“What were you thinking?“

Balancing fluency and feedback in UX copywriting

Clare Wilson
Booking.com — UX Writing
6 min readOct 5, 2018

--

I did a quick straw poll. On two of the teams I work with, we speak 15 languages between us — and everyday we come to the office and work in English. We get coffee, give training and presentations, plan sprints, develop experiments, chat over lunch, tell jokes — and give each other feedback.

Feedback is a key part of the Booking.com culture, but every so often, my manager will ping me a screenshot and ask what whoever wrote the copy was thinking. Then I find myself in an awkward position. My copywriting colleagues are experts in their areas of the business, but…maybe they’ve missed something? Maybe that light-hearted message isn’t so breezy after all — does it sound a bit sarcastic? If it’s just an opinion from one guy, should I pass the feedback on — and if so, how?

Hello, overthinking.

And so this post was born. Good UX copy should be easy to understand, so what should we do when not everyone gets it?

“A focused man working on a sticker-covered laptop in a coffee shop” by Tim Gouw on Unsplash

Comprehending the issue

Everything we write at Booking.com gets localised into 43 languages and dialects. We’re used to trying to anticipate translation head-scratchers when writing UX copy. But trying to gauge understanding of English, in English, at the same time? That’s harder, but not impossible.

First of all, let’s size this problem. You’ll know your own audience best; for us, a fairly substantial proportion of visitors to Booking.com browse in English. We can get a bit more granular with analytics*, but obviously, it’s impossible to know exactly what level of fluency our audiences have.

* … I haven’t, because we’re here for words not stats.

It’s important to remember that there are all sorts of factors which can affect comprehension — for example, cognitive impairments like autism, dyslexia, or ageing. Language isn’t always the reason that copy might be misunderstood.

What to do if people say “…I don’t get it?”

Whether it’s misplaced humour, or misunderstood instructions, you can end up taking comments like ‘I don’t get it,’ ‘this copy isn’t clear,’ or ‘this is misleading’ personally. Everyone’s a critic. But those moments are an opportunity. Remember that you’re testing your words, not yourself.

Let’s go back to first principles. Literally. Do you have Copy Principles? We have five, which are used as the basis of our Tone of Voice guide. They’re a shorthand guide helping stay focused on the right things, helpful for fending off rogue opinions and great for backing up an innovative (perhaps controversial) new idea.

At Booking.com, A/B tests are one of the (many) tools we have to understand our users. But you don’t get a sense of how people interact with your words using just quantitative data. Did they chuckle? Did they click with a sense of certainty? Did they even read what you wrote? Maybe, maybe not.

Get people in a room if you’re looking for qualitative insight into your copy. Try running some user testing for unbiased feedback, or grab friendly colleagues for quick opinions.

If technical/linguistic copywriter-to-copywriter feedback is what you’re after, we have structured ‘CopyLabs’. One person presents a copy-problem and we challenge, critique and help develop the idea that’s being presented linguistically. Opinions get left outside. It’s a great way to get new ideas, as well as thoughts on your current direction.

Lots of brains are better than one. “Several people fist bumping over a busy workspace” by rawpixel on Unsplash

Takeaway:

Know your audience. Who is reading what you’re writing? What do they get stuck on? Does your product need more instructions — or is it more effective to let your customer decide what they need?

Can you avoid issues before they affect real customers? User testing shaped the ‘voice’ for Booking.com’s Booking Assistant chatbot, as Kathy explains in her post about having greater impact with a less-chatty-more-helpful chatbot; “…there’s a time and place for entertaining, engaging speech bubbles, and it definitely isn’t when someone is trying to get something done.”

Finally, why are you writing what you’re writing? Start with what the user wants to do — which is not the same thing as what you want them to do. Sounds so basic, but it’s oh-so important.

How about if people say “…[insert description] — is it really?”

Idioms are beautifully expressive, but they’re one of the hardest things to learn in any language. As they’re very localised, they can be great to give local audiences a sense that you’re speaking directly to them.

Trying out localised version of the the same idiom — “More travel deals than you can shake a stick at” for UK audiences became “More travel deals than fish in the sea” in Portuguese

Use them with caution. This example is a fun use of idioms, but in other contexts expectation management can be problematic. Trust comes into this too. Are you selling something at its real worth, or do people perceive you’re using fancy phrases as camouflage? Are you saying more than you realise?

Takeaway:

Prioritise sharing real insights or useful info over wordplay. As an accessibility tip, run your copy through a readability tool like Hemingway (free) or Readable.io (free trial then paid) every so often.

What about if the feedback is “…that’s a bit threatening?”

Urgency is a controversial topic. There are (rightly) strong feelings about the ethics of persuasion and urgency messages in the UX community. There are legal implications for getting it wrong, and factoring in cultural sensitivities adds another layer to this discussion.

Can you imagine being on a team made up of Iranian, Portuguese, Greek, Russian, Polish, Romanian, Austrian, Dutch and English people? That’s me every day. It’s great. But it made choosing an urgency example tricky because everyone had different opinions about what is useful and what is too much.

Does urgency messaging work, or rather, does it encourage users to take the action you intended? Yes, or it wouldn’t be so widespread. Does keeping it ‘light’ in persuasion messaging translate well to different cultures? Not always.

There was a lot of discussion about the following example — both when it ran as an experiment and while writing this post. This pop-up was shown to users when they went back to a hotel page, after they started to make a reservation:

“Double-checking? Smart — but remember that our availability at this property won’t last forever.”

The intention was to address a real user problem — we can’t ‘hold’ a booking, so there’s a risk of a user losing the place they intend to book if they delay too long.

I’d like to draw your attention to the first line:

“Double-checking? Smart — but remember that our availability at this property won’t last forever.”

What might read as a casual acknowledgment of back-and-forth behaviour, can also be perceived as heavy-handed, unhelpful or even disrespectful.

It’s not always an easy one to navigate, but data is your friend in this case. Start with your customer. When a user’s behaviour around your site tells you they’re already questioning something, think about what they need from you — and solve that problem first.

Takeaway:

Be empathetic, think long term and write responsibly. Urgency messaging can be much harder to tone down or remove than it is to add, so interrogate the ‘why’. If you’re using a persuasive technique, consider it in context of what your user is doing.

Getting and giving feedback

Feedback is an integral part of Booking.com and it’s an important part of developing strong, impactful UX copy. It can be tough to get — and tough to give, especially when it’s sharing thoughts on what people could do better.

Feedback is most constructive when it’s actionable. ‘I don’t like this,’ doesn’t count. To ensure feedback is as constructive as possible, we use the BIO model: observed behaviour / its impact / the opportunity for the receiver of feedback. Not only does this keep feedback focused, it also helps reduce the chances of taking feedback personally.

One last takeaway:

Ask questions. Questions structure conversation. If I really want to know what the rationale behind a piece of copy is, it gives whoever I’m asking a starting point for their reply. And if I’m on the receiving end, it helps me find out if I’ve really missed something.

Moving forward, there’s one more thing to do — ask yourself what can you do with the feedback you now have? Hopefully, something great.

We’re always on the hunt for new writing talent, wanna join us? Apply here.

p.s. Extra thanks to the copywriters who were (and weren’t) happy for me to use their work in this post. Feedback forever ❤

--

--