An interaction designer books a holiday
Because, you know, it’s 2017 and we’ve nailed that whole “design for the web” thing, right? Right?
I want to book a holiday. But it’s not the first time I find myself yelling at the multiplicity of poor interactions, bad usability, and dark patterns employed on various travel, airline, and accommodation sites. Because, you know, it’s 2017 and we’ve nailed that whole “design for the web” thing, right? Right?
Booking.com
I try Booking.com for hotels. I enter Stockholm and my arrive/leave dates. The results page comes up and I sort the list by distance from the city centre.
The first search result is a hotel that is unavailable for my chosen dates. It takes up around one third of the page on desktop and is the only visible search result. This is not ideal.
I know Booking.com are committed to exploiting the principle of scarcity here, to make me panic and buy something quickly. I know this because as well as telling me this first hotel is unavailable, there are two separate alerts on this page telling me that there is high demand for my chosen dates.
Questions I would ask:
- Should we show unavailable hotels at all?
- If we must show unavailable hotels, could we build in some logic that prevents them appearing first? Do we want the first (and only visible) outcome of a search to be “Here’s something you can’t have”?
- Should unavailable hotels take up as much space on the screen as available hotels?
- At the very least, could we provide a call to action to change dates right where people are reading about that hotel’s unavailability? And not just at the top of the page, where I have not even looked (a) because of banner blindness, and (b) because of course I went straight to the search results?
- Which is more important: bombarding customers with information to panic them into buying something, anything, or creating a long-term sustainable customer base by making a website that is easy to use, and which surfaces available hotels based on the information you have given? A web service that people want to come back to and use again?
Trivago
To be honest, I had forgotten Trivago existed, but I saw an ad for it on TV, and I’m pissed off with Booking.com now, so I think I’ll try that. See, advertising does work.
It’s unusual that Trivago’s date-picker UI defaults to a date early in August (next month), and I have to click back to choose dates for July. But maybe their user research shows that most people book 3 weeks ahead; I don’t know.
I am looking at search results and there are prices next to them. I have questions. Is this price per night, or for the entire stay? Customers shopping for the best deal probably want to compare per-night prices, but I’ve entered dates indicating that I’m staying three nights, so I’m kind of expecting to see the total sum. Also, I don’t know how competitive Stockholm hotel pricing is. I live in Norway, one of the most expensive places in the world, so prices in EU countries always make me a little giddy, like that time in Germany when the bill for three glasses of wine and a beer came to under 20 Euro, and I had to sit down.
Come to think of it, are these prices in Swedish Kronor (destination) or Norwegian Kroner (where I live)? The page just says ‘kr’. This matters less at time of writing, when the NOK is in a bit of a slump and closer to 1:1 against SEK, but still. I need to understand if I can get a better deal here or somewhere else.
By clicking on a hotel name and fighting through a lot of information I don’t want or need as I scroll down that next page, I eventually get the total price for the nights I have chosen. I had to work for it, though.
Questions I would ask:
- Where are customers looking when they are thinking about price?
- Might customers who have entered dates be interested in the total cost, as well as the per-night rate?
- How can we help customers who are not familiar with local exchange rates?
- Where customers are based in a country that uses the same currency name (crowns, dollars, yen, lira, etc) as the destination country, how can we disambiguate?
AirBnB
I have a look at AirBnB. I start out in the mobile app, because it’s super-easy to browse from the sofa. I add a few nice-looking Stockholm apartments to my ‘holiday’ list (something AirBnB invented for me, which I like). I also take the opportunity to update my AirBnB settings to reflect that I now live in Norway, not the UK, and my currency is NOK, not GBP.
Later, with a view to maybe booking something, I check back on desktop. Except … I can’t find my holiday list. Any list. They’re not under For you or Home, which is where all the lists of things live. (Sidebar: what is the difference between Home and For You? Weirdly, the stuff under For You doesn’t feel personalised at all.) I’m genuinely quite confused by how the navigation equates between desktop and mobile: what looks like top-level navigation on desktop only exists within one tab on mobile.
(Two days later, because it’s still bugging me, I finally find my holiday list on desktop, under Trips. On mobile, the Trips tab only shows bookings I have fulfilled in the past; lists live under Saved, which doesn’t exist on desktop at all. Again, information architecture all over the place.)
According to the desktop version of AirBnB, my currency and country settings are all still British. I check the app, which still shows Norwegian, just like I updated it. Also while rooting around in my account settings, I find that AirBnB have default-checked two settings, ironically under the Privacy section, that I would never have said yes to: share all activity with Facebook friends who are also on AirBnB, and include my profile in search results from Google, Bing, etc. I’m not on Facebook, but if I were, I would not choose to spam my friends; and I’m not a host, so have no need to appear in search results. Plus, hi, informed consent. Dear AirBnB: please educate yourselves about the General Data Protection Regulation legislation that takes effect from 2018.
Questions I would ask:
- What evidence exists to support different navigation and information architecture for desktop and mobile?
- How does our development team structure support a single, simple, joined-up user experience across all devices?
- What evidence exists to support ‘Home’ and ‘For you’ as separate navigation items, and what data is used to inform the content that lives in each?
- Might it be worth the development pain to ensure that data entered by users on desktop and mobile updates the same ‘single source of truth’ database?
- How could we build trust with users (and thus secure a faithful long-term customer base) by ensuring we get informed consent to share their information?
So what’s the problem? It’s not like these services are actually unusable, after all. Do you just like being angry or what?
Well, there’s usable and there’s usable. Could I complete a transaction on these services? Absolutely. Could everyone? Probably not; I’ve worked on digital services with broad customer bases. Some people don’t know things you think are basic, like the difference between main page content and footer content. I didn’t even talk about accessibility here — simple, basic things, like making sure focus is visible when navigating the site using only a keyboard, as some people need to do.
My point is, the more usable by everyone these services are, the more long-term customers they will gain. Show me a company that doesn’t want to sustain its existing customer base while also exhibiting long-term growth. I dare you.
Also, I’m not actually angry. I am quite frustrated, though. We can do better.
So how do we do better?
We educate management.
Per the agile manifesto, and per usual, I’m going to assume that everyone who made these things did the best they could within the existing constraints and with the information they had at the time. But what I can never assume is that:
(a) there was a managerial commitment to conduct early user research in order to identify user needs, and to assess continuously through ongoing testing how well the proposed design solutions actually worked for real people;
(b) there was a managerial commitment to prioritise consistent and honest user experience, in order to grow a long-term-sustainable customer base;
(c) there was a managerial commitment to release new features and functionality sensibly and continuously, and not to some arbitrary big-bang deadline, set in ignorance of what well-researched design looks like.
All of the above depend on management being educated about how to build digital products and services sensibly and sustainably. If you don’t have that, you can’t do this stuff well. Although at least if you understand that you don’t have that kind of expertise, you can choose to hire it in.
We already talk about technical debt, but perhaps we should start talking more about user experience debt.
I get that as you grow a product (and a customer base), stuff gets shoehorned in ad-hoc, until eventually you end up with something like The Winchester Mystery House, desperately in need of interface and information-architecture refactoring. Product and service owners are mostly reluctant to start over, maybe due in part to the sunk-cost fallacy. But also, understandably, nobody wants existing users to freak out if everything changes.
Maybe though, maybe it’s possible to refactor the UI without giving customers a scorched-earth user experience. There’s a fine line between nurturing an existing customer base and being too scared to execute necessary change. It’s definitely worth exploring; we interaction designers love a challenge. And if we’re occasionally cranky, it’s only because we care about good user experience. We hope we can persuade you to care about it too.
