Support · Diligence
What a target IT estate review should actually produce
Most diligence estate reviews produce a spreadsheet. Here is the shorter list of artefacts that survive contact with the integration programme.
Every acquisition produces an IT estate review. Most of them produce the same thing: a large spreadsheet, assembled under time pressure, with columns added by whoever was awake at the time. It is delivered to the deal team, cited once in the integration plan, and then quietly abandoned because nobody can tell which rows are still true.
The problem is not effort. It is that a spreadsheet is a snapshot without provenance, and the programme needs a record with provenance.
The artefacts worth paying for
A review that will still be useful after close produces five things, in this order.
1. A scope statement. Which entities, which tenants, which subscriptions, which sites, which countries. Written before collection starts, and revised in the open when the scope changes, because it always changes.
2. A source register. For every system touched: what was read, with which credential, at what time, by whom, and with what permission scope. This is the artefact that answers the security question in the data room, and it is the first thing missing when a review is done by hand.
3. The estate record. The consolidated entities, with each attribute traceable to the source that produced it. Not a merged average. A record where a disagreement between two sources is visible as a disagreement.
4. A conflict register. The rows where sources disagree, each with an owner and a decision state. Open conflicts are information. Silently resolved conflicts are future incidents.
5. An assumptions and gaps note. What could not be read, why, and what the programme should assume in the absence of a reading. A review that does not state its gaps is claiming a completeness it cannot have.
What to stop asking for
Stop asking for a single “final” spreadsheet. The request guarantees a document that is wrong within a week and cannot be corrected without a new version.
Stop asking for a completeness percentage. The number is invented at the point of presentation and means nothing without a denominator nobody agrees on.
Stop treating the review as a diligence deliverable rather than an integration input. The same record should serve the diligence appendix, the wave plan and the post-close baseline. If it is built only for the appendix, it will be rebuilt for the plan, and the two will disagree.
A test for any review you receive
Ask one question: show me where this number came from.
If the answer is a source system, a timestamp and a credential, the review has provenance and can be maintained. If the answer is a person’s name, the review is testimony. Testimony is fine in a meeting. It is not a baseline.
Ask a second question: what could you not see?
If the answer is “nothing”, the review is wrong. Every estate has a system that would not respond, an application with no inventory, or a site nobody could reach in the window. A review that reports no gaps has not looked for them.
Disagree, or want the detail
Write to hello@acqi.tech. Corrections are published with the article rather than quietly applied.