# What executives should ask before approving a network transformation > A transformation programme arrives at the executive table as a vision, a supplier, a budget and a date, and the technical detail underneath it is exactly what the people at the table were hired not to need. They do not need it. They need twelve questions whose answers tell them whether the detail was done, who is carrying which risk, what the organisation will be unable to undo, and what the first ninety days will cost in attention. The questions, why each one matters, and what a good answer sounds like. Source: https://ronutz.com/en/learn/questions-executives-should-ask-before-a-network-transformation Updated: 2026-10-03 Related tools: https://ronutz.com/en/tools/change-blast-radius-mapper, https://ronutz.com/en/tools/change-window-runbook-builder, https://ronutz.com/en/tools/health-snapshot-comparator, https://ronutz.com/en/tools/digital-transformation-tracker --- ## What the table is being asked to do A network transformation, whatever its name this year, is a decision to replace the way the organisation connects and protects itself, over a period of one to three years, at a cost that is visible, with a disruption that is not. The executive who approves it is not being asked to understand the technology. They are being asked to judge whether the people proposing it have understood it, whether the risks are where the proposal says they are, and whether the organisation can carry the change while doing everything else. The [digital transformation article](https://ronutz.com/en/learn/digital-transformation) is about what the word means; this one is about the signature. The twelve questions below are the ones that find out. Each has a reason and a description of what a good answer sounds like, because the quality of the answer is the evidence. ## About the decision itself **1. What decision are we actually making today, and what are we deferring?** A good answer separates the irreversible commitments (a contract, a cut-over) from the ones that can wait for evidence, and names a date for each deferred decision. A proposal that needs everything decided now is usually a proposal with a discount that expires. **2. What has to be true for this to be the right decision, and have we checked?** A good answer is a short list of assumptions with a word next to each: verified, or not. The [architecture-review method](https://ronutz.com/en/learn/review-a-network-architecture-before-approving-it) calls this the assumption register, and it is the single most useful page in any proposal. **3. What would make us wrong in two years?** An acquisition, a regulation, a supplier's roadmap, a change in how people work. A proposal that cannot name anything has not been thought about as a bet, and the table is being asked to place one. ## About the material in front of us **4. Which parts of this proposal were written by the supplier, and which by us?** The honest answer is usually "most of it by them". That is normal; what matters is whether somebody on our side has rewritten the claims into our terms and tested them, which is what an [independent review](https://ronutz.com/en/advisory) is for. **5. What was measured, and what was estimated?** Traffic, users, sites, growth, savings. A good answer attaches a date and a source to each number. A number nobody can source is a negotiating position. **6. Who else runs this exact combination, at our scale, and have we spoken to them without the supplier present?** References are chosen. The useful reference is the one that resembles us and the one that left. ## About what we lose and what we cannot undo **7. What will our own people no longer be able to see or do, and what do we get instead?** Every modern design moves functions into things the team cannot log into. A good answer names the diagnostics the team loses and the ones the supplier exposes, in writing. **8. What does the exit look like, and what does it cost?** At the end of the term, or earlier if it goes wrong: what do we own, what can we export, in what format, in how many days, and what would it take to run without this supplier? A good answer is a plan that has been written; a bad one is a paragraph that says there is one. **9. Which of the promises in this proposal are in the contract?** Service levels, renewal prices, data location, export, support hours, the roadmap items we were shown. A good answer is a list with page numbers. ## About carrying the change **10. What does the first ninety days cost in people, and whose people?** The overlap period, when both the old and the new exist, is paid for twice in money and three times in attention. A good answer names the team members who will do the work and what they will stop doing. **11. Who on our side is accountable for the technical outcome, and do they have the authority to stop it?** Not the sponsor; the person who can say the cut-over does not proceed tonight. If nobody has that authority, the supplier does. **12. How will we know it is working, before the business tells us it is not?** Baselines taken before the change, measured against after; the [snapshot comparator](https://ronutz.com/en/tools/health-snapshot-comparator) is built for that comparison. A good answer names the measures, who reads them and when. ## What a good set of answers produces Not a yes or a no. A conditional approval: these commitments now, these decisions deferred to these dates with this evidence required, these questions answered in writing before the first cut-over, this person accountable, this review date. Written down, so the approval survives the people who gave it and can be examined when the programme is halfway through and somebody asks what was agreed. The [example decision memorandum](https://ronutz.com/en/advisory/decision-memorandum) shows what that document looks like for a programme of this kind. ## The question behind the questions Every one of the twelve is a version of the same question: has somebody who is not selling this, and who understands it technically, read it on our behalf and written down what they found? If the answer is yes, the twelve will be answered quickly from that document. If the answer is no, the table is being asked to approve a proposal it cannot read, on the word of the people who wrote it. If a transformation programme is coming to your table and that reading has not been done, it is the kind of work I do professionally: [independent technical advisory](https://ronutz.com/en/advisory).