I asked Claude to interview me 100 times. Here's what I found out.

I tell my clients to do this all the time. Sit down. Think about who you are. Think about what you stand for. Think about how you want people to feel when they interact with your brand.

Simple advice. The kind of thing that sounds obvious when you say it out loud.

Turns out, I'd never actually done it myself.

Not taking my own advice

I spend my working life helping people discover their brand voice, their identity, the emotional thread that ties everything together. I ask them questions they haven't thought to ask themselves. I help them articulate things they can feel but can't quite put into words.

And yet, when it comes to my own brand, my own voice, my own story; I've been winging it. Not badly, I think, but without the depth I'd demand from a client.

So I decided to put myself through the process. Properly. One hundred questions about how I think, how I write, what I believe, what I can't stand, what makes me tick. The twist? My interviewer was an AI.

Why AI?

I know, I know. An article about how great AI is from someone who uses AI. Groundbreaking.

But hear me out. The reason it worked wasn't because the AI was clever (although, credit where it's due, some of the questions were uncomfortably sharp). It worked because I don't have these conversations with anyone. Not really. Not in this depth.

My wife knows me. My brother knows me. My kids know me. But none of them have ever sat me down for two hours and said: "You say you value authenticity, but your LinkedIn posts sound like everyone else's. Why is that?" Nobody has ever said to me: "You keep hedging with 'I think' and 'maybe' when you clearly have strong opinions. What's going on there?"

The AI did. And it didn't let me off the hook when my answers were vague.

What I discovered

Some of it I already knew, deep down. I knew I was warmer than most people in my space. I knew I cared more about emotional connection than SEO rankings. I knew I'd rather sound like a real person than a polished professional.

But there were things that surprised me.

I didn't realise how much my teaching background still runs through everything I do. Fifteen years in the classroom left its mark; some of it painful, some of it invaluable. The instinct to explain why something works, not just tell someone it does. The need to make things accessible without dumbing them down. The belief that learning should feel alive, not like a dissection. All of that is still there, just wearing different clothes.

I didn't realise how consistently I undermine my own authority. The AI pointed out, mid-conversation, that I'd answered a question about whether I hedge by hedging three times in a single sentence. I said: "I think it's a bit of both... I definitely think I still think about... maybe I like confidence." Three hedges. In a sentence about hedging. You couldn't make it up.

And I didn't realise quite how central humour is to everything. Not comedy, not jokes (well, maybe the occasional dad joke). But that playfulness, that lightness, that refusal to take things too seriously. When the AI asked me to name the single most important thing about good writing, I said one word: humour. No qualifiers, no hedging. Just humour. It was the most decisive answer I gave in two hours.

The uncomfortable bits

The AI asked me what I'd never written about, and I gave the safe answer about ethics and industries I wouldn't work for. Fair enough.

But then it asked me whether I unconsciously undermine my own sales pitch because the selling part feels gross to me. And I had to sit with that for a moment, because the answer is yes. Probably. A bit.

I've never been financially motivated. I've always been the person who helps first and worries about the invoice later. That's a lovely quality in a friend and a slightly problematic one in a business owner. I'm working on it. The blog post you're reading right now is, in its own quiet way, part of that work.

It also asked me about imposter syndrome. I don't have formal qualifications in graphic design or web design. Everything I do is built on taste, experience, a lifetime of absorbing culture, and the stubborn belief that I know what good looks like even if I can't always show you the certificate. That's terrifying and liberating in roughly equal measure.

What I'd tell you

If you run a business and you haven't done this exercise, do it. I don't necessarily mean with AI (although it's surprisingly good at it). I mean sit down, in whatever way works for you, and answer honestly:

• How do you want people to feel after they've interacted with your brand? Not what do you want them to do. How do you want them to feel?

• What would you never do, even if it worked commercially?

• What does your voice sound like when you're not trying to impress anyone?

These aren't abstract branding questions. They're the foundations that everything else sits on. Your logo, your website, your LinkedIn posts, your emails; they all flow from the answers. I know this because I tell my clients this every week. I just hadn't done it myself until last Saturday.

The result

The output of those 100 questions is a document that captures how I think, how I write, and what I stand for in enough detail that someone (or something) could write in my voice. Not perfectly. Not as a replacement. But as a starting point that's actually me, rather than a generic approximation.

It's the same thing I build for my clients when I do their brand work. The difference is, this time the client was me. And it turns out, I'm just as bad at articulating my own brand as everyone else is.

The good news? That's exactly what I'm here for.

If your brand feels like a collection of parts that don't quite hold together, or if you know what you want to say but can't quite find the words, maybe it's time to have that conversation. I promise I'll ask better questions than you'd ask yourself.

Help is, as always, on the way.