The SpoolCast with Jared Spool

The SpoolCast has been bringing UX learning to designers’ ears around the world since 2005. Dozens and dozens of hours of Jared Spool interviewing some of the greatest minds in design are available for you to explore.

Episode #170 Steph Hay - Writing Content for Usability

April 20, 2012  ·  30 minutes

Listen Now

Download the MP3

Content is everywhere. With the amount of content users are confronted with everyday it can be challenging to garner their attention. Compounding this problem is the fact that designers and developers are often tasked with writing content that end users see. This can be an intimidating prospect if you’re unaccustomed to crafting copy.

Show Notes

Content is everywhere. With the amount of content users are confronted with everyday it can be challenging to garner their attention. Compounding this problem is the fact that designers and developers are often tasked with writing content that end users see. This can be an intimidating prospect if you’re unaccustomed to crafting copy.

Luckily, Steph Hay, Co-Founder of FastCustomer comes to the rescue in her virtual seminar, Writing Content for Usability, full of her tips for writing compelling content. She explains that choosing the right words and tone are essential to getting users to take action. The audience asked a bunch of great questions during the live seminar and Steph joins Adam Churchill to address the ones that we didn’t have time for in the podcast.

Tune in to the podcast to hear Steph answer these questions:

Full Transcript

Adam Churchill: Welcome to the SpoolCast. Recently, Steph Hay joined us for her seminar "Writing Content for Usability". Designers and developers often have to write content that end users see. In her seminar, Steph explained the four characteristics to compelling content for web and mobile experiences. She left our audience with practical techniques to write more effective, engaging copy that makes people do something like sign up, buy, or simply come back for more.

"Writing Content for Usability" has been added to UIE's user experience training library, which presently has closing in on 90 recorded seminars from wonderful topic experts just like Steph Hay giving you the tips and techniques you need to create great design. Hi, Steph. Thanks for joining us today.
Steph Hay: Thanks, Adam. I love being here.
Adam: For those that weren't with us during your virtual seminar, can you give us an overview of your presentation?
Steph: Absolutely. Really what I was getting into was how to go about writing compelling content. As you said already in the intro, and compelling content here is a sort of action-oriented content, the stuff that makes somebody want to do something. So what I outlined was the sort of process by which you might arrive at writing content.

That's, number one, to focus on audience, medium, and network, and audience being really specific people. One person, if you can imagine that person sitting across the table from you, versus a wide demographic of folks from age 18 to 35. Being able to really focus on a single person sitting across the table from you would allow anybody to write content in an easier way because you actually can understand what that person might be asking, what are the sorts of questions she has, what are the answers that you can provide?

So focusing on an audience and a very specific audience is really the first step of writing compelling content. Then really considering that medium that you're going to be using to communicate with that audience is the second step. How are you going to communicate on a homepage or a website or on Twitter or on a blog?

And then, how can you also use additional media, email for example, to help tell a story and compliment your message, which really takes some of the pressure off of you having to write everything you could possibly write for that audience in that one little spot that you're probably writing content for.

Then, finally, when it came to focus considering that network of folks around you who are going to help influence your message that really needs to be a part of your defining focus, but at the very start before you write a single word on content. That first step of focus then leads into the compelling content. The four characteristics of that that I go over in the seminar and provide examples of include helpful, meaningful, result oriented, and confident. Compelling content that has these four characteristics really actually helps a user make the decision to take an action.

The third part of that process after you've focused your energies, written helpful, meaningful, result oriented, confident content is to go back and make sure everything is consistent in voice, structure, and tone.

Then, finally, what I did in the seminar is provide a step by step way, a four step approach, that folks can take if they're actually writing content so they don't stress out and actually arrive at an end result that's compelling content that they really need to write more of, in my opinion.
Adam: Steph, is there a difference between Marketing Content and Usable Content?
Steph: Yes. In short, marketing is the content creation and testing process itself and usable is the outcome of that process. In fact, I should have explained this in my seminar because, now that I think about it, a bit of the feedback made me realize that some of the practitioners who attended understand "marketing" as home page content and "usability", to them, means button content or call to action, which I'd argue is too compartmentalized to sustain.

As a reminder from the seminar, I mentioned those four characteristics of compelling content, meaningful, helpful, result oriented, and confident, and I showed home page screenshots really as apples to apples comparisons of how different organizations are demonstrating those characteristics.

Even if you're just working on a back end, for example, UI or you're writing on a technical document or something like that, that apples to apples comparison of the home page was explicitly, actively used by me in order to find that common ground between organizations.

You still have to have some forward face on 99.9 percent of the cases so that forward face tends to be a home page or a landing page. As a reminder, that compelling content that I mentioned intrigues a user to do something. So the problem is that most marketing content is written by someone who hasn't actually spoken to the typical end user.

That person writes blindly based on what she considers to be interesting rather than by using the kind of language and truly substantive content that I think the end user needs. This is really at the core of content that's usable. People who read marketing content that's well written, they get it because they think you, as the person behind that, you get them and you've already found what sorts of messages they prefer.

I just want to say that marketing content and usable content don't describe where content shows up. For example, marketing content isn't just the headline on a home page or the about page content and usable content isn't just a button nomenclature or call to action in the side bar.

Marketing content is really any content that's written for a user and usable content is the outcome of effective marketing content because more users take action and the right action that you want them to take.

Can I use a little case study of something that just happened yesterday?
Adam: Absolutely.
Steph: OK. Cool. So FastCustomer, my startup, we're doing usability testing yesterday. This is just a mobile application UI that I'm talking about right now. FastCustomer you never have to wait on hold again for customer service. So most people get it. They don't need to be convinced that they should have this product. They download it. Then they understand that having a button called "Have them call you", which was very user based language, explicit, came from the user, they didn't have a hard time understanding that, either.

They pushed the button that said "Have them call you". They knew someone at a company was going to call them if they pressed that button. But then, we quickly added the response language ourselves and the response language was "Call in Progress" which, during usability testing just yesterday, was making people literally lift the phone to their ears.

So here's an example of a marketing hook that we used using the user's language to get an action that we wanted the user to take and that the user wanted to take, but then the response that we wrote out of our own developer minds caused confusion and therefore it really wasn't usable. Once I saw the users bringing the phone to their ears when really they didn't need to be I just asked, "What did you think would happen when you first pushed the button?"

And from that, I got the language that I needed in order for the user to understand what was happening to really make it usable fully, which was simply just you can close this app any time. It'll keep working. They weren't sure what was actually happening with the call. They thought maybe they had to be on it and they thought maybe if they got another call it would interrupt it and that sort of thing.

So clarity here is a stronger indicator of reuse later than inadvertently introducing confusion because we're using the wrong lingo. Even though we want them to stay in the app, we really don't want them to close the app, at the end of the day we really want them to understand what's happening. That's really, I think, the best marketing content that we can have out there that's really, in the end, usable.
Adam: It makes me think of we follow pretty closely the folks at Marketing Experiments. One of the things that Dr. Flint McLaughlin always says is that when you're thinking about marketing copy versus content is that clarity always trumps persuasion.
Steph: Yeah, man. That was real clear yesterday. My designer, actually, had been there with me to go through the usability testing and we just looked at each other and I just said, "We just have to tell them they can close the app."
Adam: When it comes to the process of writing content, we had some folks that wanted you to say a bit more about that. You mentioned that content is considered in two places in the typical design process and Sid wanted to know is that typically what you recommend? When should content be considered?
Steph: Yeah so, when I had gone over the process of how content typically is written today it was really born from my experiences in agency and consulting one-on-one with companies where it really starts at the very beginning when you're thinking about information architecture.

Then, it concludes and it only appears again at the end when you're actually going through all the pages that you decided to have during the information architecture planning and you realize that there's content that's missing on those pages so you hurry up and populate them right before launch.

This is something that I think needs to go away. Content should be considered before the project even starts just by answering the question what do we want our users to know? By doing that, we've already considered audience and the outcomes you have to express to them in order for them to care about what you're saying. So, "what do we want our users to do?", once they know that, is then the ultimate conversion point.

Without these two questions answered, what do we want users to know and then, secondarily, what do we want users to do once they know that, we'll just get distracted during the project life cycle from the purpose. Even though there are several answers to each question, that's OK. It's just a matter of prioritizing from there.

But thinking about content so literally in the traditional design process as the site map and as the fill in the blank before the site launches, means that technology and creative design are driving at all times. And I think, you know, if I think about the content strategy side of things, inventories are only important if you're an archival organization. If users won't make any decisions based on that content, why include it? That's a real question that you need to ask yourself.

If you happen to have a ton of content and you're spending energy and time sustaining it rather than it actually helping the user to make a decision. That's why I advocate for a design process in which every deliverable is brought back to the audience outcomes and goals. Who are we speaking to, what awesome things do we promise them that our competitors don't, and what do we want this user to do?
Adam: Marley asks this question. Often our audience, our attendees, want to know from experts like yourself what tools or exercises are available for them. In this case, for copywriters and designers, are there tools or activities that you recommend that help them work together in creating design and copy together?
Steph: Yeah. I mean I can only speak from my own personal experience here, but I'll just tell you what I do in both of these cases. I start with a text file, just a plain text file. I've definitely had clients send me comps, mockups of their new site design, and say here's where we're at, we need things for this and that. Frankly, I just tend to ignore them because I can't be distracted by the design right now.

I know you're being distracted by the design, but I have to think about the user. I typically start with a text file. In fact, for my own personal website at, did this for, done this for clients' websites, oh and I know that Ben the Bodyguard, which I used in the seminar as an example,, also followed this sort of process.

I, as a copywriter, created the content in a text file in conversations with the client, with end users, and then handed it to the designer and the designer really built the site around that. It was collaborative in that way, but in this process I started with the content. Once I handed it to the designer and some comps came back, actually, I should say, I made a few UI/UX suggestions in the notations as part of the content.

Otherwise, I left it entirely up to the UX designer to create wire frames around the content itself. In all cases so far, the designer has really loved this process. It's liberated that person from having to think about the content because it's already there and that's what's ultimately speaking to the user. It's created a structure and a process around the most important stuff to be communicated without requiring the visuals to lead back to it.

Once I see the wire frames, then, in every case I've revised the content. I'm actually not thinking about UX in this case, I'm just revising content where now as I see it's structured it could be a little tighter here and there or it's missing something here and there. On a few pages, the designer has suggested different content, which is awesome, so it's very back and forth and collaborative in that way.

Because the content can't exist as vibrantly on its own, it evolves with the design and vice versa. It ends up being a much more enjoyable process and then a much more powerful outcome, too.
Adam: Let's talk about prioritizing the content and messaging. Peter wants you to say a bit more about audience versus tasks. If you take a task based approach it can help when you're trying to consider, I guess, the fact that it doesn't really matter who you are, it's more about what the site visitor wants to accomplish.
Steph: Right, right. No part of the creative process should ignore audience, in my opinion. Whether you're designing an interface or building features specific to a type of work, you have to consider who is using it and what does that person need to accomplish in whatever task they're trying to accomplish.

To ignore them when speaking to them, the written word would, I think, do two things. One, result in non-specific, static, boring content that really appeals to no one, which hence is the kind of shitty marketing content that pervades so much of what we must read today out there. Or, two, it complicates the writing process itself by making it a one way conversation which is difficult to write in the first place and, frankly, even more difficult to learn from.

So I'd argue that the audience can never truly be separated from task when it comes to choosing or writing the best content to help that user achieve a goal. That said, I do think there are some UX strongholds that all audiences understand. Using the phrase login, for example, rather than attempting to be different for the sake of being different, using something like access instead, for example. That's just interjecting weirdness where weirdness doesn't need to exist.

Or a field name called email address is totally fine, you get it, you don't have to think about it, the audience rather, than something like where would you prefer to receive email? Most content we create, test, and stress over, which is really what I am addressing a lot in my conversations with folks, isn't this cut and dry. It's not choosing that UX stronghold label, for example, that doesn't really take into consideration audience because it doesn't need to.
Adam: Brooke wants to know how you avoid redundant content if you're tailoring for more than one audience. An example here would be the features of a product, right? You might need to craft content that is specific to people that are resellers versus people that are actually the customers or the users of the product.
Steph: Yeah. I think if you have a lot of redundant content a problem may exist more with your IA than with the content itself. If your information architecture is unnecessarily broken out into audience specific verticals that force you to do a lot of redundant content and describing the same product or service then I suggest reorganizing the content around the subject matter itself rather than the audience.

Plus, once you have your content organized around the meaty stuff people are looking for then you can actually do some more granular testing of what's working and what's not. In practice, this requires that you choose a primary audience who will be reading that page and that's really probably the most difficult thing for a lot of these very information heavy organizations and websites is to prioritize a primary audience.

If you can do that, and once you've done that, then you can just write the content for that audience. Then any other audience is pulled from that page onto other pages with more detail or different detail or it's in the content itself but it's otherwise hidden until clicked, some cute little JavaScript-y thing.

That way, your page is clearly speaking to one audience but offering more info for secondary audiences and beyond who either self-identify themselves as those types of audiences or they need more information they haven't found yet and you know what that information is and you've just compartmentalized it in a way that's hidden naturally.

Either way, you've now learned, assuming you're tracking clicks, which other audiences are on that page often and what kind of content they're looking for. I think the real problem, and why so much redundant content exists, is because we assume we've got these specific audience types who need to have the content set in this exact way and, in doing so, we actually introduce confusion.

Because, you know, aside from maintaining it being difficult, it's actually difficult for the user to then sift through and figure out what is relevant for that person versus the other audience over there where it's sort of saying the same thing but in a little different way.
Adam: Can you take that a step further and explain how folks that are creating government websites might think about that?
Steph: Yeah, yeah. That was a good question that came up during the seminar. It's almost April 15, right? Having been on the IRS website I can definitely say yes, I have some tips about this. They're really similar to the ones I just described, which maybe this is just a pipe dream right now.

Anyway, government sites, they're especially tricky because the variability of audience type is so incredibly vast. But that said, the great aspect of a government site is that it tends to be a focused topic. So most agencies deal with a particular subject matter pretty fully. The best advice I can give is to take your five to 10 most common questions and to structure your site around those topics.

I mean, yes, I know there are plenty more questions that are asked but it's mind boggling how much information is actually related to just a core 5-10 questions. From a structural standpoint, that means the core pages would include information relevant to nearly all audiences. I would say, ask a question like what policies and rules apply to most people who will be on this page, which is all about this very specific one question of those 5-10.

Include only the information that applies to most people who would be on that very specific page. Then from there, offer a couple buckets, one that's oriented around audience self-identification, and one oriented around topic. In the case of the IRS, as a small business owner, I just need to know what I'm supposed to do with my quarterly estimated payments and then maybe there are some links on audience type that say if you are a disregarded entity, by the way, you are now called this or whatever.

Now, I'm looking based on audience. Or maybe there's additional content related to topic so, are you looking for how you should be paying your annual taxes now? Something like that where once I've actually taken some action, either identified myself as this particular audience or identified that I need this particular information, now that additional content of yours or I'm taking to another page.

But most of my questions have already been answered by the content that applies to me and a great number of other folks like me. The other thing is taking this approach, which is a true funnel where the most general and overlapping information starts the site and continues getting more granular with each additional action the user takes, and keeps getting narrower then, will help the user feel like she's going through an information heavy yet guided experience, rather than just being thrown every option and every detail and every link at the top level and expected to make sense of it and sift through it.
Adam: In the seminar, you talked about establishing or maintaining a consistent voice across your content. John's looking for your best recommendation for getting users to complete a process like completing a transaction. In an example like that, would you use an authoritative voice or more of a personalized? I assume that means more of a conversational type of voice.
Steph: Well when it comes to transactions there's really no hard and fast rule here, but transactions absolutely require confidence. The details are vitally important. No typos, no big swings in voice from friendly to matter of fact or from, in this case, personalized to authoritative. It just can't shift. I can't think that something has changed between the time that I've made the decision to start going through this process to the time that I'm actually completing my process.

If your marketing content here that's successfully led a user into the conversion funnel has a personalized voice, I say just keep it consistent, same with formal/authoritative. The most important thing is that it does not shift so as to make the user stop and wonder if you've really paid attention to this process in full which, oh by the way, is probably asking for my credit card, too.

Actually, for example, the other day I signed up to auto refill this Starbucks card I have on my iPhone. This, by the way, is an incredibly smart way to get my money because someone gifted me the card digitally in the first place as a thank you, which then made me download the app which then made me pay, while I was in a Starbucks, for using the app which then was so fun and cool and novel that I never wanted to not use my iPhone to pay at Starbucks which, of course, now made me do this auto reload thing so I never run out of money on my Starbucks.

Anyway, the last screen before I click yes, confirm that I want to auto reload, which I have to do on the web, and P.S. I've just given you my debit card number, there was a typo. It said "it's" instead of "its". I actually faltered. Now, had my experience not otherwise been so compelling and fun I might not have confirmed things because it's just such a boob move.

I mean, a typo from Starbucks, who can clearly afford copy editors. This was a totally preventable mistake. My point is just this, that Steve Krug said, "Don't make me think." Just don't make me think. That's not about taking an action in the first place but also about undoing one that I'm already taking.
Adam: Did that typo in this situation make you question the security or the authenticity of actually who the transaction was happening with?
Steph: You know, the Starbucks brand is so known, it wasn't really questioning their credibility. It was a literal stop in my process that didn't have to exist there where I stopped and I said, "Starbucks, seriously? You guys could not review this content?" It was an unnecessary questioning of their process because I work in the web, of their ability to actually capture these sorts of small but important details, important to me as a writer.

The whole point is just about that stop that exists. When you introduced a stop because you have created inconsistency, when you've raised a question that is in direct contrast to your credibility. Starbucks, they have the brand to fall back on. There are stores everywhere, you see them. Most of us, we don't have that, so we can absolutely not raise that credibility question.
Adam: Steph, we talked a little bit about the .gov folks. Let's switch a little bit to the .edu folks. We heard from the University of Texas libraries and they made a comment that their large academic library has dozens of subject matter experts and about 80 distributed web authors. They're actually working on style guides but are afraid that alone won't resolve redundancy issues or, as you were just talking about, inconsistencies that might arise from the tone or content quality.
Steph: OK. You know what, when it comes to voice and content consistency, I think that libraries and similar search based sites have some wiggle room, frankly. And here's why, because audiences understand that the people who have written the books and research papers and abstracts that they're searching for are different people.

So I think, in being the sort of body, the umbrella, the glue that holds all of these individual subject matter experts and authors together, the primary charge for the body above that is just to ensure that anything that's not a subject matter based page sounds like it's coming from the same brand.

A style guide absolutely can help with this, though depending on how large a site you've got, I'd consider limiting those as much as possible and having one person, a writer, edit them. Think about Google, which is arguably the biggest library, or Wikipedia. You go there and you search. That is the core of those things. Any other content around that should be relatively limited and, in that case, it should be, in my opinion, relatively easy, then, to pull one person into a project to go through and make sure there's a consistency in voice and content.

The rest, however, for libraries and real search heavy types of institutions would be to surface the content as semantically as possible, I think. Various content management systems can help with these, tie into real time content searches coming from Google elsewhere and that sort of thing, but the real core focus of users on these sites is the search.

So I think, you know, not to stress too much about consistency and that, instead more, I think, the focus should be on limiting any pages that aren't directly tied to search.
Adam: This has been awesome. Let's sneak in one more question. In the seminar you cited a FlipKey example. They had a headline, "Rooms from $49 a night" and you talked about how that stated the obvious rather than the result for the person that the headline was trying to capture the attention of. Kate wants you to suggest what a better headline might be.
Steph: OK cool. So when I was talking about compelling content one of the characteristics I said was that it need to be results oriented and this is the thing that I think is most problematic and the thing that is most missing which is that people tend to state the obvious of what you're going to get rather than what the result is of why you should get it from them. It's a step beyond.

It's like a bear hug that happens when you find out from somebody you're thinking about purchasing from that there's this whole other story behind it, there's this whole thought about you as the user and what you're going to get out of it. In the FlipKey example of rooms from $49 a night, that's not a result. It's just about the FlipKey promo.

It really has to be about what the user would get beyond what she would get if I went to Hotwire and saw $49 a night on Hotwire. Why should I buy from FlipKey versus Hotwire? I would say a better headline might be something like "The greatest night of your life will only set you back $49". That, to me, sounds a lot better than "Hotel rooms from $49 a night".
Adam: Steph, you are the best. Thanks so much for circling back with us. I appreciate it.
Steph: Thanks, Adam, and I appreciate everybody who's listening out there. You can email me questions, and that's my website, too,
Adam: Yeah, and I guess I should have mentioned this in the intro but at UIE we're pretty particular about our site copy. We needed some help with it and we turned to someone who we think was the best person to turn to and that's you. Anybody out there that needs help with their copy Steph's a great person to turn to but please don't steal her from us.
Steph: [laughs] No way. It's impossible. Don't worry, Adam. Thanks, though. I appreciate it.
Adam: For everyone listening in, thank you for listening and joining us and, of course, for your support of the UIE virtual seminar program. Remember you can get all the details on upcoming seminars at