AI Guides › Claude Mastery

The Difference Between Asking For An Answer And Asking For A Method

By Nigel Guy · 2 min read

"What should I do here?" is the most natural question in the world to ask, and it's often the wrong one, because it hands the entire judgement call over and gets back a single confident-sounding recommendation with no visible working. It reads decisively. Whether it's actually right for your specific situation is much harder to tell, precisely because you can't see how it got there.

The rule: asking for an answer gets you a conclusion you have to take on faith; asking for a method gets you a way of deciding that you can check, reuse, and disagree with on the merits.

The mechanism: reframing the request

Before sending a decision question as-is, check whether it's asking for an answer or a method, and reframe it deliberately if it's the former:

  1. Instead of "what should I do," ask "what are the factors that should decide this, and how would each push the decision one way or the other." This surfaces the actual reasoning as something you can weigh yourself, rather than a single verdict you either accept or reject wholesale.
  2. Instead of "which option is better," ask for the criteria a good decision here would need to satisfy, then apply them to your specific options yourself, or ask it to. This separates the (genuinely useful) judgement about what matters from the (much more fallible) judgement about your particular case.
  3. Instead of "write the plan," ask for the planning approach — what order to tackle things in and why, what the biggest risks are — and then have it produce the specific plan from that approach. The approach is the reusable part; the specific plan is disposable and situation-bound.

Why the method version holds up better

A method exposes its own weak points. If a piece of reasoning is thin, asking for the method surfaces that thinness directly — you can see which factor is doing the most work and question it. A bare answer hides the same thinness behind confident phrasing, and confident phrasing is not evidence of a sound method underneath it.

What to skip

Skip asking for a method on questions that are genuinely simple factual lookups — "what time zone is Tokyo in" doesn't need a framework, it needs an answer. Reserve the method-first approach for judgement calls: decisions with real trade-offs, ambiguity, or consequences, where the reasoning behind the answer matters as much as the answer itself.

Guardrails

All 751 AI guides · JulieMango plans from £17/mo