Around the time The Melting Pot hit the 135-unit mark, the fondue chain decided to reposition itself. Long four-course meals and the price tag that comes with them had turned visits into once (or twice) a year occasions. To keep growing, the 40-year-old brand was eager to increase the traffic of current customers and attract new patrons who might be more time-starved or price-sensitive. Okay. So how did they make that happen?
The franchisor realized it had reached the point where knowledge had to supplant intuition. “Our company was mostly going on gut research and had no solid information, so we wanted qualitative and quantitative information and to do a deep dive into that customer,” says Sandy D’Elosua, national director of marketing and communications for the now 142-unit chain. “We had brought on board a new senior vice president of marketing when the economy crashed, and the first thing she did was analyze what research we had, who is our customer and where are we getting that information.” Realizing that more insight was necessary, the new hire retained a market research company.
The returns on those investments were surprises about the customers The Melting Pot thought it knew well. They were younger than the chain had believed, and less affluent. “We were known for being a four-course dining destination,” says D’Elosua. “It took way too long to dine at our restaurants, and we were way too expensive.”
The Melting Pot didn’t abandon its four-course meals, but moved that option to the back of its menu. A la carte choices were added for the chain’s revised target audience, and prices were presented on a per-person basis rather than in the aggregate for a meal that serves two, a move intended to avert sticker shock.
Sales rose in test markets, says D’Elosua, declining to reveal the size of that increase. But it was sufficient to prompt a systemwide rollout in 2013. “Research was imperative to get us back on the right track and to make changes,” she says.
Actually, it was a function of getting the right research and the right interpretation of it. And that’s not the guaranteed result when commissioning market investigation, whatever the price.
Laying the groundwork
Consumer research is a tool to help you feel comfortable with the decisions you have to make when launching a new concept or revamping your brand, says Judy Hsu, marketing and consumer research manager for San Francisco-based The Culinary Edge, which has advised a number of restaurateurs on rebranding or new ventures.
Whether you work with a consultant or do it yourself, you’ll want to think first about the consumer you want to attract. That requires some unflinching discussions among company owners and c-level executives: Are you OK not appealing to all consumers? What kind of cuisine are you thinking of offering? What does your average guest look like? Will you appeal to millennials or not?
Think generally about what you want to be. “Your restaurant is still very conceptual at this point,” says Hsu. “It’s an idea about categories rather than specific dishes—creating the menu framework and essence. Should it be artisanal? Family oriented?”
Decide what questions you want to answer through your research. They may be questions about the consumer that you weren’t able to answer in your internal conversations. They may be questions to help you understand what will draw prospects in terms of food, service and ambiance. Once you know your questions, you can set about answering them through both qualitative and quantitative research.
Ask open-ended questions
Qualitative research, or drawing impressions rather than hardcore statistics, is the place to start, Hsu says. It can include talking to friends, chatting to people in restaurants or in the area you hope to open in—“anything that allows you to have an open-ended conversation,” she says.
Focus groups are popular at this stage and typically comprise eight to 10 people. “There’s a right and a wrong way to do these because you can be very leading,” cautions Hsu. If you can’t hire a market research firm for the recruitment, seek out the people you expect to be your customers. If your budget doesn’t stretch to a professional moderator, consider tapping online discussion forums where you can pose your questions.
“You’re probing,” says Hsu. “Ask surface questions such as ‘What kind of food do you like to eat?’ Then dig deeper and keep asking why, why, why.” The questions need to be open-ended, she says, “because you want the benefit of asking participants follow up questions and understanding the why.”
The insights gleaned in the qualitative stage have to be validated (or negated) through the hard numbers of quantitative research. That data is generated through surveys of groups large enough to represent the total population you want to understand. Now the questions are close-ended—for example, “On a scale of 1 to 5, how much do you like Mexican food?” or “Rank these four different cuisines in order of preference”—as opposed to the open-ended queries of qualitative research.
For a rebranding or repositioning, the questions should reveal what consumers’ feel about the concept in its current form and whether the new ideas fit the story. “We ask about brand attributes and brand clarity—what are the emotions and adjectives people use to describe it and do they match?” says Hsu. “We try to see how people perceive the brand on an emotional level.” The data should indicate what’s working and what is not—what should be preserved because it’s important to the customer base.
For a new concept or when vetting new ideas for your existing concept, Hsu suggests asking questions to measure if consumers will buy it and is the idea unique.
After you’ve done the research, go back and revisit the questions you thought of at the start to make sure you’ve answered them all, says Hsu.
You shouldn’t be navigating by gut, she adds. Use software to organize the data—Excel or something more sophisticated, like predictive analytics software. These break it all down to descriptive statistics, providing averages, means and medians “and are pretty user-friendly so you don’t have to be a statistician to understand them,” Hsu says.
“It’s still art and science,” she says—and your own experience also factors in. “At least you have the data and can decide whether to use or ignore it as you develop your concept.”