Architecture Power Hour — independent software architecture consulting

Independent software architecture consulting, in one prepaid hour

You have one technical decision to make and nobody independent to put it to. Your vendor has an opinion and an invoice attached to it. Your developers have a preference. Nobody in the room is both competent and disinterested.

This is software architecture consulting stripped to its useful core: one prepared hour, one question, and a written record of the answer you can show to a board, a co-founder or a vendor.

What you get

A prepared hour

You send the material — the estimate, the architecture diagram, the vendor proposal, the repository — at least two working days before the call. The hour starts with the question, not with context-gathering you are paying for.

A one-page decision record

The decision, the reasoning behind it, the two or three conditions that would change it, and what to watch over the next quarter. One page, because a decision that needs ten pages has not been made yet.

The alternative you did not pick, and why

Every real architecture decision has a runner-up. Writing down why it lost is what stops the same argument being reopened in three months by someone who was not on the call.

How the week runs

Day 1 — you send the question

One paragraph is enough. If the question turns out to be five questions, I say so before you pay, and we pick the one that unblocks the others.

Days 2 to 3 — I prepare

I read what you sent. If something critical is missing I ask for it now rather than spending your hour on it.

Day 4 — the call

Sixty minutes, recorded if you want it. No slides.

Day 5 — the written summary

The decision record lands within one working day of the call.

Build vs buy, specifically

The most common question this hour gets asked is build vs buy: a team wants to build something that already exists as a product, or wants to buy something that is genuinely their differentiator. Both mistakes cost roughly the same, in opposite directions. An hour spent on it before the commitment is the cheapest hour in the project.

A worked example

A founder arrived with a vendor estimate of 45 days for a multi-tenant SaaS backend and wanted to know whether it was real. It was not — the estimate assumed a single-tenant data model, and the multi-tenancy the product needed had not been priced at all. The hour cost $200. The gap was worth about 25 developer-days, which at that vendor's rate was a little over $12,000 that would have surfaced as a change request halfway through.

Who this is not for

• Anyone who wants a second opinion to use as ammunition against their own team — that is a management problem, not an architecture one
• A general technical review of everything (that is the Delivery Risk Audit)
• Anyone who needs the work done rather than the decision made

What is not included

Nothing written beyond the one-page decision record, no follow-up calls, no review of code or documents beyond what fits the hour's preparation, and no implementation. If the answer turns out to need a real audit, I will say so on the call rather than sell you another hour.