By the time the bankers are appointed, technology diligence readiness is fixed. Whatever the estate looks like at that point is what the buyer's advisors will examine, and the only remaining variable is how quickly and confidently you answer.
That is why the work belongs a year earlier, when it is cheap and unhurried.
What the buyer's team will actually ask for
Their questions are predictable, which is what makes preparation possible.
A complete application and infrastructure inventory with costs. Contracts, renewal dates and change of control provisions. Licence compliance, particularly for the databases and enterprise software where audits are common. A security posture summary with evidence: controls, incidents, insurance and its claims history. Key person dependency in the systems that run the business. Technical debt: what is unsupported, what is customised beyond upgrade, what depends on somebody who left. And the roadmap, with what has been committed and what it will cost the buyer after close.
None of these is unreasonable. All of them take weeks to assemble from scratch and days to produce if the register already exists.
Where value leaks
Three patterns cost sellers money, consistently.
Unquantified debt. The buyer finds an unsupported ERP and prices a replacement into their model, at their number rather than yours. If you have already scoped it honestly, you control the figure being discussed.
Licence exposure. Unresolved compliance questions get escrowed, and escrow is a discount with extra steps.
Key person risk. The finance systems that only one person understands, the customisations only one contractor can maintain. Buyers price this, and it is one of the cheapest things to fix a year out and one of the most expensive to explain away in a data room.
What to do twelve months out
Build the register. Applications, infrastructure, contracts, costs, owners. Keep it current rather than assembling it for the process. This is the single highest-return item, because it turns most of the request list into an export.
Close the licence questions. Audit yourself before somebody else does, and remediate quietly.
Reduce key-person dependency. Document, cross-train, and get the undocumented customisations understood by more than one person.
Make the security story evidenced. A risk register with owners and review dates, incidents recorded and closed, insurance in force and consistent with what you have warranted.
Decide what you will not fix, and write down why. A seller who says "we know, here is the cost, here is why we chose to leave it for the buyer" is in a far stronger position than one who appears not to have noticed.
Six months out
Freeze the estate. Now is not the time to start a migration, because a half-finished programme is worse in diligence than either not starting or having finished. Finish what is nearly done, park what is not, and keep the register current through the process.
The reframe worth making internally
Diligence readiness is often treated as a cost of exit. It is more accurate to treat it as part of the value creation plan, because everything on the list above also makes the business run better while you still own it.
A current asset register lowers spend. Resolved licence exposure lowers risk. Reduced key-person dependency lowers operational fragility. You would want all three even if you never sold.