I was on a call a few weeks back with someone I’ll call Priya (names have been changed to protect the innocent). Priya, the ops lead for one of our clients, was going through last month’s site numbers. Visits were down slightly at one of her busier stations. I asked why.
She shrugged it off, “Probably just quiet. Summer holidays, maybe.”
Nobody had actually looked into it, and there was no reason to. There’d been no complaint. And there was no churn alert anywhere in the CRM.
Her shrug makes complete sense once you realise the evidence is scattered across four different, disconnected systems.
The EV transition creates a difficult challenge. There’s no single moment when a customer switches from petrol to electric, just a gradual change in pattern that, six months later, turns up as an unexplained dip someone shrugs at on a call. Just the way Priya did.
By which point, somebody else has already won that customer.
Fuel retailers are sitting on genuinely brilliant data – transaction frequency, basket data, loyalty behaviour, channel preference. The kind of customer signal most industries would sell a kidney for. But the infrastructure holding it together was built for a world where people filled up with petrol and came back on the same day, every week. Like clockwork. Predictable and simple.
The EV transition doesn’t play by those rules.
Nobody switches from petrol to electric on a Tuesday and vanishes by Wednesday. It’s much trickier to spot than that…
The visit frequency drifts.
The dwell time creeps up.
The basket changes shape.
Each one of those, on its own, looks like nothing. A blip. And blips are easy to ignore, to write off as some weird anomaly.
But, if you were to put them all together, they’ll give you a clear picture of what’s actually happening. But, for most fuel retail data setups, it’s not that simple, because the same customer is living as four separate profiles across four separate systems that have never once been introduced to each other.
The till system thinks they’re still a regular. Loyalty still has them down as “active,” because they redeemed points eight months ago and nobody’s touched that definition since. The app’s given up entirely, since it hasn’t heard from them in months. Meanwhile, the actual human being behind the data, is visiting a different forecourt, and not one system in the stack is shouting about it.
The question I asked Priya, and would throw open to anyone in fuel retail that I speak to, is – Could you tell me, right now, using only the data you already have, which of your customers switched to electric in the last year? And which ones are about to?
Not after three weeks of someone in the data team stitching CSVs together by hand. Right now.
For most fuel retailers, if we’re honest, the answer’s no. The data’s probably all there, somewhere. It’s just scattered across four siloed systems, and nobody’s ever thought of pulling them together to agree on who’s who. Or been able to work out how.
That’s what fixing the data in our Agentic CRM framework, actually means in practice. It’s not glamorous. Hardly sexy. But, resolving who your customers actually are means the signals already sitting in your systems become visible. That’s actually a pretty exciting outcome, if you ask me. Important work, too, especially while there’s still time to do something about them.
The fuel retailers taking this seriously have already clocked what’s happening – EV is showing up in their data every single day. Insights that are readily surfaced and acted upon, without anyone sitting there looking for it.
Our Agentic CRM Playbook walks through exactly this, data layer first, campaign second. So next time someone like Priya gets asked why the numbers dipped, “probably just quiet” isn’t the only answer she’s got.
