Restaurant growth funnel

Growth that pays for itself.

In 2020, Fush wanted to grow without growth eating its own margin. For a full table-service restaurant, more customers usually means more staff and more cost. We built a funnel that brought in new customers profitably, then pushed that demand online so the kitchen could absorb it without the wage bill climbing to match.

Online ordering growth, 2020 to 2022

20x

Marginal profit per new customer

$18

Return on ad spend

8x

Margin per customer, vs 10% typical

40%

The funnel, end to end.

The funnel ran on Meta ads into a connected stack of HubSpot, Zapier, and Coupon Carrier, with a custom POS integration so a scanned QR code tied straight to a customer's account in store. A prospect saw an ad, signed up, and got a unique two-for-one offer by email, led with the secondary menu, fish and chips and fried chicken, so it pulled people in without discounting the hero dishes.

Paid once, came back twice more.

The clever part was what happened after they redeemed. We kept the email and retargeted them with automated follow-ups: a second offer, then a third. About 35 percent reconverted at each tier, so a single acquired customer was brought back at least twice more. Those return visits cost almost nothing to drive, no ad spend, just the overhead of the email and marketing software. So while the first visit cleared about $18, each repeat visit cleared closer to $23, because there was no acquisition cost left to carry.

Twentyfold in two years.

It all worked because we advertised to new customers only, so every ad dollar bought incremental revenue instead of subsidising regulars. With wages and overheads already covered by the existing business, those extra customers did not need extra staff, right up to a critical mass of about a 20 percent lift in sales. To hold that advantage as volume grew, we drove orders online, so the kitchen could absorb the demand while the floor stayed free for returning and local diners. Online ordering sales grew twentyfold between 2020 and 2022.

The whole funnel buys the third visit.

There was a reason we engineered for exactly three visits. Restaurant loyalty does not behave like e-commerce. As restaurant operator John Taffer describes it, a flawless first visit gives a customer only about a 30 percent chance of returning. A flawless second lifts it to around 40 percent. But a flawless third visit takes the odds of them coming back to roughly 75 percent. Three good visits turn a one-off into a regular, so the whole funnel exists to buy that third visit. After it, the customer keeps coming back on their own, at no marketing cost at all.

Per new customer.

Leads came in under $2, and 35 percent converted, so a new customer cost about $5.50 to acquire. Average spend after discount was $43.50. The food needs honest accounting: the two-for-one meant a meal worth $20 to $25 left the kitchen free, and that food still costs about 30 percent to make. So the real food cost is closer to $20, near 45 percent of the discounted bill. Even after that, each new customer leaves about $18 of marginal profit.

Marginal basis: wages and overheads sit still because the funnel only adds incremental customers the existing base already covers.

Margin is a point of view.

Around $29 a customer, before loyalty.

Each repeat visit was driven by email to a customer we had already paid to acquire, so it carried no ad cost, only the small overhead of the email and marketing software. That makes every return visit a little more profitable than the first. Blended across three visits at a 35 percent step, an acquired customer is worth around $29 of marginal profit, and that is before the long-term loyalty from the third-visit effect compounds on top.

Reconversion ran at about 35 percent per tier, with no acquisition cost on the return visits.