Archives / January 2003

I recently interviewed Elizabeth Peaslee, Travelocity's VP of Customer Experience.

Q: What's your background at Travelocity?

I started working at Sabre in 1995 and moved over to Travelocity in early 1998. When I came to Travelocity, there was no product marketing. The site had been up for two years, and the technology team was deciding what enhancements we'd make to the site. As long as the site was up and fast, that's all that mattered.

My background was in product marketing, so I approached the person who was running product marketing at the time. I said, look, it's great that the tech team is creating these new features, but there isn't a discipline for collecting customer input to help drive the direction the site is going in. We knew that while tech enhancements would drive success for awhile, with faster and cooler features, there would be a point where technology for its own sake wouldn't keep us successful. We'd have to listen to customers and make enhancements based on what they wanted us to do.

Our first effort was to change the way people shopped for hotels. Traditionally, the only way to shop for hotels was via an airport code, which to a customer makes no sense. I'm driving 50 miles from the airport to the hotel, so why do I need the airport code? We thought about how to allow customers to shop based on the hotel's location. That was our first product marketing effort: we took customer input about how they wanted to shop for hotels, and we changed how the shopping tool worked.

The project was very successful. We grew hotel bookings by 25% in one month - radical growth. People said, "Wow, it's amazing how successful you can be by listening to customers and implementing what they want." Then we did that for the air booking and car booking paths.

Overall, my role is to be an advocate for the customer, and to make it insanely easy to buy on Travelocity.

Q: How did Travelocity decide to create the customer experience team, and the VP position?

It was a fairly long process, evolving over the past year or year and a half. Along the way we've used data from our own internal successes, and from successes at other sites, to help sell the idea that focusing on customer experience can have an impact on the bottom line.

For example, we used to only sporadically do user testing. But when we did, we always learned valuable things that made our projects more successful. So I collected data on conversion rate improvements. The hotel story was the earliest we had. The 25% increase in hotel bookings was a huge deal. I also had examples where we hadn't done testing and we hadn't been successful.

I also started collecting information from people who run non-competitive websites. I asked how they did their testing, and started learning how they were implementing a discipline of user testing that could make us more successful over time.

Because of that data, it's been easier to sell the idea of the customer experience team. It's one thing to say, "Here's a great idea, but I have no numbers to prove it," and another to say, "I think it's a good idea, and here's the success we've seen and what other sites have seen." With the numbers, management is much more likely to make an investment in building a team.

We have grown the team incrementally. I'd rather build slowly than walk into a meeting and ask for 400 people. Instead, I'd ask to add one person to do usability testing. Then I'd add an IA [information architect] as a contractor for one project, and see if that cut down on development time. Then later I'd ask if we could have that IA full-time. That's different from selling a full-blown new department, which in a small organization like Travelocity would be a hard sell.

Q: How is the customer experience team organized?

Today, customer experience is comprised of four groups. The day-to-day site production/promotion group does high-volume, quick-turnover promotions on the site - home page stuff. Then there's a market research group, which handles usability tests and third-party research. Third, the site design group focuses on longer-term projects: templates for the site, the style guide, the overall "look" of the site - it works closely with our tech team. Lastly, there's a group that handles IA and site-wide functionality like registration, login, profile, account management, looking up trips you've booked, and other stuff that crosses different business lines. Those are the four main groups.

The IA team is a new role for us. We added it because it was becoming more difficult to write new requirements that take into account all the complexities of the site. The IA role has helped business owners write better requirements that handle all the scenarios, not just the customer going down the correct booking path. All deviations - like if a user clicks on the help link, or if they want to go back and modify something, or any sort of error message - used to get pushed out to developers. That was extending the timeline for projects.

So we brought in people whose expertise is use case scenarios, and site mapping, and that helped shorten the end-to-end development time of projects. It's been very successful, and it makes everyone happy. The tech team gets requirements where all of those other scenarios are already thought through, and they can deliver the project faster. And business owners avoid getting bogged down in figuring out all the different paths a user could take.

Q: Your team handles a variety of issues. Do people "get it" that customer experience is a holistic process, not just tactical usability or IA?

Overall, customer experience is recognized as being important. Though sometimes I worry that "customer experience" will become a horrendously overused buzzword, a catch-all. We have worked very hard to educate the organization on what the customer experience team does.

One reason people "get it" is that everyone here uses the Travelocity site. Everyone can remember a time before we fixed project X: there was something that bugged me, and I'm so glad we fixed that. Because people can personally identify with the frustration of a customer, it's much easier to internalize the value of people focused on customer experience.

As I tell people, our team fixes things that are broken, and then there are things on the site where we wonder, why did we do that? Or maybe it worked a couple of years ago, but not any more, because our users have gotten smarter and more comfortable with the Web. Users now are much more savvy in how travel works. Their needs for functionality have evolved and increased.

Q: What change to the customer experience did you make in 2002 that had the biggest impact?

I'm not sure it's a change to the site, but for us, it's been the implementation of this idea of the customer experience discipline. Whereas before, we'd think about doing user testing, or taking learnings from past projects, if we had the time. Now with this discipline, we WILL test: prior to development, in design, and then after launch. That approach affects all the projects we do, and they've all benefited.

We never want to skip any steps. In the past the business owner may have said, "Let's skip design, I'll just give my Powerpoint to the developer." By getting more disciplined about steps 1 and 2, all the other steps go faster, and what we deliver in the end is so much better.

Q: What are the steps in a project?

A business owner has a concept. We like to get involved early, to help them look at at the scenarios, the impact to navigation, and whether it actually makes it easier and faster to buy the product. We then help deliver things like wireframes and use case scenarios. Who is it geared toward? Why would someone use it?

And then once we have a pretty good idea of how this will meet the needs of customers and the business, we can start design, conduct testing, modify the design based on testing, and then hand it off to our tech team, where it goes through scoping, development, QA, and then it launches.

Q: How does Travelocity monitor the customer experience day-to-day?

We have an enterprise business intelligence group responsible for the dashboard, a set of metrics that the business wants to keep an eye on. They monitor and report the metrics to the organization. There are also some metrics within the customer experience organization that we watch - the error page rate, conversion rate, things like that.

Equally important to the customer experience group are the less frequent and more subjective metrics that we get through user tests. The success rates in those tests show what people are struggling with and succeeding at. We do several user tests every month.

Q: What will you be working on this year at Travelocity?

We'll continue to grow this idea of discipline. Now that we've proven to the organization that there's value in user testing and upfront planning, we'll be doing more of it and getting better at it. The big challenge of the past two years was selling the organization on this whole idea. Now we have to deliver on everything we said we'd deliver by having a focus on customer experience.

The biggest technology story of 2002, in my opinion, was the exponential increase in the number of bits and bitstreams engaged by Net users worldwide. Not just sp-m mail, which got the most attention, but all bitstreams.

In 2003 these bits will continue to increase exponentially, driven by the good experience - that is, the ease-of-use - of their publishing tools.

Therefore, to predict the upcoming year while looking back at the last, Good Experience presents this, the 2002/2003 Review of Bits.

Below are the most important bitstreams that increased in 2002 and will continue to increase in 2003.

Music bits

Music bits have proliferated because of the ease-of-use of new MP3 players, such as the Apple iPod, and "ripping" software, such as Apple's iTunes. It's now easy to centralize music bits onto one device, as opposed to the dozens or hundreds of plastic devices that most people have managed for several years - their CD collections. File-shares like Kazaa still thrive, helping to fuel the popularity of these bits and their players.

Online radio managed to survive 2002. A Congressional effort led by Jesse Helms saved webcasters from the worst of the royalty rates, but Web radio will never be the same as it was in the early, royalty-free years.

Salon on Jesse Helms, Web radio's hero.

Outbound music bits - creating and mixing new music - are still outside the reach of most users. This is due to the lack of ease-of-use in available publishing tools and the lack of users' compositional skills (relative to, say, photography or writing). This isn't likely to change soon.

For 2003: The consolidation of music bits onto single devices will continue, and will drive demand for the acquisition and storage of more music bits (via CD purchases, file-shares like Kazaa, or simply borrowing friends' CDs). The increase in storage capacity on player-devices will drive even more acquisition.

My own experience bears this out: Since I bought a 20-gig iPod a few months ago, I've bought more CDs than I have in years. There's something about the iPod's gaping, available space that makes it seem hungry for more music.

Photo bits

2002 saw a rise in photo bits, but nothing like what's in store for 2003. This year we'll see much wider adoption of mobile phones with cameras attached, allowing any user to take a picture, anywhere, anytime. Once again convenience is the trump card: I carried a cell phone and a small camera (the Canon Elph 100) throughout 2002, taking many more pictures, but most users won't want the hassle of dealing with two separate devices. The camera-phone has great potential, and given a reasonable price, its popularity will hinge entirely on the ease-of-use of its user interface.

An important consideration here is the interface of uploading photos - whether to a PC, a friend's camera-phone, or a "photolog" like fotolog.net (see below). In most cases it's still too hard today for average camera users to upload their photos.

Text bits

Text bits are supremely important on the Web, and getting more so. E-mail, IM, and weblogs - text-based online diaries - fit this category. As with other bits, their popularity is in direct proportion to the ease-of-use of the tools. Given that most users have a desire to type to someone (a friend, a stranger, a weblog reader), the only possible barrier to widespread popularity is a hard-to-use publishing tool. E-mail and IM dropped this barrier; weblogs are beginning to do the same.

Weblogs deserve a special mention because they finally gained mainstream prominence in 2002. The ultimate reason for their spread was the increased ease-of-use of the publishing tools. Many more users, possibly in the millions, will start weblogs in 2003, as major online services and ISPs inevitably roll out consumer-friendly weblog tools.

In terms of market activity, weblogs are quickly becoming today's version of the Geocities home page: lots of press buzz, lots of Net activity, and unknown market value. One thing is for sure: We won't see a repeat of May 1999, when Yahoo bought Geocities for $5,000,000,000 in stock. That's five -b-illion, if you're counting.

Still, weblogs are worth watching, for three reasons:

1. Weblogs are here to stay, not a fad. Like the Geocities-style homepage, people (to varying degrees) will maintain weblogs for years to come.

2. A small subset of the weblogs in existence are high-quality publications, well worth reading. Unlike the bottom 99.999% of all weblogs, the top 100 or so weblogs will command huge readership, press attention, and possibly, eventually, some revenue. Hundreds or thousands of niche-oriented weblogs will also thrive. Less popular weblogs will tend to link to the same top sites, thereby securing the popularity of the elites.

Here's one list of the top weblogs today.

3. This third reason is the most important, the greatest value of weblogs: they're leading to other kinds of advances that will become noteworthy in themselves. Examples are photologs (weblogs made of photos), "moblogs" (weblogs created via mobile device, like a PDA), and my favorite, meta-blogs like Blogdex and Daypop, which are indispensable guides to what webloggers are linking to every day.

So yes, weblogs are important, especially to the top 100 authors and their readers. Past that, count the weblog as an important R&D lab for future Web innovations.

As for other, non-weblog text bits: E-mail and IM had their own news stories in 2002. IM is undergoing standards battles. E-mail continues to be dominated by spam, which of course gained prominence because of the ease-of-use, and low price, of sending e-mail. The industry will continue to hack away at the sp-m problem in 2003. For now, the best solution I know of is to filter and delete it: see my free e-mail management report.

Life bits

Someday we'll be able to record everything around us, all day, and refer back to it later - for family archives, educational uses, obscenely detailed weblogs, and many other uses (imagine it as evidence in the courtroom).

2003 will see the beginnings, in garages and R&D labs, of this "life bitstream" - bits that are captured by devices attached to the user throughout their daily life. These bits are most likely to be video (from a video camera strapped to a backpack, say) or audio (from an attached microphone). They will stream constantly to be collected online, or in a local storage device (a hard drive kept at home).

Microsoft has already begun developing such a technology, under the name of MyLifeBits. Wired News reports that Microsoft wants this to be a "database application that would form part of the MS operating system." Considering the poor usability of Microsoft Access, the company's database product, I doubt I'd entrust the company with the sum of my life's digitized experiences - not to mention the obvious question marks around privacy, security, ownership, and other little details. (Almost NoneOfMyBits are in Microsoft format today, and I see no reason to change.)

Alternatives to Microsoft are already starting. One comes from Steve Mann, who used to lurk in the MIT Media Lab in the mid-90s wearing a bulky cyborg outfit, recording everything he saw. Now at University of Toronto, he and his team have created a line of "conspicuously concealed cameras" - in backpacks, necklaces, even a tank top (you'll understand when you see it).

Of course, the various devices users already carry - mobile phones with cameras attached, PDA's, hiptops, and the rest - are slowly creating a life-bits platform today. I may not capture a constant stream of every experience during my day, but I can choose certain moments to photo, video, or record. Within a few years we'll have the ability to do much more.

. . .

Also worth reading: the Guardian's technology predictions for 2003.