Restaurant Business image with AI
Loyalty programs are one of the biggest trends in restaurants right now. They have become a go-to strategy for driving traffic and frequency and for gathering information on customers. A steady stream of restaurant brands have launched or re-launched programs. There’s a sense that loyalty is now table stakes for modern restaurant marketing.
But something is holding loyalty back.
A new report from consultant Restaurant Loyalty Specialists (RLS) reveals a disconnect between what restaurants want out of their loyalty programs and what they are actually able to accomplish. RLS founder and Principal Olga Lopategui talked to leaders from 53 chains about their loyalty strategies, and their responses paint a picture of a technology that has clear potential, but is currently limited by a lack of resources, insufficient tech and ROI that can be difficult to measure.
The report goes into great detail, more than we can delve into here. But here are a few of the biggest challenges it identified, as well as some of the solutions it uncovered.
The ‘myth’ of one-to-one marketing
One of the big promises of loyalty programs is their ability to unlock personalized or “one-to-one” marketing. Once a customer enrolls in a program, the restaurant can begin tracking their preferences and ordering habits and then sending them marketing messages tailored to those behaviors.
Restaurant brands talk about personalized marketing all the time. At last year’s Restaurant Leadership Conference, the topic came up so often that I wrote a whole column on it.
But according to the RLS report, few restaurants are actually doing it. Just a quarter of the brands Lopategui spoke to said they are using their loyalty programs to enable targeted marketing, while the rest are still sending identical “batch and blast” emails to large groups of customers.
“Personalization is a myth,” said one marketing VP from a 100-unit casual-dining chain. “We say we personalize our email, but in reality, we don’t.”
It’s not hard to collect customer data; using it to power personalization is another matter. Forty-five percent of operators told RLS that they lack the staff or resources needed to do the sort of sophisticated analysis that one-to-one marketing requires. Another 35% cited technology limitations.
“The data is there, and technically we have the ability to do it, but it’s actually much harder in practice,” said the VP of marketing for an 80-unit fast casual.
Other obstacles: data quality/visibility (20%) and creative assets (20%). Indeed, if every customer is getting a bespoke email, the restaurant has to generate that many unique images, subject lines, body copy, etc. It can be a lot to ask, especially for a small team.
On the bright side, marketers that are doing personalized marketing seem to be coalescing around the best way to do it: Talking to customers about things they’ve ordered in the past. The report refers to this as “product propensity” messaging and concludes that it is the holy grail for marketers.
“Let’s sell sushi to the people who love sushi, and tacos to people who love tacos,” said the senior director of loyalty at a 500-unit QSR.
The AI trust gap
I can already hear vendors’ response to some of these concerns: “That’s what the AI is for!”
It’s true that many loyalty providers have developed AI tools that allow restaurants to query and analyze their customer data and even generate and run entire marketing campaigns on their own. It’s just that most operators don’t trust them.
Brands interviewed by RLS noted concerns over AI hallucinations, limited contextual awareness and a lack of accountability, as well as poor data infrastructure on their end. For those reasons, 51 of the 53 respondents said they are not ready to launch AI-powered loyalty campaigns.
“There is no accountability for AI yet,” said the head of loyalty for a large fast-food chain. “If it fails, they turn off the platform; if I fail, I lose my job.”
That said, most operators believe that AI will be the answer—eventually. Sixty percent said AI will be key for scaling personalized marketing. And the report itself concludes that restaurants should be using AI to automate personalized offers.
“It would be incredible to just tell the platform what you need—like targeting all the ‘queso lovers’—and seeing the campaign materialize instantly without the usual technical friction,” said the CMO of an 80-unit fast casual.
The ROI conundrum
At the end of the day, the true test of anything a restaurant does, whether it be a loyalty program, restaurant prototype or menu item, is whether it generates some kind of return.
When it comes to loyalty, ROI remains murky.
Part of the problem is that many operators (40%) said they are not even attempting to measure the ROI on their loyalty programs. Another 35% are studying ROI at the campaign level, while just 15%, or eight of the 50-plus operators surveyed, say they are tracking the ROI of the entire program. This last group is mostly enterprise-level chains, the report says.
“I’ve not done deep ROI analysis on anything,” admitted the CMO of a 50-unit fast casual. “I feel like loyalty is table stakes we just need to have. And so we picked [platform redacted], it was the lowest cost option.”
Most vendors offer some sort of estimate of the ROI their programs deliver. But some restaurants (more than a dozen, in this case) simply do not trust them.
According to the report, loyalty providers sometimes advertise up to a 15% lift from their programs while ignoring basic inputs like food costs and discounts. Restaurants’ own calculations typically show an ROI of around 4%, per the report.
“Even if [the vendor] did provide a number and it was positive, I would not believe it,” said the fast-casual CMO.
To be sure, the ROI on loyalty programs can be hard to measure. Do loyalty members really visit more often, or were they already frequent visitors to begin with? It’s a classic chicken-or-egg situation. And it’s related to the fact that the vast majority of restaurant transactions are still from non-loyalty customers. Your loyalty members may seem to visit a lot, but it’s not always clear how their behavior compares to the average customer.
Brands that are measuring the ROI on their programs have tackled this problem in a couple of ways. One common strategy is to create “global holdouts”—control groups of loyalty members who don’t receive any marketing messages. This allows chains to see how their campaigns perform in comparison.
Another strategy is to analyze customers’ visitation before and after they join the loyalty program, which is possible by using a customer data platform (CDP) that ties together various data sets into a single “source of truth.” This also allows restaurants to compare loyalty and non-loyalty customers.
The good news is that operators using CDPs do tend to find that loyalty members’ annual spending will double or triple after joining a program, according to the report. This is a major validation of loyalty programs’ effectiveness generally.
Without that visibility though, it’s hard to know for sure how any given program is performing or what it could be doing better.
“Without a CDP, I can’t compare members and non-members. There is no baseline,” said a senior director of loyalty at a 200-unit QSR.
It’s no surprise then that 38% of respondents are using or in the process of launching a CDP, while another 48% said they’re interested in using one. The few skeptics note that CDPs can be expensive and complex to use.
The bottom line here is that loyalty programs are popular for a reason: There’s compelling evidence that they get customers to visit more often.
To take them to the next level, restaurants need to prove their value, and vendors need to improve their technology. If these things happen, marketers stand a better chance of getting the resources they need to reach their program’s full potential.
