The most underrated skill for getting results from AI


Read online

Welcome to this week's issue of Unpacking Meaning. If you received this from a friend and enjoy it, ​subscribe here.

You know what's funny about working with AI? The more human skill you bring to the interaction, the more "intelligent" the AI seems to become.

There's this misconception that AI tools should just work magically with minimal effort. But the reality is quite different. Just like you wouldn't expect a new team member to deliver exceptional work without proper onboarding and communication, the same applies to AI.

When Anthropic (the creators of Claude) talks about building AI agents, they highlight make a great point:

"Put yourself in the model's shoes. Is it obvious how to use this tool, based on the description and parameters, or would you need to think carefully about it? If so, then it's probably also true for the model."

This is essential for anyone trying to get quality results from LLMs. The better you understand how these systems "think," the more effectively you can communicate with them.

The key elements of “AI empathy”

Based on my experience working with these systems daily, here are the key components of effective AI empathy:

1. Context management

LLMs don't have working memory like we do. They process text sequentially, and their ability to hold context diminishes over distance. That's why breaking complex tasks into manageable chunks works so well.

For example, when I write copy for a website, I always share the overall outline first before diving into section-by-section work. This gives the AI the big picture before focusing on details.

2. Clear instructions

AI systems can't read between the lines or understand implicit requests. Being explicit about what you want—including format, tone, length, and purpose—dramatically improves results.

3. Iterative refinement

Think of your AI interaction as a conversation, not a one-off request. The best results often come through multiple rounds of feedback and refinement.

4. Understanding limitations

Recognizing when an AI might struggle (with complex math, highly specialized knowledge, or maintaining consistency across long outputs) helps you structure your requests appropriately. And it’s also the mindset you need to know which model to use for what purpose.

My AI empathy process for copy projects

That said, here’s the exact process I use to create high-converting B2B SaaS copy with AI assistance:

  1. I first consider the user journey context—where does this page fit, what have visitors seen before landing here, and where will they go next?
  2. I use the AI as a brainstorming partner, exploring different angles and approaches.
  3. I always request an outline first (using a tailored template prompt + my comprehensive research and strategy as a starting point).
  4. I refine the outline iteratively with the AI until the structure feels solid.
  5. Rather than requesting the entire page copy at once, I feed the AI one section at a time, which allows for much deeper, more focused work on each component.
  6. I edit and refine manually, plus use the AI to iteratively improve specific sections.

This structured approach respects the AI's strengths and limitations while maintaining human oversight for quality and accuracy.

As AI tools become more integrated into marketing workflows, the differentiator won't be access to the technology—it'll be how skillfully you can use it.

The better you do it, the less time you'll spend fixing AI-generated content and more time leveraging it for strategic thinking and creative exploration.

So, start noticing how you interact with AI tools. Are you providing enough context? Are your instructions clear and specific? Are you working with the AI iteratively, or expecting perfection on the first try?

Small adjustments to your approach can lead to dramatically better results. And as AI continues to evolve, your ability to collaborate effectively with these systems will become an increasingly valuable skill.

DISCOVERY

🎙️The psychology and decision-making behavior behind effective copy

I recently joined the B2B Growth Blueprint Podcast with Mark Osborne to discuss the psychology behind effective copywriting. We explored how understanding user behavior and decision-making processes leads to higher-converting messaging.

Key highlights included the balance between AI assistance and human creativity, the value of qualitative research in tailoring messages, and specific website elements that create conversion momentum versus introducing friction.

If you're looking to strengthen your messaging approach, check out the full conversation for practical frameworks you can implement today.

Listen to the full episode here →

video preview

RESONANCE

"mastery is not about being special or more gifted than anyone else. Mastery is a direct result of pigheaded discipline and determination."

Chet Holmes, The Ultimate Sales Machine

Have a great weekend!

Cheers,

Chris

Chris Silvestri

Founder & conversion alchemist

🙌🏻 Let’s be friends (unless you’re a stalker)

When you're ready, here's a few ways I can help

🔍

Get a copy & UX audit

Look at my page →

🙌

Let's work 1-on-1

Get a copy coach→

✍️

Turn words into gold

See if we're a fit →

Hi, I'm Chris, The Conversion Alchemist

I'm the founder and chief conversion copywriter at Conversion Alchemy. We help 7 and 8 figure SaaS and Ecommerce businesses convert more website visitors into happy customers. Conversion Alchemy Journal is the collection of my thoughts, ideas, and ramblings on anything copy, UX, conversion rate optimization, psychology, decision-making, human behavior, and -often times - just bizarre, geeky stuff. Grab a cup of coffee and join me. Once a week, every Friday.

Read more from Hi, I'm Chris, The Conversion Alchemist

Read online Welcome to this week's issue of Unpacking Meaning. If you received this from a friend and enjoy it, subscribe here. I'm writing this as I'm about to fly back from Edinburgh after attending Turing Fest 2025. Among many brilliant speakers discussing AI, one presentation particularly sparked my thinking: Sharon Zhou, PhD, Founder & CEO of Lamini (a developer platform focused on making AI not hallucinate). In her talk about fine-tuning AI models, Dr. Zhou showed how even AI agents can...

Read online Welcome to this week's issue of Unpacking Meaning. If you received this from a friend and enjoy it, subscribe here. When most people hear "CRO," they think about tweaking button colors, A/B testing headlines, or optimizing form fields—all with the singular goal of bumping up that conversion percentage. But this approach has always felt incomplete to me. It's like focusing on the destination without considering the journey. That's why at Conversion Alchemy, we're adopting a...

Read online Welcome to this week's issue of Unpacking Meaning. If you received this from a friend and enjoy it, subscribe here. I've been working with a client recently on validating and optimizing their existing website copy, and I stumbled upon something that's been hiding in plain sight all along. While most copywriters are busy studying competitors' websites and running surveys, there's an underutilized goldmine of insights sitting right under your nose: product team interviews and user...