Scott Moorhead is Executive General Manager – User Experience and Innovation at Australian company Wotif Group. Wotif are the leader in their market at selling hotel accommodation online, but the reason I conducted an interview with Scott was due to one of their newer initiatives, a very interesting OTA called Wotflight.
Martin: Scott, Thanks for making the time for an interview and offering to share some of your experience with my readers. I’ve had a few different people in recent months telling me to take a look at Wotflight and suggesting that the user experience (UX) was quite innovative. But before getting into some of the details of how and why you have set up the site the way you have, can you please give me a bit of background on Wotflight, the parent company Wotif, your role at the company, and I suppose an idea of some of the thinking behind moving away from just hotels and into selling flights.
Scott: Thanks Martin. Glad to let you know some more about our Wotif.com and Wotflight. Wotif.com is Australia’s largest online accommodation provider and has been so for more than 10 years now. We offer accommodation in over 55 countries and our customer base is continually growing worldwide with nearly 4 million monthly visits to the site.
I head up the User Experience and Innovation team for the Wotif Group and our team is tasked with ensuring that our evolution of existing services and adoption of new ones retains a focus on usability. One of the things that sets Wotif apart from the many other OTAs is the fairly unique grid-based display that enables users to quickly compare prices between many properties across a wide time period. That display has been emulated many times now amongst other Australian travel sites though is still fairly novel elsewhere. It is a clear differentiator for us and one of the things that customers mention as a factor that keeps them coming back.
Wotif.com remains an accommodation specialist, but the group as a whole is expanding beyond property through other domains and Wotflight.com is one of them. Carrying the ‘Wot’ name, we knew customers would expect Wotflight to duplicate that core element of Wotif – a unique display approach that aided comparison rather than favoured advertisers or any individual provider – something that offered a visible advantage over other similar services.
Australia actually has two of the world’s top 10 busiest flight routes: the Sydney to Melbourne route offers close to a thousand flights a week. When you include the multiple classes of ticket, there can be hundreds of flight options on any given day. It’s also a big country and some routes present options whose total flying time can vary by several hours.
For most users, even business users, choosing a flight requires trying to balance price, departure/arrival times and length of flight. Most flight sites tend to heavily favour only one of those factors by listing flights in order of one of those attributes. The result is very unbalanced, even if the user can change which attribute drives the sort-order. And none of them are ‘glanceable’: the majority of those hundreds or thousands of data points are usually presented as text. That’s a lot of reading even if much of it is peripheral – hundreds of prices, times, flight lengths, stops – all presented as text.
So the goal was a search result page that could show 150 flights for a single day; that didn’t require the user to scroll their browser window; and that could impart at a glance the relative departure and arrival times, length of flight, number of stops, waiting time, and relative cost without the user needing to read any more than a few data points on the whole page.
Martin: Scott, I take your point about the Wotif hotel display being unique and it has certainly had a big impact on training consumers to search differently for hotels than they do in other markets. I did read a quote from your CEO Robbie Cooke at the NoVacancy conference earlier this year saying 60% of the hotel business was direct to the site and under 10% was from paid-for traffic, so hoping you could share some insight on where the Wotflight traffic is coming from. And following on from this, how you see the service fee model in Australia given that major OTAs elsewhere have dropped air booking fees aiming to make up for it in hotel commissions. Maybe the AUD$20.00 hotel voucher you are currently offering is kind of from the same school of thought, but I’m not sure if that is a promotion or something more long term? Either way, it sounds like a reasonable idea to build up some decent early volumes on Wotflight and get some customer familiarity and loyalty for the unique UX – I’d be interested to hear your view on how customers have taken to the on-screen display and if you’ve had to tweak anything since launch based on feedback or user testing?
Scott: Much of the success of Wotif.com was built on the word-of-mouth of happy customers. In fact, there was no traditional advertising for Wotif for several years and we’ve never been a big advertiser.
With Wotflight, we’ve also taken something of a quiet approach to the launch – really only exposing it to existing Wotif.com customers and natural SEO traffic. We weren’t reliant on Wotflight making a big initial impact and we already had waiting customers and a channel for reaching them through Wotif.com. So a soft-launch has given us a chance to trickle-in the users while we monitor and refine the system and prepare it for some bigger changes to come based on our own plans and the feedback of users.
The display has had some mixed feedback, though the vast majority has been positive. Straying as far as we are from the flight-display conventions, I would have been dumbfounded if we didn’t get any negative feedback. But the few negative comments we’ve had were largely anticipated, so haven’t changed the way we think about the service – they’ve really only been about the fact that “it’s different”.
That difference is also what’s driven the majority of the positive feedback we’ve received: people telling us that they can find and book suitable flights in “half the time” they experience elsewhere, or that the display makes it much easier to pick out the short flights from the long ones.
We’ve had lots of constructive feedback too, in pre-launch user testing and post-launch from real customers. That feedback has initiated some design tweaks and changes. We’ve changed the way we display flights that cross time zones, turned some two-step processes into single-step ones and now offer two views of results – one with flights sorted by price and one with them sorted by carrier (cheapest carrier first, and flights by price within each carrier group).
In regards to service fees, lots of people will happily pay a few dollars if their time spent searching for a flight is cut in half – just as they would be if you could cut their time grocery shopping, washing dishes, or driving to work in half. That said, our service fees are already lower than the major competitors in this market and we are offsetting those with Wotif hotel vouchers. You are right in suggesting that the voucher offer is helping us build up early traffic volumes, familiarity and loyalty, but the voucher deal isn’t a temporary promotion, it is an ongoing incentive for customers to stay within our network.
Martin: I’d like to focus a little on your comment about Wotflight straying from the flight-display conventions, as that point really gets to the core of this interview. Your flight duration / timeline screen (availability page) reminds me a little of the ITA Matrix display, but you’ve put a nice skin over the top and also made a good effort to display the fare families and associated price points for the same flight. I did notice on an Adelaide to Coolangatta search that you don’t show mixed Qantas – Jetstar itineraries, but that would primarily be a strong leisure route and maybe you are more focused on business travel. I also came to this conclusion about business travel when I noticed you are only showing fixed date search, whereas the trend in recent years has been to expand the date range of the calendar to show lowest price over a wider range of dates to assist more flexible travelers. Am I correct in assuming the site is primarily aimed at unmanaged or lightly managed corporate travel, or are there other novel features that are attracting a reasonable number of leisure travelers to the site? And one more thing before I hand back to you for an answer – which websites do you look to from within the travel industry or maybe even non-travel retail websites and think that they are doing a good job on innovation when it comes to user experience?
Scott: I hadn’t seen the ITA graphical display before now. I can see similarities in that both have an underlying approach that resembles a Gantt chart, but there’s strong differences too. You’ve spotted a couple of the things that make Wotflight unique; the skin that provides some breathing room between the data while highlighting the difference between elements on the page, and the display of multiple fare families or classes, rather than a single “from” price for each flight.
The Wotflight display provides some anchoring cues to orient the user so that they don’t need to keep referring to a legend or scrolling up and down the page: Airline logos, highlighting of rows on mouseover, and animation that gradually removes parts of the page when the user makes a selection.
One of Wotflight’s real strengths though is in clustering the display of flights so that several flights sharing the same carrier and price attributes can be shown on the one horizontal line, and that multiple horizontal lines can share a single label for the carrier and prices. This conserves space and can dramatically reduce the amount of text data that would otherwise need to be displayed on the page. I’m looking at a page of results for a Sydney to Melbourne weekday flight and Qantas is offering, amongst others, 17 flights that all share the same carrier, departure & arrival cities, 5 fare classes and their associated prices. With Wotflight , for 17 flights, that’s 8 types of data displayed as text or labels just once – 8 things to read rather than approximately 135 labels that would arise without that clustering.
With the 70 flights in total on the page, Wotflight uses around 73 text labels and logos (not including the timeline), and only around 61 when the results are sorted by carrier – in stark contrast to around 500 that would be required without clustering. In the more typical table or list display there’s at least 3 more data points per flight (departure, arrival and duration). Assuming an average of 4 classes available per flight, a rough estimate for a page of 70 flights puts the number of displayed pieces of text at around 900 in the typical flights table versus only 61 for Wotflight (15 times less labels for the same amount of data). That difference makes the page much easier to visually scan by allowing the user to easily consider or reject whole groups of flights at a glance.
The particular Adelaide/Coolangatta Qantas/Jetstar flight combination you mention isn’t available through any OTA that I am aware of. Qantas itself offers the combination through their own site (being Jetstar’s parent company). Wotflight offers many other alternatives for the route to suit both business and leisure customers.
We haven’t taken a particular preference for business travelers over leisure travelers, and in fact always set out to design a site that worked equally well for both. This is another reason why the timeline display works so well. It allows a leisure customer to easily focus on budget by looking at prices down the left of the page, and business customers needing to get to a meeting on time to focus on times by scanning down from a certain place on the timeline. One feature that does work well for business travelers, or, more precisely, for individual travel arrangers within an organization, is the voucher deal. The accommodation vouchers that come with every Wotflight booking don’t have to go to the traveler, and some savvy travel arrangers out there booking flights for several employees are collecting a number of accommodation vouchers that will no doubt go towards a great personal holiday. That’s fine by us, as corporate travel arrangers are obviously a good group of people to have on-side.
As for showing results for a wider date range: that is something we do want to incorporate – along with a bunch of other features. From a UI point of view the aim is to make sure that the value of any additional feature outweighs its contribution to overall complexity.
There’s no single travel site that I see as being a holistically good example of innovation in user experience – though many have individual features that I appreciate (adioso.com has an interesting natural-language search box for instance that makes their home page very simple). There’s a lot of showy innovation on some travel sites, though it doesn’t necessarily make the experience any better for end users.
I tend to pick-and-choose my user experience inspiration from a wide variety of sources that include retail sites, books and sites about interface design & customer experience generally, and from non-web and non-electronic sources including everything from games and hardware design to signage, printed tickets and charts. I think that for too long the web world never looked much further than software design for inspiration, and the more radical ideas were often crippled by technical constraints. There’s always going to be constraints, but now there’s no reason why a site needs to look or work like 1990’s software or static print design, and we can make better use of the knowledge of so many other design fields, techniques and tools.
Martin: I’ll be interested to see the calendar when you launch it, as I assume based on what you’ve done so far with Wotflight that there will be some twist to try and differentiate it from current calendar displays. I’m also curious to know if you’ve looked at adding international routes, as I suspect journey based taxes would add a whole new level of challenges in the browser if you were maintaining one way display pricing as you’ve done domestically. But the real question here, and it kind of follows on from the calendar discussion, is how you view the current hot topic in travel search of how to inspire and direct people who are still very open minded about when and where to travel – the undecideds as I call them. Wotif have an extremely strong hotel offering, you are now selling flights, and in December 2009 you purchased destination content aggregator GoDo, so in my mind you’ve got many of the pieces of the puzzle under the one roof and I wouldn’t be surprised to see you popping up in this space at some point in the future. I’m not asking you to reveal future plans, unless of course you want to, but could you just give a more general comment on your view about travel inspiration and how you see this evolving as more travel technology companies, OTAs and supplier websites move further into this space.
Scott: We have plans to push the Wotflight offering into the international arena – how we do that and when is something that will be revealed soon (sorry, I can not be more specific). International flights though do present some new UI opportunities in terms of simplifying the display of concepts that don’t play as strong a part in the domestic space.
The opportunity to tap into a very qualified flights and experiences audience (via our GoDo business) is something we are very attracted to. There too though, I can’t reveal more at this point. What I can say is that new users and new technologies will ensure that there’s always room for UI and user experience improvement, and that makes a team like mine tick.
Martin: Thank you for giving such detailed answers on the UI aspects of Wotflight, and I understand why you can’t reveal too much about the future plans of Wotif at this stage. If what you end up doing in the travel inspiration space is anything like the innovation you have brought to Wotflight, I’m sure it will be worth looking at, just as I’m sure the readers here will be keen to take a look at how you implement a calendar display in future.