Care to Comment?

Larkspur Hospitality brings analytics to hotel guest surveys IF A COMPANY only relied on feedback from customers who were either delighted or infuriated, chances are its strategic thinking would be as schizophrenic as its input.

And if those decisions were based on a small sample – say, 1% – of its client base, and a self-selected one at that, the reactions would be even more skewed.

Yet this is the case with hotel comment cards. Unless a guest has had either a wonderful or terrible experience, the cards will likely remain blank.

Larkspur Hospitality Co., a Corte Madera, CA-based lodging chain, is using a combination of hard and soft science to collect usable reactions from its guests, while at the same time avoiding pitfalls a self-selected sample would offer.

Larkspur has been working with a hotel guest-satisfaction measurement firm, Market Metrix LLC of San Rafael, CA, since the fall of 1998. With Market Metrix’s help, Larkspur selects guests before checkout, allowing the chain to solicit comments from a sample that represents its total client base. At the point of departure, while room statements are being printed, these guests are asked to complete a brief survey.

“Before, we had a desk comment card that we created,” says Larkspur Hospitality’s head of guest relationship management Lisa Bellavance. “It was nothing statistically significant. But guest satisfaction is critical. We wanted to focus on getting the most feedback from [them] in the most efficient manner.”

Through focus groups, Larkspur has identified secure customers, which it characterizes as repeat guests likely to recommend the hotel to friends and associates. Bellavance is well aware that increasing the percentage of secure customers will have a positive impact on the bottom line.

These guests show high correlation with certain measured attributes. Loyal customers tend to react well to friendliness at the front desk, or to employees who exhibit a can-do attitude. Using this information, Bellavance can let her property managers know that these are valued attributes, and make sure they highlight these to their guests.

Larkspur requires that managers at any property garnering an overall score under 85 out of a possible 100 points put an action plan in place. This could include employee incentives or procedural changes.

But at Larkspur such plans aren’t needed very often. Among the three branded property lines the company manages (Larkspur Landing, Candlewood Suites and Hilton Garden Inns), the average overall satisfaction rate stands at 91, compared with an industry average of 84.

Larkspur’s ranking would likely be higher, says Bellavance, if not for consumer misperceptions of the Hilton Garden Inns. The other two brands lend themselves to longer-term guests, who have more opportunities to familiarize themselves with offered amenities. Guests at these locations also know not to expect full-service hotels.

But the Hilton name tends to attract a clientele familiar with other Hilton properties, and when they get to the Garden Inns they’re confused by the lack of a bell captain or other amenities. When they realize that these offerings aren’t available, the hotel’s rating on the “value for price” category drops, even though these properties tend to be competitively priced.

Larkspur collects between 60 and 100 completed surveys per month per property in order for Market Metrix to perform its analyses. Survey quantities vary, as several of Larkspur’s locations are geared toward guests who visit for longer periods – fewer checkouts means fewer opportunities to collect surveys. But the direct solicitation approach pulls a 72% completion rate, compared with the 1% response to in-room cards.

The higher completion rate isn’t the only benefit of the personal appeal approach. “When you’re doing a traditional telephone or mail survey, you have recollection bias. You don’t remember the specifics of your stay, which is essential if a hotel is giving feedback to its employees and it wants them to learn and improve their performance,” says Jonathan Barsky, a partner at Market Metrix.

Requiring a minimum number of responses per location allows Market Metrix to do site-by-site analysis, as well as generate scores for the entire chain. Barsky recalls that one of his clients was having less success with a program focused around staff addressing a guest by name when he or she checks in. Site analysis showed one weak location, and further investigation revealed that the staff was not following the program’s principles.

Once crunched, a combination of numbers and charts are put up at a Web location unique to each Market Metrix client. Each hotel chain determines which executives can gain access to particular site pages.

Market Metrix generates easy-to-read charts and graphs with month-over-month comparisons. But certain kinds of raw data are presented: Written comments provided by guests on the surveys are typed into the system and coded by relevancy to department, whether housekeeping, restaurant or billing.

For Larkspur, knowing the comment card feedback was reliable allowed it to incorporate some of the data gathered into its advertising. “We feel comfortable tooting our horn based on the results,” says Bellavance. And in cases where certain numbers are off, such as with the Hilton Garden Inns, Larkspur is considering ad campaigns that would guide guests’ expectations before they check in.