I Built AIgnite's First Real Case Study Before Asking a Newcastle Café for a Penny

A few weeks ago I sat down with a café owner in West Jesmond, asked her what was actually eating her time, and built something to fix it — before I ever asked her for money. No pitch deck. No sales call. Just questions, then a working system. Here's the honest, slightly messy story of how that's actually gone.

It Didn't Start With a Pitch — It Started With a Question

I think most "AI automation for small business" stories start the wrong way round. Someone builds a tool first, then goes looking for a business desperate enough to need it.

I did it backwards, mostly by accident. I already knew Alex, who owns Awesome Franks — a genuinely brilliant, Wes Anderson-inspired café tucked into West Jesmond Metro Station, all warm teal walls and terracotta accents, the kind of place that photographs itself. Through a Marketing Society connection (I'm the Treasurer, more on that in a second), I ended up sitting down with her properly, just to ask: what's actually slowing you down right now?

Not "would you like a chatbot." Not "have you considered automation." Just — what's annoying about running this place day to day?

A few weeks earlier, I'd had a genuinely game-changing 30 minutes with Simon Taylor, a business strategist I got connected with through Northumbria's SGE Incubator. One line from that conversation has stuck with me since: audit for free first, build trust before building revenue. So that's what I did. No pitch. Just questions.

What Awesome Franks Actually Needed (Spoiler: Not a Chatbot)

Here's what came out of that conversation — no growing email list despite a genuinely strong Instagram following, no reliable way to track whether marketing effort was actually working, and Google reviews that could use a real, honest push.

None of that needed AI wizardry. It needed a system that quietly does the boring, repetitive bits so Alex could focus on the parts of running a café that actually need a human — the food, the space, the people.

So that's what I built. Three connected pieces:

• A landing page where a student scans a QR code and instantly gets a discount code — no payment wall, no faff
• An automated email sequence — a welcome message, a reminder if the code's unused after 24 hours, and a Google review request a day after they actually redeem it
• A live dashboard, so Alex can see, in real time, exactly how many people signed up, redeemed, and how many review requests went out — labelled honestly as requests sent, not confirmed reviews, because there's genuinely no way to auto-track the latter and I'm not going to pretend there is

The Bit I Got Wrong First (And Why That's Actually the Interesting Part)

Here's the honest bit most "case study" posts skip.

My first version made students pay £5 for Northumbria Marketing Society membership before they got their café discount. Made total sense on paper — bundle two wins into one action, get a society sign-up and a happy student at the same time.

It was also, I realised about two days in, exactly the wrong call. Asking someone to hand over money as their first ever interaction with a brand-new system is a genuinely huge ask — way bigger than it feels when you're the one building it. So I ripped that bit out. Now every student gets their discount instantly, zero payment required, and the £5 membership gets a separate, much warmer invitation once they're already in the WhatsApp group, already feeling good about the free coffee they just got.

Turning Up in Front of the Camera, Not Just Behind the Code

A few days ago I found myself on the wrong side of a camera for once — playing a genuinely frazzled student who'd just missed the Metro, wandered into Awesome Franks in a mild panic, and left twenty minutes later with a coffee, a discount code, and a noticeably better afternoon. Small thing, but it reminded me that automation is never the whole story. The actual experience still has to be worth someone's time.

The full clip is on Instagram if you want to see it.

What Happens Next — And Why I'm Not Going to Fake the Numbers

Northumbria's Freshers Week ran from 21st to 27th September. That was the actual test — a real stall, a real QR code, real students, and a dashboard to tell us, honestly, whether any of this genuinely worked.

I'll publish the real numbers once they're in. Sign-ups, redemptions, and hopefully a real, visible lift in Awesome Franks' Google reviews from wherever they're sitting today. No inflated stats, no vague "hundreds of happy customers" language. If it works, I'll show you exactly how much. If it half-works, I'll show you that too.

This is AIgnite's first real, external case study — not because I finally found a client, but because I finally have real numbers coming, built the honest way round: understand first, build second, prove it third.

Update, October 2026: the numbers are in. 48 sign-ups in a week, 37 of them in one afternoon, 182 emails delivered and zero spam reports. Read the full case study →