Here’s a question worth sitting with: if you wanted to tell your best fifty customers about a new dish tomorrow, could you? If you’ve spent the last few years on aggregators, the honest answer is usually no. You’ve served those customers dozens of times, you know their orders by heart — and you have no way to reach a single one of them. That’s not an accident. It’s the business model.

Aggregators sell you access to customers without ever letting you keep them. The customer belongs to the platform. Their contact details, their order history, their loyalty — all of it sits on the other side of a wall you’re paying a commission to stand behind. The moment you move to direct ordering, that changes: the list is yours. The interesting part is what most people then fail to do with it.

Renting customers versus owning them

The difference between a rented customer and an owned one comes down to a single question: can you start the next conversation?

With a rented customer, you can’t. You wait for the platform to surface you again, and you pay for the privilege each time. Every order is a fresh acquisition you’re renting all over again. You could have served someone forty times and still be a stranger to them — because the relationship was always with the app, and the app has no reason to introduce you.

With an owned customer, you can reach out. A new special, a quiet Tuesday you’d like to fill, a “we’ve missed you” to someone who’s drifted — these are only possible when the customer is yours to contact. That’s the whole prize of direct ordering, and it’s the one most easily left on the table. Owning the list is worth nothing if it just sits there.

Building the list the right way

A first-party customer list is one you’ve gathered directly, through your own ordering, with the customer’s knowledge. Done properly it’s an asset. Done carelessly it’s a liability, because this is personal data and UK GDPR has opinions about it.

• Collect it in the natural course of ordering. Every direct order already gives you a name, a contact, an address and a history. You don’t need a separate data-harvesting exercise — the storefront gathers it as a by-product of doing business.
• Separate ordering from marketing consent. Someone giving you their phone number so you can tell them their food is ready has not agreed to receive your promotions. Those are two different permissions, and conflating them is the most common way food businesses get this wrong.
• Make marketing opt-in explicit and specific. A clear, unticked box that says what they’re signing up for. Not buried in terms, not pre-checked, not assumed. The customer should actively choose it.
• Honour opt-outs instantly and keep the proof. Every marketing message needs an easy way out, and when someone takes it, that’s the end of it. Keeping a record of who consented to what is what turns “we think we’re compliant” into “we can show we are.”

Compliance here isn’t red tape for its own sake — it’s what keeps the asset an asset. A list built on shaky consent is one complaint away from being a problem, and customers can tell the difference between a business that respects the inbox and one that abuses it.

The second order is the one that matters

Not all customers are worth the same effort, and the data tells you where to spend it. The single most valuable moment in a customer’s life with you is the gap between their first order and their second.

A first-time customer is a curiosity — they tried you once. A second-time customer is the start of a habit, and habits are where the money is. The jump from one order to two is the biggest predictor of whether someone becomes a regular, which is why it deserves a deliberate nudge rather than hope.

• Treat the first-order window as active. A well-timed follow-up after a first order — a thank-you, a reason to come back, a small incentive on the next one — converts far better than the same effort spread thinly across everyone.
• Reward frequency, not just spend. A deal that gives money away to people who’d have ordered anyway is a discount, not a loyalty scheme. The ones that work reward the behaviour you want more of — coming back — rather than simply the size of a basket.
• Notice the lapsers before they’re gone. A regular who’s gone quiet is a customer you can still win back, but only if you spot the silence. This is where a simple recency view earns its keep: someone who ordered weekly and hasn’t in a month is a far better target than a stranger.

Segments beat blasts

The lazy version of owning a list is emailing everyone the same thing. It works about as well as you’d expect — the loyal regular and the one-time orderer and the lapsed customer all get the identical message, and it lands right for none of them.

You don’t need marketing-department sophistication to do better. A handful of rough groups is enough to be useful:

• New customers — one order, still deciding whether you’re a habit. Goal: earn the second order.
• Regulars — your core. Goal: keep them, and gently lift what they spend. Don’t discount people who don’t need discounting.
• Lapsing — used to order, have gone quiet. Goal: a reason to come back before they forget you exist.

One relevant message to the right group beats a generic blast to everyone, and it beats it by a wide margin. The list you now own is only valuable in proportion to how well you use it — and using it well starts with not treating every customer as if they’re the same.

Moving to direct ordering hands you something the aggregator years never could: a direct line to the people who already like your food. The mistake is to treat that as the finish line. It’s the starting line. The list is the asset — what you do with it is the business.