weeks buys you is the ability to decide not to proceed.
Sellers who start with an adviser and a mandate acquire momentum, and momentum makes it awkward to conclude that the right answer is to wait a year. Sellers who do the preparation first frequently find that retention needs two quarters of work, or that the perimeter question has an answer the board has not agreed, and they postpone deliberately rather than discovering it halfway through a live process.
An asset that goes to market and does not sell is harder to sell afterwards. The buyer universe is finite and it remembers. Eight weeks is a cheap way to be sure.
Divestitures.com Research · Published for orientation, not as advice on a specific transaction. Any figure cited is orientation, not a valuation. See market notes.