What happens when you invite AI into the insights process? You get faster framing, bolder brainstorms, and the occasional Spice Girls lyric—plus a surprisingly reflective creative partner who never needs coffee breaks. This post is part experiment, part confessional, and entirely co-written with a large language model. Verdict? AI didn’t replace the work. It helped me get to the good stuff faster. Would collaborate again. Possibly would buy it a coffee out of sheer gratitude.
Jessup Smith
Sr. Insights Lead, Nottingham Spirk
Partnering with AI: Practical Use, Curious Impact, and a Dash of Wannabe
When you work in insights at a company that builds across design, engineering, and strategy, you get used to navigating ambiguity with a smile and a spreadsheet. You learn to translate ideas between disciplines, decode vague stakeholder asks, and sometimes say, “That’s not an insight—it’s just an opinion in a blazer.”
So when AI landed on my desk—courtesy of leadership enthusiasm and a deck that included the phrase “transformative intelligence revolution”—I approached it with cautious optimism. You know, the way one might approach a goose. Confidently, but with a healthy respect for chaos.
The pitch? Faster synthesis. Smarter output. Fewer late nights rewriting the same bullet three times to make it sound just provocative enough. Fine, I thought. Let’s see what this thing can actually do.
Cut to me asking it to reframe a product insight and sneak in a Spice Girls lyric without blinking. (We’ll circle back to that.)
What I found wasn’t magic. It was something weirder—and more surprisingly useful.
How AI Has Fit Into Our Work—Practically Speaking
Here’s where it’s actually delivered (yes, for real):
🧠 Synthesis Support
Dozens of stakeholder interviews. Piles of transcripts. Cross-functional chaos. I used AI to pull themes, contradictions, and summary drafts in a fraction of the time it usually takes. Did I still have to interpret the nuance? Of course. But it gave me room to think instead of drowning in quotes.
🧩 Framework Drafting
I’ve fed AI scraps of notes and let it spin up draft journey maps, needs-based segmentations, and behavioral clusters. Sometimes it nails it. Sometimes it confidently invents a framework I’ve never seen before and calls it “established.” But it always gets the ball rolling—and that’s half the battle.
🗣️ Story Framing
Crafting narratives for different audiences is where AI has genuinely shone. I’ve used it to tailor the same story three ways: insight-driven for product, emotional arc for design, and “what’s the risk if we don’t do this” for leadership. It even suggested I open one presentation with, and I quote, “Picture this: frustration, confusion, and unmet needs walk into a usability test.” Honestly? Not bad.
🧪 Hypothesis Testing
When I feel stuck, I prompt it with a scenario and ask, “What am I not seeing?” It responds like a thoughtful-but-unfiltered colleague. Occasionally brilliant. Occasionally off-the-wall. Always pushing me to think harder. The Stewart to my Squirrelly Dan, if you will.
People of the world, spice up your life! (Photo: Chat GPT)
What AI Doesn’t Do (And Shouldn’t)
Let’s be clear:
- It doesn’t catch nuance in team dynamics.
- It doesn’t know when a client’s “That’s interesting” is actually code for “please stop.”
- It doesn’t track unspoken context or organizational politics.
- And it definitely doesn’t know when it’s talking too much—which makes it the opposite of Wayne.
The real insight work—pulling signal from noise, finding what matters, making it land with humans who have deadlines and budgets and opinions—that’s still on us. But AI gives us a mental exoskeleton. It handles the repetitive strain so we can focus on the good stuff.
Tell Me What You Want, What You Really Really Want
If we’re serious about “unlocking the value of AI,” let’s name what it’s actually good for:
- ✨ Speeds up the boring bits
- 🪞 Reflects and refines half-baked thinking
- 🧲 Generates momentum when ambiguity stalls us out
- 🎤 Helps me reword an insight using Spice Girls lyrics and not make it weird (jury’s out)
The real power here isn’t automation. It’s amplification. It helps us be more human—not less. Like a chorus in your head that’s decent at metaphors and terrible at eye contact.
And Now, a Word from the Director
Written by the human who asked the AI to write this post. Ghostwritten by the AI that was asked by the human. Yes, it’s turtles all the way down.
When I first started working with AI, I figured I’d use it once, roll my eyes, and go back to working the old-fashioned way: with too many docs open and a barely-warm cup of coffee. But that’s not what happened.
Instead, I found a kind of creative partner. One that doesn’t flinch at uncertainty. One that can draft six intros in under a minute. One that helps me think faster without thinking for me.
So when I was asked to write about using AI in insights work, I figured I’d write it with AI. Let it riff. Let me edit. Let it try a metaphor about rollercoasters. Let me delete that metaphor about rollercoasters.
And then, naturally, I told it to take about 10% off there, bud.
This post is the result of that partnership.
I wanted to show how AI fits into the work.
It wanted to make a Spice Girls joke.
We both got what we wanted.
That’s what I appreciates about it.
—The Director (Me, but also... not entirely)
Submit a comment