There's a lot of advice out there on how to 'get better results from ChatGPT' (including this piece from me on how to get ChatGPT to sound like you). Most of it looks something like:
Refine your prompts.
Tell it your tone.
Use frameworks.
Be more specific.
And while those things help, they won't fix this: If your inputs aren't grounded in what your audience actually thinks and feels, your outputs will always fall short.
This is where Voice of Customer (VoC) research changes everything (and not just for AI).
What is Voice of Customer research?
Voice of Customer research is about listening before you write. It’s the process of gathering real, specific, emotion-rich language directly from your ideal clients. Think:
Exact phrases from interviews, testimonials, and reviews
Questions your audience is already asking
Objections they express in their own words
Words they use to describe their pain points, desires, or transformation
It’s discovering your brand’s messaging goldmine, straight from the source.
Why this matters (with our without AI)
Even if you never used AI, VoC would still be one of the most powerful tools in your marketing toolbox. But if you are using AI tools like ChatGPT, it becomes your distinct advantage.
Here’s why:
How ChatGPT works (and where it falls down)
ChatGPT is trained on broad data from the internet. It’s incredibly good at patterns, but not at nuance. It doesn’t know your audience unless you feed it their words.
If you’re prompting it to write:
A headline
A landing page
An email
...but you haven’t given it the language your people actually use, you’ll get something that feels flat, vague, or worse, templated. The result? Generic messaging that sounds like everyone else.
What happens when you feed VoC into AI
When you take real customer language and use it in your prompts — or even better, embed it into your brand messaging framework — AI can finally:
Mirror your audience’s thoughts and emotions
Write in a way that feels truly resonant
Hit the right tone and messaging for your specific niche
Create copy that connects (not just converts)
How to do Voice of Customer research
Even a simple process can yield huge insight:
Read your reviews or testimonials
Look for repeated phrases, emotional triggers, and specific results.Interview past clients
Ask open-ended questions: “What made you decide to work with me?” or “What were your priorities when it came to choosing a service provider?”Check online communities
Reddit, Facebook groups, forums — anywhere your audience talks without being sold to.Run a survey
Ask what they’re struggling with, what they’ve tried, and how they’d describe a successful outcome.
Then… capture the language. Build a doc. Highlight emotional phrases. Copy and paste. Use it in prompts. Bake it into your website. Repeat.
Or, let me do it for you
I offer Voice of Customer research as a done-for-you service that helps you uncover what your audience really cares about — and how they talk about it.
You’ll get:
A VoC insights report
Messaging themes and word-for-word quotes
Copy-ready insights you can use across your content
Prompts for ChatGPT that lead to more compelling, human-first output
This is something that will pay off across every piece of content — whether it’s AI-generated, copywritten, or spoken aloud in a podcast or pitch.
Great messaging requires input from real voices
Voice of Customer research is the difference between:
Guessing vs. knowing
Sounding like everyone else vs. standing out
Content that talks at people vs. copy that speaks to them
If you’re serious about making AI work for your brand — or just writing better, more resonant content — this is the place to start. If you want to learn more, or you're ready to book in your VoC project, get in touch for an initial chat.
Or check out my (copywriter-approved) brand foundations system which gives you the full package with everything you need to get great outputs from AI tools.
