You have ten minutes with an executive team to get approval for an architectural investment. How do you structure it?

What is being tested

Whether you structure for a decision and adapt the altitude to the audience.

The structure

1. The ask, first (30 seconds). "I am asking for six weeks of two engineers' time to address the checkout reliability problem." Executives want to know what is being requested before they hear why. Building up to it risks running out of time before reaching it, which happens constantly.

2. The problem in their terms (2 minutes). "We fail about 4,000 checkouts a month. Roughly 30% of those customers do not return in-session. At £28 average order value that is around £34,000 a month, and the trend is up."

Numbers they already track. Not error rates.

3. Options with costs (3 minutes). Not one proposal — two or three, with what each buys. Executives make trade-offs for a living; give them one to make. It also demonstrates that the recommendation was reasoned rather than preferred.

4. The recommendation and why (2 minutes).

5. Risks, honestly (1 minute). What could go wrong and what you would do about it. Concealed uncertainty is discovered later and costs credibility; stated uncertainty is credible.

6. What you need from them (1 minute). A decision, a budget, a priority change, a conversation with another leader. Be specific.

Leaves 30 seconds of slack, which you will need.

What to leave out

Component diagrams. Technology names. Implementation detail. The reasoning that led you here — they are buying the conclusion and its justification in business terms, not the derivation.

The practical points

• One idea per slide. A slide with four ideas conveys none.
• No diagram they cannot read in ten seconds. If it needs explaining, it is the wrong altitude.
• Numbers with context. "£140,000 a month" means nothing; "£0.004 per order against a £28 order value" is actionable.
• Anticipate the obvious objection and address it before it is raised.
• Know what you will cut if you get five minutes instead of ten, which is common.

The most common failure

The technical deep-dive to executives, which loses the room in the first two minutes and produces "thanks, interesting" rather than a decision.

If nobody needs to decide anything, write a document instead — a meeting to convey information is a document read aloud.