Published: 25 August 2019
Gerry Gaffney
This is Gerry Gaffney with User Experience podcast.
My guest today is a UX content strategy consultant and UX writer at Google. She's had a splendidly varied career that includes freelance fiction writing, rocketry, teaching high school and physical product design.
She has a Bachelor's degree in physics from the University of Washington and a Master's in curriculum and instruction from Seattle University. Her recent book through O'Reilly media is Strategic Writing for UX: Drive Engagement, Conversion and Retention with Every Word.
Torrey Podmajersky, welcome to the user experience podcast.
Torrey Podmajersky
Thanks so much for having me, Gerry.Gerry
Recently, I was eavesdropping on a conversation at a client's office and somebody was expressing doubts about the very existence of UX writing as a thing. Now maybe you can begin by telling us a little bit about what UX writing is and how it differs from, say, UX design on the one hand, our technical communication or tech writing on the other.
Torrey
Yeah, I'd be happy to. So UX writing does get some doubt in the industry still where people haven't yet seen us around. But I want to assure people that I have been doing this professionally for the past nine years. That has been my full-time job. The difference between UX writing and UX design, this is where there is actually very little difference in terms of the day to day activity, the core activity of imagining where are users, where are the people who you need to use our experiences are coming from, what they need, what they want out of the experience, and how we might delight them or engage them or move them forward to what they need. And we do that using the language in the experience. We use it not only at the information architecture level but the individual words on the page, much like an interaction designer might use or devise controls to use on an interface or a visual designer might choose a scheme of colours that would work together and embody the brand.
UX writers use language to make sure the utility is available and understandable and also that the brand comes through of whatever that experience is supposed to be. Now it is… Many, many UX writers came from the technical writing world. And the main difference there, I think there's two big differences. One is brevity of individual phrases and sets of language that's actually being written. And another part is that technical writers generally have two different audiences. One is either the sets of end users who will be using and troubleshooting and want to learn more about what's going on. And also the developers, either future developers in the same organization or third party developers who need to understand the API, need to understand that the underlying technology and make use of it. So UX writing is sort of different from both of those and very related.
Gerry
Now you mentioned, you talked briefly about brand and I guess one of the things that we think about when we think of brand is voice. You wrote in the book that "informal descriptions of voice are only as strong as the consistent understanding of that voice." How do you create that consistency or promote that consistency within the language used in the UX?
Torrey
Yeah, it is very, very important to have all of the different pieces of content that come in an experience sound like it's coming from the same experience, right? Like it shouldn't, your error messages shouldn't sound like they were written by a different company or a different organization than the people who wrote the splash screen, and the very sexy copy when you first land, although it often does sound extremely different. Making that consistent is about telling the story internal to the organization of how do, like, how do we want people to feel, what is important about the language? What are the important ideas to communicate? How do we want to communicate those in language and grammar and even the number of words. And what I've done in the book is provide a way that's been helpful to me at other companies that I actually break down a chart and the chart can have any number of columns. And the six rows that are important are the concepts, the verbosity, the grammar, the vocabulary, the, oh, I don't have it in front of me… The capitalisation,
Gerry
I don't have it in front of me either, unfortunately. [Laughter.]
Torrey
So what's important is that the every column is, starts out with what is the principle, what are the various principles of this product that need to be spoken. And sometimes it is like the one product of that organization. And sometimes it is an organization that has many, many products and the principles might be… So in the book I, I use three example apps or experiences to show that range of voice and the, and how those voices apply differently to different text patterns. And one that I use is actually kind of my favourite is a transit system app. So the transit system app needs to be usable by, say, anyone in that transit area. And so the product principles for that transit app are things like being fast and efficient and being on time and saving and being green essentially.
So the voice that goes along with that means talking about concepts surrounding that, using vocabulary that supports that. And for the transit app, you want to use a simple grammar so that even if people are not good readers of things, that they'll be able to understand what the different labels and messages mean. So all of that can be defined in a simple document that then can be used by the company to make decisions with. Like, it's not something that every writer would use every day to be like, "Well, let's see what the chart says about writing this notification." But it's about how do we differentiate and make sure that our voice is landing in all of our text the best way we can. And when we have great opportunities different great options, how do we choose between them?
Gerry
It struck me as being a fantastic tool. I've never used it and I hadn't come across it before. What's the origin of the voice chart? Where does the idea come from? Do you know?
Torrey
The idea came from… So I had, in some ways I came up with it, but I mean, using a table is been around for a long time. It's this idea of what are the different criteria, orwhat are the different ideas we're trying to get across. And that was the product principles. And then how are the, what are the different things I can change about the text? Well, I can change what I talk about how I talk about it. And in that "how" I've got choosing the different words, choosing how many of those words I use, how those words are capitalized and punctuated, you know, those are the things that I can really change to dial in that language.
Gerry
I guess a lot of people would see that as being the role of the style guide or perhaps a controlled vocabulary or word list.
Torrey
Yeah, absolutely. Both of those things are very important, but go beyond the voice chart. So, a style guide will tell you if you have decided to consistently use the word cancelled with one L or two Ls, both of which are totally legitimate in US English at least. A style guide will also tell you where to put the comma. Is it inside of a pair of quotation marks or outside? All of those are great style considerations that you wouldn't want to be sort of sloppy and inconsistent but don't tend to have a big effect for most people to pick up on the voice.
Torrey
Yeah, I guess the voice charter would be more likely to be read than the style guide also. Style guides can be a bit tedious.
Torrey
They, they can, this is, so the style guide is also designed to fit on literally one page.
Gerry
The voice chart.
Torrey
I'm sorry. Yes. The voice chart. I've never seen the style guide on one page.
Gerry
And, and you've got three specific purposes that you say that the voice chart supports. What are they?
Torrey
So there's really, so there's the, the very first one is to design it in the first place. Like, having to make those decisions and put them down on paper or virtual paper so that it is being discussed. Like what is the effect of the capitalization and punctuation on how our brand is perceived, what is the effect of how many words we choose to use, whether hyper brief or wordy on how our brand is received. So doing that design work in the first place, it gives that design work a structure and an outcome and an artefact. And then once you have that artefact and you've had those conversations and made those design decisions, it helps you train additional people. Like when a new person comes on board, whether they're a full-time UX writer or they're just the designer or engineer or product owner or marketing person who needs to pick up and write a new UX text. How do you get them on board? How do you communicate what the voice should be to this person? This gives you a framework to do that.
But I think the most powerful place to use it is in tie-breaking, the idea of having multiple good options. And I've seen this in many, many conference rooms where we're making design decisions or, you know, we've brought options or doing a design presentation to leadership and it comes down to, oh, I don't think these words get there. Well, here are some different options. How do we decide, especially if we can't A/B test them or we can't simultaneously do user research. How do we decide which is the right option to go out the door? Well, how did they all measure up against the voice chart?
Gerry
You write that our words are not there to be read, savoured and appreciated. That's kind of a hard message, isn't it, In some ways for, for writers and I guess designers in general across the board, in other contexts too, to take on board.
Torrey
I think it is. One of the hardest things is when you are in love with a great solution and it turns out that that solution doesn't work, right? Like you can write something beautiful and clever and meaningful and delightful, and it just, it is pure poetry, and it doesn't work or it doesn't take off.
So there's that side of it. Like, we all, every creative person I know wants to make things that are beautiful and sparkly in whatever way. But at the end of the day, that's not what we're there for. In fact, UX text, for the most part, it should disappear from people's memory. It should just evaporate because they are there in this experience to accomplish a task or play a game or, or create their own wonderful writing or their own wonderful art. They're not there to read those buttons and think, wow, what a great button label!
Gerry
Now of course UX writing I guess as opposed to tech writing where you're typically communicating out, with UX writing, it is very frequently a conversation. How do you about designing a conversation?
Torrey
The conversational design is one of my favourite things to do early in the design process, like, early on you're thinking about, hey, we are going to make an experience. And sometimes you start from the problem, like, we need, you know, people want to be able to do this, this thing. And we found a way to do it technically in the backend somehow. But how do we create the user flow to get them there? And sometimes it's started with customer user journeys or critical user journeys. And it's that idea of a journey but sort of distilled to… I like the metaphor of if this app had to be replaced by a human sitting at a counter and having the user come up to that human and having a conversation about what they need, how would that conversation go? Who would initiate it? How would they know what they were talking about?
So those, that core, sometimes it becomes the core terminology of that experience, is how do we identify what you're here for? Where do you, what are you envisioning is where you want to be? And designing that conversation out from there. So what I have done in the past in a few circumstances at different companies and then also I've taught this method is actually physically getting up and play-acting it. So having somebody take the part of the app and a different person take a part of the user or the person using the experience and have them say, okay, you're coming up to the counter. How do you ask for what you want to do? And actually documenting that flow and trying different things out, saying what if you brought up topics in a different order, App?
Gerry
Yeah, I really, I really liked that in the book totals worked examples you had. You know, one of them was, was that that sort of body storming for conversational design. And of course the voice chart was another one. Then there's a lot of, there's a lot of very hands on advice now.
One thing that surprised me a bit and I found very interesting. I guess just because it seemed odd to be able to put a number on it, but you, you write that there are 11 basic types of texts that almost every experience must use. And I'm not suggesting we sort of talk about every one of them, but perhaps you can pick a couple out of this list and talk a little bit about them. You had: titles, buttons and other interactive text, descriptions, empty states, labels, controls, text input fields, transitional text, confirmation messages, notifications and errors, which was a really interesting framework I thought for describing all of these things.
Torrey
Well, thank you. It's funny, I am sure that there are more than 11 basic types of texts. But these were the ones… I was trying to not make necessarily an exhaustive list, but a get you started list, like if you don't have a dedicated UX writer who you get to work with and have them do that creative heavy lifting. If you have to start somewhere. Maybe you're a product owner, your fifth language is English, and you're saying, you know, how do I even start saying that in this language? These are patterns that should be able to help. And I think that I have yet to come across an experience that doesn't at least use buttons. You know, like the one, the one trigger you know, start. Great. Mm. Maybe they also need titles. Descriptions are very handy even if it's just one or two words.
I mean these are all, these are all extremely basic. And certainly not every app needs all of them. But it is, I think that the hardest one to write about paradoxically was the descriptions because there's so much that people think about writing being that is about, well perhaps the real writing there is writing the description. Like there's an error message or not even an error, but you know, some sort of dialogue or some option that has a title and then a description that tells them what it is. Well, that description needs to be a usually very brief, but not always. It needs to be extremely relevant to whoever is using it right then, which also means it drives other design considerations like could you possibly have multiple users needing very different things on this page? Well, which of those things are you going to describe? So all of these considerations, there's like, there's a lot to distil there and yet it's kind of the most disposable text on the page because it is the least likely to be read.
Gerry
Yeah. That's a hard message to get across, isn't it, people spend so long writing, you know, this form is about, or this page is about or this organization is about and time and time again in usability testing or observation, you see people just skip blithely right past it.
Torrey
Absolutely. Yeah. And I do give a couple of views about it, like keeping it below 50 characters wide, keeping it less than three lines long, like no more than two and a half lines and using like the first words and the last words of those phrases, you know, to let people's eyes at least grab something as they skim their eyes across it and never fully read it.
Gerry
Yeah. And, and also not misinterpret it. You know, if you make it longer, if it increases the chances of misinterpreting, which kind of brings us to the importance of editing, you know, which is a, I think a not sufficiently highly regarded activity. You talk about four stages of editing. Can you tell us what they are?
Torrey
Yeah. First is to edit for a purposefulness. Like is it actually meeting the purpose that it's there for, how many ideas does this need to get across? … First of course is drafting it. Whether you use a text pattern to do that or are you just you know, write it down like, oh, it should say something like this. And then you say, how many more of these ideas, you know, what, what kind of weight does it need to lift here? And you try to cram all those ideas in and then it's usually too long and very complicated. So then you edit it again for… Actually I'm not sure which, again, I don't have the book in front of me.
Gerry
Nor do I. That's an oversight on both our parts. [Laughter.]
Torrey
Yeah. I'm, I'm actually on vacation in New York state right now.
Gerry
We'll forgive you. You can get them out of sequence and we won't tell anyone.
Torrey
So purposeful definitely first because you want to get it as long and chunky as possible and then you want to make it concise. So which of those ideas can you remove, can you shorten, can you combine in some way so that you can say it shorter and shorter and shorter.
And then you're going to make it conversational, like probably then you've made it too terse to be totally usable. So then how would a human actually say it? Like you've got the right ideas in the right order now, let's lighten it up a bit. And not in a voice-y way of, of getting the brand across necessarily because that you're doing is part of the purpose and getting those concepts in that represent your voice. And then, so then you are making it conversational and then you're just doing that check to say, is this clear?
And when it's editing, it's not just polishing the one sentence or the one phrase over and over and over again. It's very different than my fourth grade teacher and my third grade teacher thought of editing, you know, let's just polish this apple till it glows. Instead, it's very much more the UX design method of let's just keep making copies and iterating and make, you know, maybe we'll make eight different versions or four different versions or 18 different versions. So then we can choose the clearest among them and we can find the best among them. And especially if they are very different from each other, we can put them against each other and bring those back to the team and say, I think, you know, here are the trade-offs between these ones. I recommend, you know, option one and options two and three are also excellent.
Gerry
And then of course, you can further iterate through testing.
Torrey
Absolutely.
Gerry
One of the things that occurred to me when I was thinking about this, this topic I guess is that it's really hard to usability test or to test lengthy texts, traditional texts if you like, whether they be on-screen instructions or whatever because you don't, I mean, you can test them as stand-alone things, but you don't know whether people are actually going to interact with them or not. Particularly something like help text for example. You often think, well, we're usability testing something that maybe nobody is ever going to see. Whereas with UX content in a way you've got the luxury of being able to test something in a more realistic or live environment.
Torrey
Absolutely. I think there is, it is almost impossible, although I've done some of it to to accurately test help text as you said. There is a pitfall in UX text testing for text because in the testing environment, especially if they've come into a special room and there's a one way mirror and it's very fancy, people tend to take more time and be more careful and read more of the screen. I have been part of some testing pieces where the part I actually want tested is presented as either, you know, a setup task or some sort of meaningless paperwork task so that they can really approach it in a more realistic way. But that is very hard to do and can't be done with all experiences.
Gerry
Now the book is a great practical guide for anyone producing or wanting to produce good UX content.
For someone who wants to get into this area, perhaps they're working in technical communications or you know, in some sort of related area, if they want to get into specifically UX writing and I guess strategizing UX writing, besides reading your excellent book, what, what other activities should they undertake?Torrey
I think one of the most important things is to create a UX portfolio and a UX writing portfolio specifically. And the way to do that…
I mean, there, there are definitely classes available. There are definitely there's somebody who am, I forget his name at the moment is producing a sort of daily challenge for UX writers to go through how would you write this and explain your reasoning behind it. I think one of the biggest challenges for portfolios and presenting portfolios to potential hiring managers for this role is that technical communicators have had very little training in what that looks like. Whereas for UX designers, that is part of the fundamental training. What I recommend to people is to, you know, do the work, have a point of view about it and be able, and practice talking to UX designers, having career conversations with UX managers and practicing talking people through the UX decisions that were made in order to take the text from this before state to this after state. And why that's an improvement in what the expectations are there.
Gerry
Do you think that the skills that somebody would have somebody like yourself would have, well, a UX writer in general would have, translate well into voice user interfaces, conversational devices and the like.
Torrey
I think it does because the core skill is first and foremost working on user experience design, whether that's with the visuals or the onscreen interactions or the voice. All of those skills about empathizing with the customer and really helping them get where they're going are critical. And what the people with a language background and that facility in playfulness with language, that translates well into voice user interface design. There are significant differences in the tools that are used for voice user interface versus UX design or UX writing design and traditional writing. But those are tool differences that are getting, I mean there are always going to be different tools, but we can learn new tools.
Gerry
And it occurred to me that the voice chart would be very useful in fact, for doing that sort of design as well.
I'll remind listeners that Torrey's book is called Strategic Writing for UX: Drive Engagement, Conversion and Retention with Every Word. And it's very much a hands-on practical book and I would highly recommend it to anybody who's interested in getting into this area. It's very clear and useful. A transcript of this episode is available at uxpod.com.
Torrey Podmajersky, thanks so much for joining me today on the user experience podcast.
Torrey
Thanks so much, Gerry.