---
title: "You're Not Prompting Badly, You're Delegating Badly"
description: "Prompting badly is a delegation problem, your partner is the force multiplier you keep overlooking, and why saying no to a new tool still costs you time."
canonicalURL: https://aaronlynn.com/force-multiplication/youre-not-prompting-badly-youre-delegating-badly/
date: 2026-10-20T14:00:00.000Z
---

![Layered papercraft scene: on the left a paper couple climbs hand in hand up a steeply rising paper curve marked with a small ² symbol; on the right a lone paper figure slides down a shallow drooping √-shaped slope, with paper clouds and hills behind.](https://assets.aaronlynn.com/emails/newsletters/cnt-6475/papercraft-couple-climbs-exponent-curve-lone-figure-slides-root.webp)

_Power of two, or square root. Choose wisely._{.center}

## 1. Delegate to AI the Way You'd Delegate to a Person

Most of the time when someone tells me _"the AI got it wrong"_, the wording of their prompt was fine. What went wrong was the "delegation".

You already know how to delegate to a person. If you've ever managed anyone, you've done some version of [situational leadership](https://aaronlynn.com/business/effective-delegation-situational-leadership/) without calling it that. For something a person is new at or weak at, you direct heavily: a detailed spec, a worked example, and you check every step. For something they're reliably good at, you hand over the whole job with a clear definition of done and spot-check the result.

The difference with delegating to AI is that it's good at everything. As I said a couple of weeks ago, we've already reached AGI.

But for actually getting working done, we're still heavily dependent on [context and harnesses](https://aaronlynn.com/force-multiplication/ai-doesnt-know-what-good-looks-like-you-do/), and the more specialised the task, the more of a harness you need to give the AI.

Take recruitment. There are business and statutory requirements, and the AI has no idea what those are for your company. You have to put them in the harness, via skill files, rules or [review gates](https://aaronlynn.com/force-multiplication/ai-can-do-it-cant-think/). The nice part is that AI can build those for you, so hand it your existing HR policies and hiring guidelines and have it draft the rules file itself.

Check-in frequency works the same way as it does with people. If the details matter, write the prompt with checkpoints and steer it as it runs. If they don't, let it run on its own and review the finished product.

I go through the full delegation model, applied to AI, in [_ChatGPT for Managers_](https://aaronlynn.com/courses/chatgpt-for-managers/).

## 2. Your Partner Is Your Greatest Force Multiplier

There's a line about romantic partners you may have seen around online:

_"They either raise your life to the power of 2, or they square root it."_

A [force multiplier](https://aaronlynn.com/productivity/force-multipliers/) is anything that gets you more times the result for the same effort, or the same result in less time. And when you have one, it impacts everything: your finances, your health, your learning and your relationships.

We usually reach for the obvious ones, like technology, learning a new skill, picking up a new tool, or using AI. But the biggest one is the person you share your life with.

On the practical side, you get shared understandings and ways of doing things at home, a second person you trust to bounce ideas off, and if you [align your goals](https://aaronlynn.com/productivity/strategic-planning-for-couples/) you can always help each other out.

But the less-practical, more-emotional and daresay-motivational side is the bit that does the real multiplying. A partner gives you something greater than yourself to do things for. You'll work harder, smarter and longer, give up fewer times and push through more.

In general, no matter what is politically correct to say, [people are best motivated by selfish reasons](https://aaronlynn.com/productivity/motivation/#1-internal-vs-external-motivation): their own advancement, their own gain, or reducing their own pain. The two exceptions are your partner and your children. If you really care for them, you'll do and be more for them than you ever would for yourself.

I believe the reason is because when you genuinely care for someone, you subconsciously understand: "This person's future is important to my future as well."

And this results in a great sustained surge of motivation and feats you didn't know you were previously capable of.

_This does not seem to apply to parents, other relatives, a company or even the deity you believe in._

It works in reverse too. The wrong fit is a force divider, and will drain your energy in every way possible.

That's why your partner is your greatest force multiplier.

## 3. Saying "No" to a New Tool Still Costs You

Last month I set aside some time to look at 3 of my existing tools: my automation platform, my project-management tool, and how I run my coding harness. I ended up keeping the automation platform, dropping the project-management tool and updating the coding harness.

The one where I changed nothing felt free, because I didn't have to migrate anything. But I still spent time evaluating it, and that's the cost of every new tool that comes along, even the ones you end up not switching to. The trial and error adds up.

So given that there's a new tool or AI advancement almost daily now, how do you go about this?

Put a hard time-box on any evaluation before you start.

Say, one 90-minute block, and no more. And write down the decision rule before you open the tool: _"It has to save me X hours a week or it's out."_ When the box is up, decide and move on.

Be strict about which tools even earn a trial. Only trial a new tool when you have a specific problem the current one can't solve — [right tool, right purpose](https://aaronlynn.com/productivity/right-tool-right-purpose/).

"It's new and everyone's talking about it" is not a problem statement. _(It's the same reason I tell people to [stop shopping for AI models](https://aaronlynn.com/force-multiplication/stop-shopping-for-ai-models/).)_

And track how much time you spend saying no. Once you can see the number, you get a lot more selective about what gets a trial in the first place.

I cover the whole tool-selection process in [_Next Level Productivity_](https://aaronlynn.com/courses/next-level-productivity/).

## 4. Courses

I have a range of online courses that teach business people what they need to know about productivity and AI:

[**1. ChatGPT for Managers**](https://aaronlynn.com/courses/chatgpt-for-managers/)\
See how AI can solve complex management challenges in less than 30 seconds. Full prompt library and examples included.

[**2. Next Level Productivity**](https://aaronlynn.com/courses/next-level-productivity/)\
A practical, straightforward course that teaches you how to achieve elite-level personal productivity in today's constantly interrupted world.

[**3. The Complete Claude, Claude Code & AI for Work Productivity**](https://aaronlynn.com/courses/ai-for-work-productivity/)\
A comprehensive course taking you from beginner AI concepts like prompt and context engineering, to cutting-edge AI productivity using terminal-based AI tools like Claude Code for non-coding office work. Usable with ChatGPT, Claude, Gemini and other current LLMs.

[**4. The AI Playbook**](https://aaronlynn.com/courses/ai-playbook-for-ceos/)\
This is my longer premium course on how businesses can deploy AI tools and technology across their processes and teams.

[**5. The AI Agents, Automations and Agentic Workflows Guide**](https://aaronlynn.com/courses/ai-agents-guide/)\
This non-technical course shows business people and non-coders how they can build and use AI agents in ChatGPT and Zapier.

That's it for this week!

— Aaron

> Related course: ChatGPT for Managers — $97 — /courses/chatgpt-for-managers/