Ms. Frank

Website / Online Mechandising / SEO

Setting Priorities for Hotel Website Improvements

By Tema Frank, President, Web Mystery Shoppers International Inc.

Site Goals Are Key

How you set priorities depends on what your goals are for the site. If your main goal is to get more conference and event bookings you'll focus on different aspects of the site than if you are primarily after foreign tourists, which will again differ from a target market of repeat business travelers.

Website Statistics

One of the most useful ways to get clues about what changes are needed is to look at your website statistics. These are also called "metrics" or "analytics" , and you probably have access to some of this data already through your internet hosting company (or "ISP", in the industry jargon).

At a minimum, your hosting company should be able to tell you where people found your site, how many visitors you are getting, and how many visited each specific page. If there's a page you think is crucial but almost no one is seeing, that's a good indication that either your site structure needs changing, or what you are saying is simply not compelling enough for people to want to read it and continue on your site. Change is needed.

The cost of site analytics packages ranges from free to tens of thousands of dollars, depending on how sophisticated you want to get. At the more expensive end, you can find out what exact paths people are taking through your site, and details such as where their mouse traveled and/or where most users' eyes rested on each page. This will give you further clues about where trouble spots are.

Study Customer Feedback

Next, you want to combine that insight with customer or user feedback to figure out not just where people are getting stuck or turned off, but why. You'll be able to collect some of this user feedback from talking with the people who deal directly with your customers, such as call centre staff and front-desk staff. But unless there are glaring problems and your customers are loyal enough to be willing to tell you about them rather than clicking over to a competitor's site, you still won't know how typical these customer experiences were, nor exactly where on the site they ran into problems.

Even more important, you won't hear from those who were so uninspired by your site that they skipped over it with barely a glance. People who volunteer feedback tend to be at the extremes. If they love something or hate something they'll tell you, but if they are only mildly positive or negative they won't, even though their mild skew will affect how they respond to your sales efforts.

Usability Testing

Formal usability testing can help you figure out why people are responding the way they are, and provide ideas about how to make the necessary fixes. If the testing only involves a few testers, you have to look at all the recommendations that come out of the research and use your judgment about the seriousness of each of the issues identified.

Often you'll get contradictory views expressed, which poses its own decision-making challenges. I was chatting with a friend the other day who was telling me how he loved the feather pillows at a hotel chain I had stayed at recently. I'm allergic to feathers, so I was actually annoyed when I couldn't find any non-feather pillows in the room. How do you know which of us is more typical of your customers? Likewise, looking at your website, some will love the home page image; others will hate it. Should you change it or not?

There are two main ways of approaching this problem: trial and error, or test with a statistically significant sample size. If your analytics or small sample size usability test have pointed to a possible problem page, you can make some changes to the page, put it up on the Web and measure the impact of the change on conversion rates (the percentage of site visitors who call or book online). If it has little impact (or a negative one), you know you still haven't addressed the key issue(s).

With a significant sample size test (statisticians usually insist on a minimum of about 30 people in a sample), you get a much clearer sense of which issues are affecting the majority of your target customers, and how to change your site to meet their concerns. You'll be able to find out if most people liked or disliked the home page image, for example, and act accordingly.

Return on Investment

There's one final, but crucial, issue to consider in setting website priorities: the expected cost/benefit tradeoff of making the changes. Let's say these two recommendations emerge from testing:

Which should be your priority?

First, try to figure out how many of your target customers would be affected by either change. If your target customers are families, # 1 may be the more urgent issue to fix. If yours is a hotel that is mainly aimed at busy, impatient business travelers, implementing #2 may be the higher priority.

Next, look to your research to find out which of these issues is actually causing significant problems, and try to estimate the increase in bookings if the problem were solved. If you discover that a large proportion of users begin to fill out forms but do not complete the process, it is likely that many of them are getting errors, losing their data, and are too annoyed to re-enter their information. If fixing # 2 could get even an extra 10% of those who begin to fill out forms to complete them, you can easily calculate how much that would contribute to your profit margin. How likely is it that changing issue #1 would yield a similar increase in your sales conversion rates? If your target market is families, it might have an even bigger impact than fixing #2. Parents may be so thrilled by the level of detail you now offer on kids and pets that they'll go to the effort of phoning you or trying again if their data gets lost due to an error message.

(Sadly, there are not yet any industry standards that can predict exactly how much of a conversion increase you would get from any particular change. Norms will probably evolve over time, but the Internet is still too young and evolving too quickly for that. We recommend you start tracking the impact of your changes though, so that you'll be able to tap into that knowledge in future site revisions.)

While your marketing folks are making these calculations, have your technical staff estimate how much it will cost to make each change. Put the cost and marketing projections together and, bingo, you'll be able to project where the greatest return on your investment is likely to be. That then becomes the basis for setting your website revision priorities.

Tema Frank is president of Web Mystery Shoppers International Inc. Her company's proprietary website assessment uses an ever-changing panel of some 50,000 people testing web sites from their own computers and providing detailed feedback. Her company has produced research insights that have been valued by companies such as Expedia, Sabre Holdings (Travelocity), Travelweb, OctopusTravel, and many others. She serves on the research committee of the Web Analytics Association and on the Editorial Board of User Experience magazine. Ms. Frank can be contacted at 780-444-5645 or tema@webmysteryshoppers.com Extended Bio...

HotelExecutive.com retains the copyright to the articles published in the Hotel Business Review. Articles cannot be republished without prior written consent by HotelExecutive.com.