Booking vs. Agoda: What makes a better hotel search experience?
Intro
Last year was a disaster for the travel industry. However as many countries’ vaccination is progressing well and economies gradually re-opened recently, the tourism sector is expected to resume gradually.
Taking China as an example, the country has taken successfully controlled the spread of the virus since mid-2020 and the tourism industry (across Mainland China) has soon recovered to its original level in late-2020.
According to a report from McKinsey & Company published in October 2020:
People (in China) of all ages are willing to travel again. About half of the respondents expected to take their next leisure trip around the National Day holiday at the start of October.
I believe the online traveling business is expected to resuscitate soon especially domestic trips in countries that reclaim control over the epidemic. As a product designer based in Singapore, I conducted observations over the top two regional players Booking.com and Agoda.com, focusing especially on their search experience. From my point of view, I think Booking could polish its search experience to a better extend.
My story with Booking.com
The key to a high conversion of hotel booking agencies is to provide appropriate properties to users either from suitable recommendations or accurate filtering (Why is that so? I’ll explain later).
In my personal experience as a traveler, the default list always includes accommodations beyond my budget so I’d always go for the filters to explore efficiently.
The filters are not available on the first screen, so I scrolled down a bit further and retrieved it. This is a classical layout, where filter on the left and content on the right so I got no surprise. I got a budget filter immediately, which is nicely designed with different levels of budget and also an optional scroll bar indicating the distributions of properties at various price levels.
I noticed that below the Budget filter is the Popular filters, a collection of popular choices based on Booking’s user data.
Taking my last trip to Tokyo as an example, I did a quick test on the relevance of filters in this popular filter section:
- Breakfast? 🙅Nope, I am fine to dine out!
- Free cancellation? 🙋♂️ Yes!
- Double bed? 🙅 Nope.
- Hot tub/jacuzzi? 🙅 Nope.
- Less than 1km from my search location? 🙋♂️ Yes!
- Parking? 🙅 Nope
- Hotels? 🤔 I am open to it but I prefer hostels
- No prepayment? 🤔 Not sure about this better to leave it alone
As a result, much reading and thoughts were spent but only 2/8 options were ticked. I checked the properties again hoping I could get what I want, but I am still curious about what else in the filter action I can play around with.
Then I found the filters are like 5-page long! 😱
How many options in total there? I don’t know as I lost count. All of them are formatted in the same way — a title, with repeating checkbox/text/number combinations. I have to read before knowing what they are about actually.
With an experience of the low hit rate of popular filters, I quit reading at them in the middle. Instead, I hit my keyboard and called the browser search to look for properties like customer rating (I’ll go with 8 and above) and property type (Hostels is my top choice).
Browser search doesn’t always work as I expected — especially when I don’t get the right keyword or there are too many matches. This is what happened to me:
So, what’s the problem with Booking’s searching experience?
The structure of Booking’s property list page is like this:
Though we are not part of Booking’s user research team, we could still deduct the major user path on Booking’s property list, from its page layout:
As we are talking about the browsing & filtering experience to a user, let’s stand in a user’s shoes and brainstorm what they are like when they stay on the property list, adjusting the filter options.
- They may be very clear about what criteria to apply or they still await inspiration
- For certain criteria, they may be very easy or strict
- They may want to get a particular result list to save time or as many options as possible to choose from
- …
I don’t think that Booking has offered the best experience to cater to such needs. From the page layout and filter design, some usability problems could be spotted:
- The long repeating filter sections with the same text format could be overwhelming to recognize
- The information retrieving process solely relies on scanning and reading, which might be counter-effective
- Some expressions may not match users’ mental model
Now let’s take a look at how Agoda does its job.
On Agoda’s first screen of result list:
- The search entry retained a horizontal layout (as it was on the homepage) and stays on top of the screen
- Most filters are curated into button-like entries and reside below the search entry
- Map view entry resides on the top-left corner of page content, followed by a location rating filter.
Why filters are important to both sites?
Before diving into the details, I’d like to stress the importance of filters, or facet searches, on both sites. There are quite many similarities shared by both sites strategy-wise:
- Both sites are leading regional market players targeting a wide audience with abundant properties
- The information in a property in the list is both detail-oriented to entertain various facets of interest from the audiences, allowing only 2~3 items to be viewable within one screen.
As a result, narrowing down the scope of properties makes more sense for the sites to deliver precise options to a user based on their preference (via filters) or profile (via recommendation).
Agoda’s approach
Easy-to-spot filters
Unlike Booking, Agoda placed filters in the heading area (also sticky as page scrolling down) to ensure it’s noticeable and accessible, no matter what browsing behavior the user poses.
Curated filter groups with a concise look
Compared with the 5-page long filter options from Booking, Agoda curated their filters into four button-like groups:
After playing around with these filter groups, I feel it solved some notable problems with these design tricks:
- Enclosed content in four entries to reduce cognitive load
- Utilized iconic imageries to reduce cognitive load
- Extended content layout horizontally to make fast scanning possible
Let’s take a closer look:
Agoda’s also evolving
Actually, what we see today is not the first version of Agoda’s filter design. This is a screenshot I captured from one year ago.
Comparing with the current design, Agoda used to have a Popular filter group just like Booking does. The disappearance of it might cohere to my conclusions from Booking — Popular filters is highly likely to be under-utilized as it’s might not be so relevant to each individual.
Location, instead of Payment options, resides as the third filter. This decision id made probably due to the raised concern of booking flexibility during the pandemic.
By removing the Popular & Location filter, Agoda kept only four entries upfront. Quite some neuroscience studies have backed up for it:
Working memory capacity varies in the general population. The average capacity of working memory was shown to be more like four plus or minus one — that is, three to five items (Broadbent, 1975; Mastin, 2010)
Less is more.
Synergy another search in search results
Another innovative approach Agoda adopts is the search box embedded within the filters.
Though still working with some hiccups, the search box is intended to provide one more viability for users to access the intended facet of property directly.
Summary
I am writing this case study not to condemn Booking but just to share my understanding of the interaction design in the context of the hotel search experience on online traveling agencies (OTAs). Though Booking has accounted for the largest market share worldwide, Agoda demonstrated a smarter way of organizing the search & browsing experience with their filter design, by
- Enclosing filter options in only four entries
- Utilizing iconic imageries with text
- Customizing content layout within the horizontal display
“Satisficing infovore” & our limited working memories
The searching experience is like predators hunting — or, we as internet users could call ourselves as “infovore”. The strategy normally taken is satisficing — meaning that we don’t end up with the optimal result as we exhaust all options, but aims for a satisfactory or adequate result with rather less effort.
Searching hotels on a website may not happen so frequently for most of us, thus when we interact with the site we mainly use our working memory (or short-term member), which is also quite limited.
In my understanding, a better searching experience should respect these facts and facilitate those infovores to reach their goals with the minimum cognitive burden. And from that, I believe there is still plenty of space for Booking to explore and improve (before the tourism recovery 😉).
What’s more
As a product designer, being an armchair strategist is not my way 🕶️. I am currently working on a redesign project to Booking with the insights from the case study, so keep tuned by subscribing to my Medium column if you are interested.
References
- “China’s travel sector is undergoing a nonlinear recovery: What should companies do?”, McKinsey & Company, March 23, 2021
- “What can other countries learn from China’s travel recovery path?”, McKinsey & Company, October 15, 2020
- Jeff Johnson. Recognition is easy; Recall is hard. In: Designing with the mind in mind
- Broadbent DE. The magical number seven after fifteen years. In: Kennedy A, Wilkes A, eds. Studies in long-term memory
- Mastin L. Short-term (working) memory. The human memory: What it is, how it works, and how it can go wrong.
- “The word: Infovore”, New Scientist, Issue 2561, July 22, 2006
What’s in your mind?
💭 Comment or drop me a personal note and let me know your thoughts, doubts, suggestions!
👋 Connect with me on LinkedIn
