profile

The Weekly Signal

Decide before you prompt


The Weekly Signal

September 05, 2026

A short weekly note for better work, clearer thinking, and practical systems.

Hi Reader,

This is The Weekly Signal, a short weekly note for better work, clearer thinking, and practical systems.

This week, we are looking at a common trap in how we use AI—and why the missing ingredient for better output usually isn't a better prompt, but a clearer decision.

Here is this week's useful idea, one simple system, one sentence worth saving, and one question to carry into the week.

Signal: The problem may not be the prompt

I have been using AI regularly while rebuilding ysamphy.com, writing articles, reviewing book chapters, and organizing projects.

It can make the work much faster.

But I keep noticing something.

When I am unclear about what I want, AI does not remove the confusion. It often turns that confusion into polished writing.

The result may sound professional. It may be well structured. It may even look finished.

But something still feels wrong.

I used to assume this meant I needed a better prompt. Sometimes that is true. More often, the missing piece is a decision I have not made yet.

Who is this really for?

What should the reader understand?

What am I trying to change, improve, or accomplish?

What does this piece need to do?

These questions cannot always be delegated.

AI is very good at expanding an idea, generating options, finding patterns, and helping us see alternatives. But when we ask it to begin before we know what matters, it can give us more material without giving us more direction.

That creates a strange kind of productivity. We produce more words, more plans, and more possibilities, but remain unsure what to do next.

The work becomes faster, but the decision remains unfinished.

I am learning to pause before prompting.

Not for long. Just long enough to decide what the work is supposed to accomplish.

System: Write a three line brief

Before asking AI to draft, revise, plan, or recommend something important, write three lines:

This is for: Name the person, reader, customer, or situation.

The purpose is: State what the work needs to accomplish.

The most important thing is: Choose the one idea, outcome, or constraint that should guide the answer.

For example:

This is for busy professionals who struggle to return to unfinished work.

The purpose is to give them one system they can try this week.

The most important thing is that the advice feels realistic, not idealized.

Then open the AI tool and include those three lines in your prompt.

This takes less than two minutes, but it changes the quality of the conversation. It also gives you something useful for evaluating the answer.

Instead of asking, “Does this sound good?” you can ask, “Does this accomplish what I decided it needed to accomplish?”

That is a much clearer test.

Optional resource

Related reading: AI Productivity Systems for Knowledge Workers

Sentence

AI can accelerate a decision, but it cannot spare us from making one.

Self Check

Where are you asking a tool to give you clarity before you have decided what matters?

Have a thoughtful week,

Samphy

P.S. Have you found a small habit that makes AI more useful in your work? Reply and tell me about it. I would be interested to hear what is working for you.

About This Newsletter

You are receiving this email because you subscribed to The Weekly Signal from ysamphy.com, a weekly note on better work, clearer thinking, and practical systems. Each issue includes one signal, one system, one line worth saving, and one question for the week.

Unsubscribe · Preferences

1243 S Olive, Los Angeles, CA 90015

The Weekly Signal

Subscribe to receive The Weekly Signal, A short weekly note for better work, clearer thinking, and practical systems.

Share this page