Kristina talks with research and information design expert Erika Hall about the art of conversational design and the challenges—and opportunities—of how we use language to interact with users in online contexts.
Jonathan Foster leads Windows & Devices Group’s Content Intelligence team at Microsoft. Their work includes defining content experiences powered by and instantiated in AI, writing conversational interactions for the virtual agent bot on support.microsoft.com, as well as Microsoft’s digital assistant Cortana in the US and international markets. He and his team are responsible for the continued development of Cortana’s personality, crafting fun, personal, and culturally relevant experiences across the globe, as well as building an ML-powered conversational layer to support Cortana interaction. Follow him on Twitter at @jonathanbfoster.
Kristina: Hi, welcome to The Content Strategy Podcast. I’m your host Kristina Halvorson, this podcast is brought to you by Brain Traffic. Brain Traffic is a content strategy consultancy headquartered in Minneapolis, Minnesota, and serving clients worldwide. Visit us online at BrainTraffic.com.
Hello again, my guest today is Ms. Erika Hall. Erika is the cofounder of Mule Design, and the esteemed author of, Just Enough Research, and the brand-new, Conversational Design, both from A Book Apart.
I feel like I am doing an auto sales commercial right now for your book, welcome Erika.
Erika: Hello, thank you so much, I appreciate it.
Kristina: I’m setting you up for success is what’s happening right now.
Erika: Because that’s what you do, that’s what people hire you to do.
Kristina: That’s exactly, sustainable business success if you will, and I will. Hey Erika, tell me about your new book.
Erika: It is so exciting, because it is done, it is out there. You’ve written a book, you know, you know the feeling.
Kristina: Oh I do, I often liken it to you get pregnant, and you’re like, “I’m pregnant, I’m having a baby,” and then seven or eight months into it you’re like, “I’ve changed my mind, I want to go home, this was a terrible idea.” Then once the baby’s here you’re like, “That wasn’t so bad, I would do that again.”
Erika: Maybe at some point, and so the funny thing about this particular book, especially what I experienced when I was writing it, is a lot of what I say in the book is that the absolute worst thing you should do is lock yourself alone in a room and try to make words come out.
Kristina: You’ve written a book about writing and design that tells people to stop writing?
Erika: Well, it’s not exactly about writing and design, writing is a part of it, it’s about using language definitely, and it’s about the fact that talking, like what we’re doing now, and maybe this is controversial, but maybe you’ll agree with me, that having this conversation is so much easier than sitting down and writing.
Kristina: Everything is easier than sitting down and writing, why is that important right now in the design process?
Erika: This is important, because we’re at this transition point, this inflection point, this great hot mess of a lot of things coming together, where we have all sorts of powers being brought to us by cloud computing, and machine learning. We still have the old timey Internet made of webpages, and we have all of these things, and I think people are still looking to old ways of doing things to do these things. In particular for the use of language, there are a lot of people still using traditional writing processes, and traditional more documentation-oriented processes to create interactive experiences.
That’s what I wanted to address, was getting people to collaborate to design things in a multidisciplinary device independent way, that really starts from looking at what you want to say to your audience, your customers, your users, whatever, like how you want to help them, regardless of what channel, what device, whether you’re talking, whether you’re typing, you have to step outside of that, and say, “How do we make this interaction as easy as having a conversation with somebody who knows what they’re talking about, and is friendly like you?”
Kristina: I feel like we could just wrap on that point right there.
Erika: Yes, let’s wrap.
Kristina: No, and yet we have a podcast to record, what got you thinking about this?
Erika: Well, I first started thinking about this way, way back in 2007. My first conference talk was called “Copy is Interface,” and it was about the idea that interactive designers were really hyped on the visual parts and the behavioral parts of whatever they were designing, and the words just got left, the words got lorem-ipsumed out. My point was that, “Hey, the words are often the thing that defines the interaction,” like you can’t just make a container and say, “Oh, we’re going to worry about the form, and then fill it with meaning at some point down the road.”
You have to start with meaning, and you have to start with really respecting somebody’s time, and you can’t do that if you don’t worry about the language you’re going to use until some point months later.
Kristina: Do you think that, that has changed over the last 11 years?
Erika: Let’s say, I think some people are more ahead, like some organizations, some individual designers are more on top of this than others. I think especially now in the last couple of years, and this brings me like how I got from that talk like 10 years ago, to the book that just came out, is I thought about writing a book about that for a while, about really thinking about how what we’re using the Internet for, is just a lot more about having conversations with each other, and going back to this like oral culture that we used to have, and there were ideas around that I was talking about.
Then all of a sudden everybody started talking about chatbots, and voice interfaces, and I went from, “Hey, everybody should get more conversational in their language, to whoa, whoa, whoa, that’s not what we should be focusing on when we talk about having conversational interactions,” because if you’ve dealt with Siri or Alexa or Cortana, or any one of these little pseudo-AI chatbot things, you might’ve noticed that they’re not actually usually faster, or easier to interact with than just clicking things on a menu.
Something got lost, people started focusing on the wrong parts of what makes a conversation work. Just because you’re having a voice interaction, doesn’t mean that’s conversational.
Kristina: What are the parts that make a conversation work?
Erika: I start from, if people are into linguistics and philosophy, I start from Paul Grice’s Cooperative Maxim, which essentially boils down to a shared goal and a willingness to help each other accomplish that goal in a cooperative manner. This is the reason why you can walk up to any random person on the street and ask for directions, and be pretty sure that you can have a reasonable exchange, as long as you both speak the same language, even though you’ve never met them before.
They will help you meet your goal, and you don’t have to go up to them and make any explicit agreement, this is implicit in every conversation. The idea that it works, because we both implicitly agree to help each other.
Kristina: Hypothetically that all makes perfect sense, but obviously when you’re dealing with what did you call it? These little AI, pseudo-AI chatbots, is that what you called them? That’s what you called Alexa? There’s the challenge of not being able to be standing in the middle of a sidewalk and seeing somebody walk up to you, and seeing somebody who’s clearly lost, go and start to ask you for help.
Where would you have advised the designers who are currently working on these chatbots, like where are they going wrong in terms of this is not any easier, or this doesn’t necessarily have less friction than just going and typing something in and looking it up that way, where are they going wrong?
Erika: Well I think Alexa’s probably one of the better cases of this, because it really is like ... Amazon is a fantastic company to look for as a model of design and using language and learning, and they’re always running experiments. I think Alexa for them is a pretty successful experiment, in that they’ve gotten in people’s homes, and it does do some things well. I have one in my kitchen, and it’s the world’s most complicated and expensive timer for when I’m cooking.
It plays music, it reads audiobooks, but if you look at how Alexa fails, right? Amazon is an amazing machine for taking my money, and as a shopping experience, as say a touch point for shopping, Alexa totally doesn’t work, because you have to have other sensory input really to shop, because that’s a really high commitment. You don’t just want to sight unseen say some words, and have your credit card be charged for hundreds of dollars.
I think the difficulty that they’re finding, is how to cue people to what’s possible. There’s no way, aside from the things I know how to do with Alexa, there’s nothing that tells me what I can do, right? That’s a really, really hard problem to solve, if you go to a website, right? Say you Google something, you express your intent, it sends you to a website, that website usually at this point now will do a pretty good job of telling you explicitly, or through implicit navigational cues, or using various design conventions, will give you an idea of what is possible for you to do there.
Alexa has no cues, and it doesn’t have any of that suggestion or recommendation engine, so this is what’s so curious to me knowing what Amazon does really well, is that Alexa never suggests anything. If Alexa notices that do something, and maybe they might be doing this very intentionally to not creep people out, because I know Alexa’s been really creeping people out lately, because it’s been talking unprompted, and so maybe that’s an experiment to see if Alexa offers suggestions, how freaked out people will be.
There’s nothing to let you know what you can do, and so people are just going to keep doing the same set of things they know how to do with it, and never expand their experience. Sure, if you’re really keen you could look up different Alexa skills, essentially the little app for Alexa, but do you have one of these things? Do you have a Google Home?
Kristina: No, because I still have yet to be convinced that they’re not actually listening to me, and it’s so stupid, because honestly Amazon already has every ounce of personal data that I could possibly feed into a shopping site anyway, so I why would care if it’s listening to me I don’t know, but regardless I don’t have one.
Erika: I can understand that, because Jeff Bezos, his look has really shifted towards Bond villain, so that combined with the fact that he owns the Washington Post, sure, yeah, there’s a speaker. I feel good with that.
Kristina: See, it’s like the default is for me to grant access to that constant ... I don’t know, anyway I do not have one. I have enjoyed very much watching other people try to get Alexa to do what they want done.
Erika: Siri, some people still try to use ... I never really bothered with Siri, because I could just hit a few buttons and do whatever, and I don’t like that feeling I’m feeling. Then with the chatbot interactions, about a year ago everybody was coming up with some little bot that you could text, and a lot of this was because people saw the success of WhatsApp, right?
Especially in China, I think a lot of people in business and technology say, “Oh, what’s happening in China, and can we copy that?” One of the big differences between AmErika and China, is that Google didn’t really take off. Google was blocked there for years, and so people started finding other things to use to accomplish the same things. I think the texting and chatting about everything really took off in that market in a way that doesn’t totally translate to the context that we are in.
It’s not necessarily that just having little apps be little chatbots sitting on top of Facebook messenger makes anything easier, because we have so many other ways of accomplishing things that we’re used to.
Kristina: This is a really interesting question actually, that there’s so many different ways of accomplishing things that we’re already used to, and this of course 20 years ago is what the Web was facing, why would I want to send an email when I can just do what I’ve always done and type this up and send it away? Why would I want to IM somebody who’s sitting three desks over from me, when I can just get up and talk to them? Whatever, I mean, so now we’re faced with this, “Oh, all this change is coming, VR, AI. Everything is going to absolutely change the way that we do our work.”
One thing that I keep hearing over and over, is that the website is going to die, but what I hear you saying is that it’s so much easier and so much frictionless to be able to just go and type in a few words and get what you need from a content perspective or from an activity or action or whatever perspective online, that behavior change around that is going to require some significant design breakthroughs. What do you see coming there?
Erika: I think the real design breakthrough is a shift in attitude on the part of the designer to not be so device specific, but to say, “Okay, so what is an organization?” We can start with Amazon, because Amazon is in the process of providing absolutely everything. Now they’re winning awards for their movies and television shows, that’s where we are with Amazon. You say, “Okay, I’m going to interact with Amazon,” and the interesting design isn’t the app on my phone, or isn’t the website, it’s being at Amazon and saying, “What are all of the ways that we can provide something to our customers, and get something in return? How do we do that across all channels?”
That they were really early in doing something really interesting and multimodal with the Kindle and Whispersync. Have you used this?
Erika: Yeah, so I had stopped buying or reading novels for a while, and then I started running a lot. I was training for a marathon, and I started listening to audiobooks, and the fact that Kindle Whispersync let me listen to an audiobook on my phone, and then I could switch immediately to reading it on my phone, or reading it on the Kindle app on my iPad, or reading it on my computer, or reading it on my Kindle, Kindle, and then I could pick it up again the next morning when I went for a run, and it was seamless.
This was Amazon really ahead of this multimodal interaction, right? The book as a thing, the concept of the book existed somewhere in the cloud, but Amazon always knew where I was, and allowed me to immediately pick up the interaction on any device where I was in time. I think that is the fundamental concept that designers should be thinking about. You’re not designing a mobile app, who cares? That’s boring, you’re designing something that allows people to, “Oh, do I not want to speak aloud? Do I want to very discreetly touch something on my phone, and not have people around overhear it?”
I should be able to do the thing. Am I in my kitchen and I’m cooking, doing something with my hands, and I should be able to speak the thing, and then maybe I should be able to pick it up on a different device where I type something on my laptop. It’s that place where you’re thinking about, it’s almost as if what you’re designing is the spirit that haunts all your devices, but it’s able to remember who you are and interact with you, using the most context appropriate input and output.
Kristina: As I’m thinking about this, this is all A, blowing my mind as you often want to do, B, I’m also seeing it primarily as a first and foremost design problem. As you know this is called The Content Strategy Podcast, and content strategists, one of the big things that I’m constantly hammering on, is that we’ve gotta get our content savvy folks in the room as early as humanly possible, because at some point words will need to be introduced.
Talk to me about how you see content strategists helping to solve this challenge.
Erika: Well I think the first step is thinking about the word content very carefully, and perhaps choosing a different word, because content implies a container. What the people who are currently going by the name of content strategist are about, is articulating meaning, right? That needs to happen in the beginning, so the people, and I know that the historical challenge of content strategy, or web writing, or they’re called content designers now.
There are all these words, but the people who are in charge of the verbal meaning part of things, are typically brought in at the end, right? Because of the process I was talking about, where it’s, “Okay, if you’re going to design an app, the first thing you do is draw boxes.” I have a quote about exactly this in I think one of the early chapters of my book. It’s like, okay, you draw boxes, everybody draws a lot of pretty boxes, you decide how the boxes are going to animate or whatever, and then you’re like, okay, let’s bring in the word person to fill the form with content.
What you’ve gotta do is flip that, and say okay, the people who understand how language works, how meaning works, and who really understand the value of the communication and how to work with that communication and how to create a brand voice and a personality, those people need to be at the very, very beginning of the whole process. The first question you want to ask, is what are we offering to people? How do we articulate what we’re offering to people?
What information do we need to get from people in order to provide the value that we claim to be able to offer? What’s the personality and the voice? All of that has to be decided before you pick what device you’re going to provide this value in. Who cares? Everybody’s expecting things to just be constantly available everywhere, and so what you need to do is have the people who know how to make things meaningful say, “Okay, let’s talk about the different contexts,” and so this also implies researchers need to be up there, ethnographers who understand the social context of interaction up there with the word people to say, “Okay, before we draw anything, let’s just ourselves get together in a room and have a conversation to say, ‘Okay, what interaction are we providing and whose going to pay us to provide it?’”
Kristina: What’s so crazy and difficult for me, is that this all is just forehead-slappingly obvious once you say it aloud, right? Yet so difficult for design teams and marketing and UX organizations to I think get their heads around when they’re still trying to clean up their website from 2012. You’re also in the Bay Area, and you are able to reference companies like Facebook, and Slack, and Airbnb, and these companies that are working hard to deliver this experience across devices, that are offering more of a product I guess.
Then so many organizations are still so siloed with marketing over here, UX over here, brand over here, and what you’re talking about is a fundamental shift in the way we are thinking about delivering experience and information. If I am a CMO, and I’m asking you to sit down with me and help me start thinking appropriately about not the container, but about as you said meaning and language and purpose and outcome, where do I start?
Erika: Very coincidentally you start by getting everybody in a room to have a conversation about these things, because you really have to do these things. In order to really design an interactive experience that respects people’s time, you have to start by interacting with people in your organization. I think the big challenge for these teams like you’re saying, people are in silos. People’s jobs are defined by the type of documentation they produce, and we have to get past this.
I mean this is even true of writers, people who come from a traditional writing background think about it in terms of the words taking up space. This is maybe for some people who come out of advertising maybe they get this, like advertising copywriting has to be very brief, and is often designed to work in multiple media. You have to let go of the idea that when you’re creating words, that this is something that you go off and do yourself to create.
People still do this for websites, they create copy decks that go through levels of approval and editing in an asynchronous manner, where people at higher and higher levels of the organization essentially mark up a copy deck that then gets handed to somebody to put onto the website. That is madness.
Kristina: You realize of course that, that is 90% of everyone’s lives.
Erika: Aren’t you excited that we have so much work ahead of us to do?
Kristina: Oh my goodness.
Erika: This is why websites suck, because people are writing to be read, not to be interacted with. Everybody who was talking about writing for the web, that was a half-step, we have to actually think, okay, our website is having a conversation with somebody, because that’s the speed at which you’re operating. A lot of times there’s a really simple exercise that I do with clients when they get that like, “Oh my God, you’re exploding my world, you’re telling me ...”
We really had a lot of this when we were doing design systems to help move people to responsive websites. People would come to us, and you’ve probably had this exact same experience. People would come to us with completely unstructured content, and we’d be talking about a very highly structured responsive design system. They would talk to me about content migration, and I’d say, “Nothing’s getting migrated, you’re starting from scratch, because you’re starting from structure.”
My metaphor for this, is you have soup, you have content soup, and you need to make a sandwich. You cannot turn soup into a sandwich, you cannot take something unstructured and structure it. That’s so much effort, that you might as well start from what are we trying to communicate, in what pieces, in what channels? Start there, and then maybe you can go back to the old stuff for meaning, but you have to burn it all down and start over, because nothing you’ve done has any inherent value, and if you try to save it, you’re just going to spend a lot of sadness and effort trying to rescue your content swamp.
Kristina: What’s interesting there, is that you are essentially advocating for companies that are dealing with older websites, and I realize this is not what we were talking about a minute ago, but I want to jump on this.
Erika: It’s all connected.
Kristina: It’s all connected, it all comes back to Alexa. No, it all comes back to Jeff Bezos.
Erika: Who is listening to us right now.
Kristina: Yeah exactly, who is listening to us, and is preparing to sell us things related to this conversation. You’re talking about people who need to go responsive, and that you are advocating in fact for a full-on slash and burn of the website. There’s a whole other school of thought that says, “No, website redesigns must die, we need to be working with Agile and doing iterative improvements.” The concept of and the product of website for me is just so maddening, because I think it’s so critically important, I think it is the hub of so much communication.
I can’t see it going away, and yet everybody’s got their opinion about how important it is, or how to take care of it, or how to redesign it. Tell me what your website philosophy is.
Erika: My website philosophy is that your website is fundamentally important, but that has nothing to do with how much people use it. I think this is where people get lost, is they think like, “Okay, our website is important to the extent that, that is where people interact with us.” I say, “No, the website is important, because one, it’s often the conceit.” You have to at least pretend that you’re redesigning a website in order to get all the parties around the table to talk about organizational change.
Also, you have to have a website, we’ve talked about this as the single source of truth, right? It doesn’t matter if people are interacting with your website, because the best place for people to interact with you might be on social media, it might be on Medium, it might be through another channel, it might be through your newsletter. You have to have a website as that central thing often that you drive people to, or coordinate with, right? That is maybe the hub, I don’t know, that’s probably a loaded word.
You can’t say that just because it’s important, that has to be where people have to interact with you, and it has to be webpages. The thing that I was talking about, there’s a little exercise I do with people if they’re freaking out about like, “Oh my God, my precious content I’ve made, how do I start?” I say, “Okay, say you’re just sitting across from me at a table, and I’m your best prospect in the world. I’m your best perspective customer, audience member, whatever, and you have 30 seconds to tell me what I need to know,” right?
They can always do that, they might be totally like, “I don’t know how to prioritize anything, we have too much stuff, we have too many people, we’re still thinking about the website like spatial territory to carve up.” Every single person I sit down with, even visual designers who don’t think of themselves as professionals who work with words, right? If I sit down and say, “Okay, so I’m this target audience member, and you have 30 seconds, tell me what I need to know.”
Everyone can always think of the most important things to say, but if it comes to sitting down and typing words into empty boxes, people freak out.
Kristina: I freak out.
Erika: Yeah, it’s really like looking at the blank page, because the blank page isn’t something conducive to thinking about two living entities, and that’s what you want. You want even if somebody’s looking at your website, you want to create the sense that there are living, breathing, caring, ethical humans behind it. That’s where you start, you start from what would this be like if it were just two people exchanging information following that cooperative principle?
Then you translate that into, okay, then how do we capture that essence in this medium through this channel? You don’t start from, “Oh, we have some empty boxes on a website that we have to write for, and that we hope that we’re not boring people or confusing them or alienating them or insulting them.”
Kristina: I just want to tell you right now that every content strategist listening to this podcast is crying tears of joy at this very moment. They’re wiping their eyes with glee.
Erika: Excellent, my new book is highly absorbent.
Erika: No better way to soak up your tears of joy.
Kristina: No, and what is the title of your new book again Erika?
Erika: Conversational Design.
Kristina: Were I to want to procure a copy of this book, where would I be able to do that?
Kristina: Thank you, is the book available in Kindle or PDF format?
Erika: Yes, it is available in an e-book that works across I think it’s a DRM free EPUB, PDF, mode that will work on all your devices. If you get the electronic version it will work.
Kristina: When is the audiobook coming out?
Erika: Oh man.
Kristina: When is the musical version of the book coming out.
Erika: Yeah, there is a dubstep version that’s going to drop in June.
Kristina: I’m going to unplug for the entire month, and just pretend it never happened. Erika, where can people find you online?
Kristina: All the things, because you know what? You’re not content, you’re meaning, that’s why. You’re not device specific, you are the spirit that haunts my devices Erika Hall.
Erika: I hope I always will be.
Kristina: I do too. Thanks a lot for being here today.
Erika: Thank you.
Kristina: That does it for this week’s episode of The Content Strategy Podcast. I’m your host Kristina Halvorson. The Content Strategy Podcast is brought to you by Brain Traffic. Brain Traffic is a content strategy consultancy helping clients worldwide move from content chaos, to sustainable business success. Visit us online at BrainTraffic.com. Thanks a lot, we’ll see you next time.
The Content Strategy Podcast is a show for people who care about content. Join host Kristina Halvorson and guests for a show dedicated to the practice (and occasional art form) of content strategy. Listen in as they discuss hot topics in digital content and share their expert insight on making content work. Brought to you by Brain Traffic, the world’s leading content strategy agency.