Des Traynor is the COO of Intercom.io, a company that helps people communicate with their customers. In this episode we talk about the dangers of data, a framework for thinking about UX, and the role of microcopy.

TOPIC DES TRAYNOR COVERS

→ He is the co-founder and CEO of Intercom

→ Intercom is a communication platform for web products to communicate with customers

→ Provides tools to track user behavior and communicate with customers based on that behavior

→ The Goal is to provide targeted communication to the right users at the right time

→ Targeted communication to the right users is important

→ Over-communication can lead to users ignoring messages

→ Intercom allows for user filtering and tracking user behavior for targeted communication

→ Intercom is set up by adding a small piece of JavaScript to the website

→ Intercom is a communication platform for web businesses to communicate with customers

→ Can replace helpdesk, email marketing, CRM and marketing automation tools

→ Focuses on specific communication jobs such as announcing a feature or asking what’s going wrong

→ A tool for talking to customers in general

→ And a whole lot more

LINKS & RESOURCES

→ Des Traynor’s LinkedIn Profile

→ Intercom.io

WATCH THE INTERVIEW

READ THE TRANSCRIPTION

Bronson: Welcome to another episode of Growth Hacker TV. I’m Bronson Taylor and today I have Dez Trainor with us. Dez, thanks so much for coming on the show.

Des: Thank you, guys.

Bronson: Yeah, absolutely. You are currently the co-founder and CEO of Intercom, and that’s an Intercom IPO, which is a pretty awesome company that’s been mentioned a few times on this program already, actually. But go ahead and tell us from yourself, what is Intercom.

Des: So effectively in terms of communication platform for Web products to speak with their customers of any sort? So what that usually means is everything from talking to your customers right after they signed up either inside your product or by email to communicating new feature releases all the way through to dealing with any issues your customers might have with using your product. And if customers look like they’re slipping away, how can you talk to them to learn why they’ve stopped using your product? And frankly, you know, what we do is we provide the information you need to understand how our users are behaving, and then we provide you the tools that you need to communicate based on that behavior.

Bronson: All right. So it’s part understanding of what they’re doing. So you’re actually watching what your users are doing on your site. So it’s kind of analytics. And then the other part is just communication, which is where the intercom idea comes in. It’s a way to speak to your customers at different phases of their lifecycle after they take certain actions, just at set intervals. Kind of however you want to talk to them, right?

Des: Exactly. Yeah. So you’re totally right in terms of it’s called intercom because it’s basically, you know, the idea is like it boils and speak to your customer as you’re using your product, whatever. But doing data that alone leaves you blind to who you’re speaking to. So what you need to have is a sense of, you know, I want to tell people to use their reports generating feel more. Well, do you want to say or do you only want to say that to the people who are using it? And how do you know that’s even a problem? Well, let’s see how many people aren’t using it. Okay. Looks like 50% of the user base is using it. Let’s talk to them about it. Let’s not bartered it or 50% who are busy using it every day. I think what companies miss is like that over communicating the wrong message to the wrong people. So if you’re trying to upgrade people who are already on your premium plan or if you’re trying to tell people to use a featured are already using already already. So they don’t want.

Bronson: Mm hmm.

Des: What you do is you train them to ignore you and what ignoring it looks like, because it takes a few different forms. Some people just, you know, they are quite good at Gmail. They set up filters. They just say, like, you know, whatever this guy mails me trout into bin, other people just set up a mental filter, which is whenever this guy emails, we archive it on site or people unsubscribe. But if actually what you’re doing is you’re training your own customers not to listen to the messages you have, which is a really dangerous thing. So that’s why we always try to encourage people to like it is almost never a good reason to communicate to your entire user base like it’s maybe a once, twice a year type thing, but inevitably anytime something happens you like, Oh, we had an outage. And so they think that’s mandatory user base and like, well, let’s be able to go there on Sunday. Maybe they care. I don’t think any of us gives a shit, you know. Yeah. So it’s like, really like, you know, tell the right people the right message at the right time and you’ll get results if you do the opposite. What you get is like archive and unsubscribe, and ultimately you lose a channel of talking to your customers. The real danger there is, and this is a point I was make and you should talk to your customer as well, because if you don’t, one of your competitors will. So it’s often I talked to founders who are in a situation where they’re like, all these guys, let’s try for managers. And I’m like, You know what? They have an unsubscribe button. What are people saying? So, you know, you’re now in a position where it’d be like, Imagine you’re dating somebody and the only person they won’t listen to is, you know, like it’s like you don’t stand a chance, like, yeah. So, you know, it’s really like targeted communications are really important and you can’t carry a that data, which is why that part of the product exists.

Bronson: Yeah. No, that’s great. And all you have to do right. Is just put JavaScript on your site and then you can filter it on intercom. So on intercom you can set up those filters. You don’t have to set them up beforehand, right?

Des: Yeah, exactly. So you just basically it’s all a little bit of JavaScript like Google Analytics or any of those tools, and you choose what information you want to send us. So if you’re Instagram me and send us, you know, photos, likes, friends, whatever, if you’re like base camp, you send us like project teammates, messages, files. And the more information is that it’s typically the more visible it gets. But then we’re just, you know, once you have that information there retrospectively, you can now go and say, I wonder, has anyone ever not used our final feature? Well, just filter the files equal 0 to 6000 of them. Let’s find out why they’re not using it.

Bronson: Mm hmm. That’s right. Now, on your homepage, it kind of mentioned some of the things that Intercom can replace in certain situations. And here’s a list of them you can replace helpdesk, email marketing, customer relationship management, CRM Marketing, automation tools. Tell us how those how people are using that, how your customers are using it to really replace some of those things.

Des: So what happens within it? So I’ll just go go broad for a second and I’ll give you a specific. So typically what happens is when a, when a tool is brought into the workplace of any source, people use it to do different things and. If it takes off, lots of people copy it. And then what you have is what was a product turned into an industry. So you’re right, the email marketing industry, radio, and then the industry starts to define the tool, even though the tool to define it in the industry. And when that happens, people lose track of what the fundamental jobs that people were actually trying to do with the two of us. So people have in their heads that like, you know, a helpdesk solution has to have priorities and tags and flags and ordering and assignment and death and etc.. And that’s because that’s what they are there on top. Whereas if you actually really drill into the job that people are trying to do with a helpdesk or with email marketing or with any of these tools, you realize that an awful lot of the sort of foundations have changed a lot. So, you know, the perfect example would be the notion of CRM for a web business is different to the notion of CRM for a traveling salesman in 1995. But the assumptions and some of this entity, sensitive CRM, is killing any of the same things, whereas in reality what they need, like you have to do the same job, they actually need a different set of things these days. So in our case, what we’ve built is a communications platform for talking for web businesses to talk to their customers. Mm hmm. The talking to customers tends to fall into lots of different categories, but we don’t really think of it in that way anymore. We just think that in terms of what’s the job we’re trying to do, are you trying to announce a feature to your customers? Well, maybe you might use an email marketing tool to do that. Well, for us, it’s just that’s just an announcements to a lot of people. Conversely, are you trying to ask a customer what’s going wrong? Well, maybe it was a helped us to do that or first that’s just talking to one person. So we just specifically focus on the different jobs that people tried to do as they’re running a web business are how we make sure that we support them. And in that sense, we’ve kind of we’ve you know, people try to pigeonhole us into specific measures, but that’s not really correct. That’s not how we see ourselves. We are the tool for talking to your customers.

Bronson: Yeah, which makes even more sense. The name Intercom. It’s kind of the perfect name to sum up that it’s about communication. It’s not about one, you know, pigeonholed kind of communication. I’m talking about some of the case studies. What are some of the companies that have seen growth from it or seen some kind of benefit from the business? Talk to me about the sales pitch for a second.

Des: So that we feature a lot of these on our blogs. Our blogs are really popular. It gets lots and lots of traffic and we feature, you know, the best applications or best practices of. Intercom So I have a guy called Nathan. Nathan Company has a product called Draft, which is a tool for improving writing, and he basically increased his revenue by 200% just with a well-timed series of messages inside the product. So when people logged in, he’d remind them to upgrade or tell them about all the new features that he has built for them and say, Hey, if you don’t upgrade to the premium feature, it’ll work. And you know, he said, that’s a great graph that we featured on on the blog that just showed revenue. Like just, you know, as Peter Along has just exploded. And that was just a simple like part of this. And part of the reason that works so well is because we made a lot of these things easy to do, because if you ask any typical web business owner, would you like to be able to show a unit inside your product that has this new idea, which is, is that going on? A developer is going to be JavaScript in check across browsers, whereas when it’s actually a tree clicks and like typing one sentence, it becomes, you know, insanely easy. So as a result, people do it a lot more. So that was one example. A lot of people do clever stuff around support. So what they do is they, they do the Mecca of support as proactive support, which is you get it, you, you come back to your customer with the solution before they even report the problem. And this company called Circle, which do continuous integration, they’re called the circle name. And what they do is they realize through intercom that the biggest leading indicator of them losing a customer was a fail to build. So continuous integration is basically in, but the details aren’t really super important. But like one of their key thing Serco should be able to do a few is like build your product. And they knew if somebody signed up and couldn’t get a product to build tree circle, the whole thing would fall apart like they used to customer base basically. But I want to focus another part. So they were like, right, well what are we going to do this so that you have have a triggered intercom message about something, Hey, there’s a problem with your build. Click here to contact us which would again open intercom or read this dot page and it kind of builds or does emails I think. I can’t remember the numbers, but I think like they’ve, you know, they rescued maybe like 45% more of what would have been failed customers, purely biologists identifying an issue and creating a message associated it does that there’s plenty of other examples I’m just trying to take number guys called off zero damn they do it in a slightly more interesting take and support they recommend they’ve automated some backend processes to make it real simple basically and get it emails people any time anything is about to go wrong so they can step in and solve it. So again, you know, I really like a lot of the realizations are. Just around. You know, there are certain attributes of behavior that like that happen or that exist before something goes wrong and we potentially stand to lose customer. We can identify them and actor will keep the customer. And then there’s other use cases around like we have a great one himself. If somebody uses intercom as loud in more than 25 times, we just suggest they should follow us on Twitter. And by doing that, like we like our Twitter account went from like again, just real floppy to long connect this sort of hockey stick, almost all activity. And it’s just, you know, all of these. Ideas that we have, like, you know, their experiments effectively and a lot of them work out. A lot of them don’t work. But the great thing about it is you can just try to ten users. If it doesn’t work, don’t do it, you know, whereas previously before Intercom in the world I think it was like try it on, try to I try to carry you the rates and if it doesn’t work, oh well I’ll probably lose 1% of those subscribers. Whereas I, you know, now the punishment for, for a bad idea is pretty minimal. And the rewards for an execution of a good idea can be anything from like saving a lot more customers to like a 200% increase in revenue.

Bronson: So yeah, you know, it seems like for the sake of our audience here on growth of TV, it seems like communication is a growth hack. You know, being able to communicate accurately at the right time with the right message to the right person can actually increase revenue, increase visits, increase retention. You know, it’s a it’s a trial and error. You have to try certain things and see what works. But communication is really a huge aspect of growing a website, right?

Des: I think that’s totally true. And the one point I was making, it was like your goal. What would intercom are when you’re communicating to users? Most of the time you want them to go and do something that they’re like, Oh, here, use this feature, go here and upgrade, go here and sign up or whatever. But the other secondary goal, which intercom is great for to, is like if you can’t get them to do that, find out why. Because you’ve got to call the first part of growth IQ. I’d call this part of product talk, which is like one of the things I love about people who set up intercom Well, it’s like they’re always like they’re more impressed. They’re always impressed. We’re like, you know, we’ve got a 3% or 10% increase in this or doubted it or people are using or to only rent or whatever. But a lot of them come back and say, you know what’s great is that because of this auto message that’s like sending out every day I like have a drip feed coming into my company are reasons why people won’t use this and all we ever do when we sit down at the start of week is we say, right, what came back this week? Oh, it turns out people we have it’s a Firefox blog. Let’s figure it out. You know, they kind of I think I can’t imagine like not having this because we’ve had it for so long or for so many companies, the product talk of it, it’s kind of imagine it’s just every week while, you know, why do you sleep while you eat why you party intercom is busy gathering reasons why people won’t use your product or for the ones who will. It’s promoting them unprompted to do it. Yeah. So communications w growth hack where like the outcome of communication doesn’t just have to be a percentage increase, it can also be and we now have really, really good data on why people don’t do this or what. So it’s better feature or whatever.

Bronson: Sure. What are some of the most clever, bizarre kind of implementations that you’ve seen? Have you seen anybody use Intercom and just some really outlandish way to do something.

Des: As well to some interesting ones? I mean, if people use this in a weird way as technically that we wouldn’t have expected, like, you know, Intercom exists in a lot of browser extensions or, you know, we don’t have like and we don’t, we don’t have like an OS X client, but people have managed to somehow get that working through clever combinations of web views and stuff like that. And a lot of people use it to do everything is like so, you know, like Black Monday in the US or Cyber Monday, sorry, whatever it’s called. Yeah, sales day and like that totally caught me off guard because, because again, it’s, it’s not something that I don’t experience either here in Dublin or in general. Yeah. But like are like messages sent went through the roof, right. What’s going on and what it was was like ever an icy word was they got around to this is a good idea but ever this basic is your entire non active user base discount codes and saying hey I, I can use this product you know here’s a massive discount and I could just get would be like you know, two months free or whatever. But again, like it just they resulted in a significant bump in like signups and activity. And, you know, I always think of like Cyber Monday as being like a half price stereos and televisions. I never it never occurred to me like that. Like, you know, South UPS could like have an equivalent, like, you know, a party pack if you like. So that was that was kind of interesting. There’s other like, you know, what I look at and go, most are better than our customers. Most days I think that they teach us that we’ve basically created a platform for experimenting and we kind of stand to benefit so well and we try to distribute the knowledge back. But one thing was like a lot of people talk about retention in software as a service because basically the big problem is you can’t retain customers. You don’t have to solve this house business. You know, you’re basically going to die within six or 12 months or whatever. So retention is always an issue. And again, my argument is always if you can’t bring them back, find out why they left their best guys who’ve done to they’re like a music based startup in New York. And what they did was they have a message that goes that people who have been inactive for the past 30 days, but what it does is it includes features and screenshots and videos of everything they shipped within the past 30 days. So while the reason that works as opposed to, Hey, we miss you, please come back because that stuff doesn’t work. I can’t quantify it just if anything, it just in order to be further and but what’s good about you know did this take is a rather than just trying to nag people they’re actually giving them a good reason to market like so here’s five features we’ve built in the past 30 days. We think this will take you to your turn. If not, please tell us why. And they get like they got two great effects in our one. They start claiming back customers instead of patching up a leaky pocket. But secondly, and more importantly, they get to learn what has gone wrong. So people don’t realize that you don’t have the customers while you can write them off in a lot of ways. Joe is definitely like that. Every so often it can be one back, but there are really good source that lets you know what’s going wrong, either in your product or in your onboarding or at the highest level in your in your you’re marketing itself. So. Another example is we had a customer who was like, hey, look, we got like 8000 sign ups and only like 50 of them stuck around. What else got wrong? And I was like, Well, that is pretty important. So we went looking and what we did was and this is a very clever thing, people sent us the original referring URL from where people have signed up and what they realized. But this is a weddings website, a very popular wedding website, and they had basically to be in TechCrunch, which meant they got a obsolete shed load of new signups. Well, as you’ve guessed, the audience of TechCrunch dot com isn’t necessarily the guy that’s who you’re helping to plan your dream wedding. So by, you know, I mean, I’m sure like as I said, they got 50, but like 58, about 2000 is probably about right for people spontaneously looking to plan a wedding a year. And so those sort of things are just interesting like that. Like, well, we realized there was well, it turns out like, you know, if you do get TechCrunch, please make sure that you stand to benefit from the traffic. Because basically, like the sources of like a lot of like brides to be or couples looking to get married is in TechCrunch dot com it’s much more likely to be a weddings for and so yeah that was interesting too.

Bronson: Yeah. The quality channel matters. Now let’s talk about data a little bit. You know, I want to talk about because Intercom is all about data, you’re giving data and allowing people to act upon that data. But also you have some unique views on data because we have a lot of people on this show and they talk about how important data is. Why are you got to be data driven? But as I’m looking at your stuff online, it seems like you have a broader view of the role data plays in an organization. So I want to talk about that a little bit. When is data not the point? Because I saw a blog post you recently wrote where sometimes, you know, we present data and it’s not the point.

Des: Yeah. So there’s a few different angles they have on this. One of them is like my opposite background was, you know, I used to be a researcher in university and my area was like mixed method analysis, you combining qualitative and quantitative data and I kind of, you know, since moving into the tech industry and specifically into like startups, you kind of your people throw data at data all the time. And the first problem, what it is for me is that so much it’s just really it’s a really appalling analysis of basic, you know, what, what should be basic stuff. And people just fundamentally misunderstand it to like, you know, to an embarrassing level. But that’s like that’s only part of it. Secondly, I think that in the whole startup industry does not give enough waste to the value of like actually like talking to customers. They’d much rather analyze 65 data points, pick up the phone and ring them. And it’s just, you know, just so much is lost when you relegate someone’s experience, when it’s not a pure product to a set of data points, as in they got to follow step two. I’m like, Okay, Grant and they’ll make all these crazy assumptions about like, what went wrong. You know, we will. We need to work on our uploader better. Like, why is like, how do we know? Like, well, can security equal no problem to make? Did you want to upload? You know? So there’s that side of it too, which is like, I think they need equal footing. And noticing that it’s related to that is like the, you know, people and reverse engineering hypothesis every dataset which is just basically bad science, you know, it’s not the right way to do things. You should start off with hypothesis and see if the data backs it up, not look what the data says, then work out. You know, it’s okay to look what the data says and if it seems to be interesting, then go talk to the people involved and see if that matches the reality. But it’s the whole like confusion of correlation and causation. You know, when I see firemen, I see fire as therefore firemen cause fires, you know, so like it’s people just need to explore it out a little bit. I think when data isn’t, but basically what data is useful is when you’ve got like large volumes of activities and you’re specifically looking to optimize what I feel that most startup people they like optimized, premature and so like they might have, you know, maybe a thousand monthly active users and maybe they hear home page gets like, you know, 800 hits a day and they’re spending your time and testing. And I’m just like, you know, I just know that’s not going to continue in anything meaningful. But like, you know, at the very best, you’re finding like the smallest local maximum that you could possibly find. Well, I’m like it’s really like the piece that people don’t look at there as like, yeah, that’s but it is a 1% improvement going to say, right. But you have to like if you really believe in this approach, you have to ab test your investment in AB testing against you, entertains your product, you could be doing it, building a better product, talking to customers, do whatever, you know, trying to sell your product. You know, there’s lots of different ways to make it down, you know? And like tinkering with shades of red is only one of them.

Bronson: All right. We have to break that down a little bit. Abby, test whether Abby testing works, right? Yes. You know, there could be other other things you could be doing that have nothing to do with Abby testing that might move the needle in more significant ways. Is that what you’re saying?

Des: That’s exactly what I’m saying. And the area why? That is a symbolic start. So you know what? I would look forward to 6 to 7 people. Everything you do is opportunity cost me talking to you right now. There’s an opportunity cost. I could be talking to close to you know and so the idea that like, you know, like installing JavaScript and tinkering with experiments and learning different things is somehow a one way bet like it’s guaranteed to win is it’s obsolete. Nonsense. There is like, you know, most people like you’re like, right, there’s, I don’t know, exact right shade of blue or the exact right button alignment on the home page. But it turns out a slightly bigger red button works. I’m like, That’s what happens when you click, Oh, we haven’t got to the center for a minute. Shit. And they’re like, Oh yeah, like this. Like in, like what I, what I specifically mean as Abby when I say advertised Abby testing what I mean is like you only have a limited amount of resources and you have to question the aura of pursuing. Abby testing versus your way of improving your product. Writing a popular blog post and like trying to direct sales and people, you know, go and speak at a conference like, oh, these like everything you do is an opportunity cost for everything else.

Bronson: Yeah.

Des: And I’m like, for some reason that gets cast aside when people say we invested 44 areas in Abby testing and we got 2.5% increase. And like, first of all, that’s a positive result. And I’m like, Yeah, if you got like 1.5 million users, that is a positive low and you’ve got people who are dedicated. Abby Testing people. Yeah, that’s a positive result. For example, neither of those things are true.

Bronson: So yeah, I know, it’s great. I mean, just hearing you talk, it makes me think sometimes, you know, we get so deep into the product and so deep into the numbers and so deep into the datasets, they forget just to go outside, get some fresh air and use common sense, and then approach the product just, okay, I’m trying to move the needle in significant ways. What can I do to do that? And the answer may not be whatever bells is telling you to do, you just have to approach it with a lot of just groundedness. So, so I love your attitude about it and it fits in with the philosophy of Intercom. I mean, obviously communicating with customers sometimes and many times will move the needle more than Abby testing. But I think a lot of products, people are afraid to talk to people, so they go to the data instead of going to the people, right?

Des: Yeah. I think people try will always try to add product. And I did McKenna had a sort of epic Twitter around about this. Like, you know, the product is like deified way above anything else, like distribution or customer communication or any of those sort of things. And that people just think like, you know, like there’s just some truth and some evidence that like you can’t build a product focused company that is 100% focused on product that has no marketing department. So like, you know, there’s some anecdote that they’re able to get that like if you look across all the great APIs, all that, it’s certainly not true. Like there’s not there is an excitement that all of you need to do with a product. And I like, you know, as a simple like, I mean the classic case for me is that people are like, alright, well I’m trying to get people to try to get my users to use this new feature called Revenue Analyzer. I’m like, okay, ground by the way. So I’ve tried these nine different types of buttons and I’m also trying this and I’m also turned down the like, okay, why don’t we try putting a message inside the product to people who often use it saying, please use it, and people I know. Or I’m like, Yeah, well like is there a goal here to have a good AB test or is our goal to achieve an outcome? Once you start with the actual outcome you want, like you realize there’s lots of different ways to get there and you know, don’t be don’t like biased your view just because you’re most comfortable with like with JavaScript and you know testing tools.

Bronson: Yeah you know it makes me think about this. One of the things I’ve learned is how important copywriting is that sometimes just saying something really clearly matters and that it’s not about all the things around it. It’s just they come to the site, what’s the headline, what’s the subhead, what’s the call to action and intercom? Oh, almost takes advantage of just the latent power of words themselves. I mean, what can do more than the word an English language, if that’s what you speak like? That’s that’s their communication to all these all day. Every day is their own language. So why not give them communication in the way that they’re used to it and then move the needle?

Des: Yeah, that’s that’s ultimately it. Like, for me, it’s just, you know, you like there’s no like, you know, the more care directory it gets more surprising, the better results you get. And as I said, like, it’s not always necessarily great results, especially responding to like both sides work for like the worst case, like, you know, worst case scenario if you try to talk to people and convince them to use your new feature and they don’t is you leave like a standing where they won’t. The worst case scenario you try to ab test your way toward the landing page that little get people do use in a feature is that you basically have a lot of messy data, no significant results and you know new advisor yeah or best case you got a 2% increase or 3.2% increase and you know, the Greens are slightly better off green or whatever, you know.

Bronson: Yeah, absolutely. Now on a similar note, you’re an expert on micro copy, which I think goes hand in hand. All the stuff we’re talking about right now. I mean, giving presentations or webinars on micro copy. Tell us real quick, what is micro coffee?

Des: It’s kind of a term that was introduced to describe like coffee that is specifically used in enterprises such that it has to be abbreviated. So it’s the opposite of long form coffee, which is like, you know, imagine writing a paragraph that fits on a book. Well, not to be ridiculous. So you kind of you know, it’s an acknowledged micro copy of the term as an acknowledgment of the fact that you need to be brief, concise and clear in UI that for a variety of reasons. So it’s it’s almost like a new dialect of English, if you like.

Bronson: Yeah. And why is micro coffee so important? I mean, I believe it is, but. But tell me, why is it so?

Des: Most of the time when people get stuck in a UI, it’s because they don’t know where they are or they don’t know. They can’t find out how to do one of the things they want to do. And most of those problems are best solved with words. And the reason it’s important is because if you kind of look at some of the lines in a product and their own experience is sort of is, say, just icons or just colors or like really unclear wording, such as like, you know, dark matter or whatever, like just crazy words that make no sense that they’re not going to know what the hell, what really are, and they’re not going to know what they want to do. And, you know, there could be a lot of examples like a a button would say a magnifying glass. And it is that surgery is on zoom. You know, like people have like, you know, just some things that, you know, aren’t obvious to people in general, but they’re things like geometry icons that universally get 100% recognition by software users are followed, O’Callaghan underlined. I looked at data before, but in general, like the reason Microsoft is important is because if you want to say something to user with absolute clarity, you have to use clear language. And that is going to be Microsoft being an enterprise. Yeah, and there are there are a lot of reasons why you fall back to icons, for example, like people kind of know if they see a pattern next to something, it probably means that it people generally know to actually see a closed beside something. It means odd or like out of nowhere or whatever. Like. So I like esthetically. That’s a much more pleasing interface. And you’ll see, I just, you know, I’m not against that per se, but there are areas in a product where party is more important than the static and in those terms feel prompted, tend to copy and you’ll really want to take your time thinking about who who you’re talking to. What’s the message? What’s the best way to say it? What information do they need? What’s the next step, given what you’re about to say to them? Because, you know, I’d like a copy of things like error goes up, right? And you’re like, Well, what is the problem? Have I fix the problem? You know, because I like when it’s only when you put yourself in the mind of the user, your eyes don’t just say error, tell me what’s going on. Or another classic is things like, Yeah, and we think a disabled button that says something like Merge, you’re like, What else is new? And why can’t I click? And you’re like, Well, you know, that’s just kind of, you know, a better wording. And like, maybe if you show how to tooltip an obvious thing, why don’t you select two people before you can mergers make you think that that’s where a copy is key. It’s just a help to user at every stumbling block along the way.

Bronson: Yeah. Now, let’s talk about UX kind of in general, because you’re UX designer, you know, you’re the CEO right now, but you do a lot with UX before and in general. And on your homepage, you outline kind of your process the way you think about U.S., the way you think about the process. And I think this is going to be really helpful. So I want to go through these five bullet points real quick here. Sure. The first thing you say you do is, is that you have to understand the domain the software is targeting. What do you mean by that?

Des: So, I mean, if you’re if if you’re not a football fan and you’re trying to build a football product, you’re basically in big trouble. And because if you’re not like if you don’t have a good familiarity of of the domain that you recognize specifically, like the actual knowledge that if your product targets marketers, you need to know a lot about marketing to speak their language. So this is most common to me when people like, you know, you see like people trying to build a product because they kind of half saw, oh, you know, how to turn their restaurants don’t really have a good solution for updating your menus. I’m going to build one of those. But like, if you’re actually if you’re not solving the problem that you experience yourself or you have to research gravy, you’re really in trouble. So the first piece is to me is like, you know, I would never take on a product of honest or serving of your consultants. Okay. If I was told, right, we’re doing UX consultancy for an online gambling product, I’m like, okay, I am going to go to a casino like that space or I’m going to go to a sports betting shop where I’m going to try and learn what the hell is going on here, because otherwise I won’t know how to write the right copy. I’m going to call things like confirm saved, you know, write it on like, you know, like, you know, place that or whatever it’s appropriate. Like, so, you know, it’s just, you know, you have to have the you can’t research it on the fly. You actually have to be one of the people you’re solving the problem for.

Bronson: Yeah. You know, it reminds me of a great essay that Paul Graham wrote where he talks about, you know. You have these ideas that you think are really good in terms of a business you could start. But yeah, good ideas are always businesses unrelated to your domain of expertize because you don’t know enough to know what they’re really bad ideas, actually. And the ideas you have in your own domain you feel like aren’t that great because you know so much, you know how things really work, but really, you know, you have to go with your own domain or you have no chance at all.

Des: That’s another 100% agree. But I do think you can you can change parts of who you are. So, like, if you’re 25 and you know nothing about soccer but you want to build a soccer app, you can get interested and you can study and hang around with the right people and go to the right buyers. And, and yeah, then you have a chance. But it has to be that level of commitment. It can’t be. I’ve got a friend. I’m going to ring him every time I get stuck. It has to be like, no, I’m the user.

Bronson: Yeah, absolutely. And then the next thing, when it comes to what you need to understand to do, you actually talk about the customers the application owner hopes for to talk us through that a little bit.

Des: So that’s just really you have to go to understanding of the people that you’re selling to. And that relates to the two sites that are one of them is a distribution like how can we get these people? So like will YouTube ads be a good way for targeting, you know, soccer fans, is the question. And if so, you know, watching their first experience be like but the key thing is just if you don’t, you have to have a really good understanding of what sort of people you’re actually designing for. That isn’t distinct from the idea of personas, which I’m not talking to here. What I’m talking about is like what is the behaviors that are likely to work with these type of people. So perfect. That would be like if you’re designing, say, a gambling product, you might naively think, Oh, we’re going to build a social network where they can all make friends with each other and livestream all their bets. But in practice you’ll realize that like people you actually best, you don’t speak about your bosses ever, and they only occasionally talk about their winnings, but they only talk in hindsight. They’ll never tell you that that they have. And this is all stopping Eric for one project. So you start to realize, well, you know, whatever that realization, it would make perfect sense to build these activity feeds and social networks inside the product. But in hindsight it’s like that is absolutely necessary. Like, you know, if I it’s actually if anything, it’s going to disturb or turn people away from a product. So unless you actually have an understanding. So part of it is like you have to be the type of user. But the other part of it is you have to understand the dynamics that actually happen within groups of people so that you can actually design for them.

Bronson: Yeah. And then the third one is understanding the technical constraints that the software must work with then.

Des: Yeah. So this is like this is we used to or in my previous career I was a consultant where like we were just actual bonafide UX people. I mean certainly at different times for different projects, you could really, really damage yourself if you were already contracted to build something by also going and designing this masterpiece that would take years to implement. Like, you know, let’s go again at any point, at any point you want, you can click it. It’ll change the color of the video and you’re like, We’re going to build that. So it’s really like it’s understanding that technical investment is limited and you have to choose where to invest it wisely. So when I see a startup and it’s asking me to sign up and it’s asking me for my data for it, and I see that they’ve got this fancy natural language. They picture what I look at. I sort of think, well, you know, maybe this is a jQuery plugin, maybe it’s not. But regardless, they did not need to invest here. This is a big investment that has to be checked across so many different browsers to make sure it works. And there’s going to be a lot of like overhead and styling it make me look good and all this and stuff and it has to be built in the first place. I’m tested for all these weird edge cases that I have to support. That’s a real waste of money as opposed to three dropdowns in a row. And but as a designer, if you’re just chasing what looks most impressive and you’re just busy trying to build off of your dribble points, you’re going to go for that. But as somebody who is actually is concerned about the product getting built, you have to actually understand, I’m not going to ask you to do that because that’s a waste of time. However, I do want to spend is over in this odor area, which is like where people edit your profile photos because we’re dating site and not really matters. Conversely, like if you’re like you know fly fox or what I’ve done is like you have new airline sites or chipmunk or whatever your entire product is basically your calendar plus your results here. So yeah, if you want to spend 5 minutes building the best there. Absolutely. That’s your product. You know, if people can’t do that, amazing, then it’s no point. So for me, it comes down to like, you know, how important is this feature? How much technical ask and how much of a technical ask is what I’m trying to do and then balancing those against each other.

Bronson: Yeah, those are great insights. And the next, when you talk about understanding the real world data that will be presented within the application, is that just really getting a handle on the content so you can design around it?

Des: Yeah, like said, there’s a lot going on there. So part of it is like. It’s the classic example of the problem, which is like if you’re designing, say something like a blog, you want to go over which article. The images of the paragraphs are written, but like to remember the railroad data for the person who’s actually writing the blog is all bullet point lists of tables and order lists and, you know, wireframes and videos and you’re like, Oh shit, if I hadn’t done that, I would have decided a little bit differently. The classic for me is always and you can see this across the table is like, you know, the stats dashboard where like every single page is like, you know, it’s like so exciting looking down on, Oh, we got everything, so everything’s there. And then you see this design and the reality is just like flat you had bought like yesterday, you know? Yeah, I think, you know what our exposure to the actual data that you’re designing for right here, basically you’re designing with your eyes close to the origins as well. Like people like can see like wireframes and PDF files in this like fancy land where like every paragraph is of equal length and headings never. And you know, like every piece of text is conveniently the exact right size of its bounding box. And you know, and another classic is like photo galleries are type things like so if you’re trying to say design a whole page for like a restaurant is like, okay, cool, we want to use flashing images, let’s do a carousel and whatever else is popular. What is actually seeing the data or the content that’s going to come in? You can’t make an informed decision. So the question has to be, show me a good these photographs are because if are stunning, they’re going to be the main feature of the website. And as shift, we’re not going to keep on the website. And after some ways in between we might have asked about some secondary page, but like you can’t actually design what I know what’s coming. The problem is that people go for is a cookie cutter template of okay well I’m a big image because that’s what everyone else is using, what’s going to be in image? And you’re like, well, I didn’t get like some of, you know, some shitty like, you know, Android 2.0, a photo of a steak that’s like burnt, I guess. But that is the leader of a generous website. But yeah, so that’s really, it’s really important to me for like, you know, what sort of, you know, what content are redesigning around and what kind of redesigning for.

Bronson: Yeah. And then the last thing you talk about is understanding the tangible, measurable goals of the design. Is it just having the end in mind and designing with some kind of goals? You’re not just designing for dribble points like you said.

Des: Yeah. So honestly, it kind of touches on a lot of different things. But like if, like, you know, if you’re designing like a, you know, Sandy message for every year, you know, it’s naive to not look at, you know, the throughput of that form from a message point of view. And, you know, it’s almost a counter to like the the points. I mean, it’s kind of it’s the celebrate side of design, which is that like your goal ultimately is to enable behavior that’s like that’s really interesting, all of that. It’s like people are here, the solution is here on your interfaces operations. And if you released an interface that like, you know, causes allows people to get across, it doesn’t matter how beautiful it is, it doesn’t work. And, you know, we used to have a problem where like, you know, most of the way was okay. And as a result, people couldn’t get free to be in a very easy with like, well, it would be like the projects failed for people to be able to file your form. But like these days it’s actually the flipside of the problem is almost, you know, people are deifying design like just raw beauty, pixel design like shadows, fonts, typography, layers, flat design, skeuomorphic design, whatever. They’re focusing so much on down on, not on. Like, how does this thing actually work? And so we’re talking to the hospital. There’s just what’s the definition of work here? So I’m like, you can really apply this out of like a sort of a microscopic level. Like so if you’re sitting down to redesign the search box and like what is your actual outcome of that? Well, are people finding what they want to do? So maybe your measure is no, repeated search is per person or whatever. But like you, you incorporate some idea. And then once you actually have the idea of like, we know exactly what this screen is supposed to do, so all you need to do is find the best way to do that. And then that kind of rises above the idea of, Oh, well, checkboxes, look over here radio of the browser, or why? Why can we just write it down? It’s just literally like of the people have it here, how many you can get there. And that’s really like whatever done you can produce works of art or terrible design or something that is actually both. And no, I have actually a quarter. Don’t you get succeeded as designer?

Bronson: Yeah, it kind of takes me back to what you talked about with the mission of Intercom where, you know, you have people on your site, they have something they’re trying to accomplish. And what stands in the way is the interface. And it’s a bridge trying to get them from point A to point B and intercom is kind of a tool. It’s like the tour guide walking them across the bridge. Hey, try this. Go here, do that. It’s just one other element of the UX of the interface to help bridge that gap a little bit, right?

Des: Absolutely. And that’s how a lot of people do use and especially people who are in a small number of large valued customers, they sort of their attitude is like, look, you know, here’s the interface. We’re pretty sure it works well, but anything you need, click this book and we’re here and like, you know if they provide a great it’s great. It’s kind of like a white glove service almost for like to refer to their groceries. And I heard that, you know, some people say, oh, well, that’s like a commodity as a crutch for the interface. But that’s not really true either. It’s it’s just really like. So customers demand more attention and if they’re more profitable, that’s okay. You know, it’s it’s not everyone’s first choice, a business model. But like, it’s, you know, it certainly makes sense to, you know, for those who use it, it is they’re really happy with doing it that way. But yeah, you’re right. Like when you it’s it’s more way to reinforce to bridge people from A to B is to make sure you’re there to answer the problems or be proactive with support or not.

Bronson: Yeah. Now you’re also a regular conference speaker. I’ve seen you online in a number of talks at different places. You have your speaker deck and, you know, things like that. And one of the presentations you gave kind of caught my eye and it was called Simplify by Orders of Magnitude. I love that title because I feel like that’s the goal of my life is to simplify by orders of magnitude ever possible, just remove a zero. But what role does simplicity play, do you think, in growing a startup? You know, you had your consultants that you talked about before, in contrast, I think is the name of it. You have intercom i o now how important is simplicity to growing something.

Des: It’s pretty high level question what’s I mean like I think Jack Dorsey is a great quote, which is a perfect every detail, but limit the number of details. And I think he’s almost something there which is like, you know, whatever you do should be really good, but you have to be fussy about whatever you do. And so that’s one side of simplicity, but it’s almost that that’s kind of tends towards minimalism rather than simplicity, I guess for me, like, whereas simplicity is of, you know, for some people or some businesses specifically, simplicity is their entire business model. Like, so, like Instagram, for example, made it super simple to share good looking photos. I think like if you think of the world pre Instagram well I mean when I say prints I’ve already been pre mobile phone on picture things because I know they weren’t going to first but they were the best. And like that world was like carrying around a, you know, kind of rumors or shit like that being connect to a laptop and it’s, you know, important as follows like the c000004 JPG You know intro to Photoshop to crop it rotated to looks the right way. Then you go to the Flickr and then you mail three people to say, Hey, look at that photo. You go to the Flickr. Yeah. And and because of that, because that was the workflow, people didn’t do it. And this is like, you know, the point of that presentation was really that when you simplify something to a large degree, you actually get like, you know, it’s like a butterfly effect. What seems like a small amount of sophistication actually causes a massive behavioral change. So the best example of how to build books and how to perfect everyone was an extensive research principal at Microsoft is fantastic but he made a point is there is a lot of history is that I went when televisions did not have remote controls every station was free to run their own scheduling at their own areas whenever it suited them. The very second the remote control became popular, they all have to think about their schedules. The reason for this is because everyone could switch channel but no one was bothered getting a better future to do it. When you give zero control the very second years televisions switches to operates, they change to another channel and the original channel program to watch that instead. So every television company realized, Shit, this is a big problem. We’re like every single time we go to ads, we lose our customers. They’re like forced to and they realize this isn’t going to work. You know, if the entire population can avoid watching adverts on television, we are in trouble. So they had a meeting and they all basically agreed worldwide that everyone had to think of their their TV schedules. So the ads are on the half hour or on to 50 minutes on the air. Yeah. And and that is a perfect example to me of like a small changes, like they interrupt you. I wasn’t bright proprietary technology. It just simplified something that everyone could already do. But because of simple, it became something people do all the time and because of the entire ecosystem have changed. And I dare other examples like, you know, I feel that Instagram was a good example of that. Like it’s been that our business built off the back of being simple and streamlined and there are plenty more, you know, even like you know just flows the popular but evolved into consoles vine could have we talked about it but like just if it’s easy to talk to see who is using your product to talk to them. If it’s actually easy you’re going to do it. If it’s tip a developer on the shoulder, pull out a CSV of email addresses, import down, introduce that a campaign, select. All right, email creative thing. Select a template from the gallery. Hey, guys, how’s the product counts and wafer? It replies you’re just not going to do it like it’s, you know, and that’s it, you know, it’s simplicity and like, how’s it? Multiplicative effect on, like, the likelihood of you actually doing something?

Bronson: Yeah, no, I think it a great point on simplicity. This has been an incredible interview. I have a couple final questions for you here. One, what’s the best growth path that you’ve ever implemented, something you’ve done in the product or something done with marketing, something that you’ve done as a team that has really moved the needle for you guys.

Des: And it’s funny, I don’t think in terms of grow talks about that answer the question. So for me, it’s okay.

Bronson: You can hate the word a lot of people. That’s fine, but I don’t know.

Des: So I wrote a post on my blog that said, If it’s important, don’t like it, which is when I when.

Bronson: I use hack, I don’t mean hack it in that.

Des: Way, but I know. And so in.

Bronson: Real.

Des: So like I feel like what people are currently afraid is going to fall into two different categories. One of them is like a really good design idea and the other one is a kind of shady. Maybe you should really do this in your product sort of idea. And, and then the sort of the delicate ones are kind of halfway on the bridge between both, you know. But I got some stuff that like, really took off that people really valued for us. With them, we found like anything, anything we find ourselves doing tends to leak its way into the product. So we have an intercom where we every day I want to see like in our early days every day I wanted to see who signed up yesterday and it was really exciting all the time. But because like when you’re really starting out, you’ve got tens or hundreds of post which you like well. And so we built this little email to send what’s an email every day and here’s you signed up. And then we like, Oh, let’s make an option for everyone else. Like Larry exploded. We got into the most valuable thing I’ve ever gotten. And the and that was a real shock to us because, you know, we thought it was cool for us. But like, I guess we should have figured it was definitely cool if I felt like the response was like even greater than we’d hoped. And then it became clear what the solution to that. And so we started it up a bit and get a lot more a lot more call to action. The knock on effects of all of all that work really was at the People and in Intercom much more often. And I think to install it was still kind of on the fence came straight back the following day when to get the email thing and 35 people signed up or whatever. So that was that was one example where it’s just like, you know, if you like, a lot of startups work for people in the background. So and by that, what I mean by that specifically is especially when you get into the world of like, you know, like developer tools and of they’re actually doing things independent of independent of user action. So it’s not like you don’t have to be in there typing or whatever. And I feel like, you know, one good thing here is just to remind can’t consistently remind people of the value that you’re delivering and gives them calls to action to return you to your product. Like that’s the sort of the abstract lesson here is like whatever your product goes for me, you know, you should be able to like summarize down on a weekly basis or whatever. Let me now add up the value you’re delivering. And if you can associate that value with calls to action, the chances of me coming back in is infinitely higher. So that’s kind of, you know, that’s where my high level keeping up with the low level tables just, you know, they report.

Bronson: Well, no, that’s great. Last question for you here. What’s the best advice you have for any startup that’s trying to grow.

Des: And try to grow in terms of users are just.

Bronson: Yeah, there either way maybe users, maybe revenue, maybe they’re trying to add employees, they’re trying to, you know, they’re trying to get bigger or whatever that means to them.

Des: Yeah. So like, I guess if, if it’s like trying to have revenue, you know, for first port of call is always be pricing if it’s churning out customers. I think the biggest thing is to work on what you’re good at. So if you’re great developers, you don’t want that, but not great writers. Don’t try a blog just because you saw someone else do it. Because if that’s not your forte and it’s it’s really a dopey point, but like focusing on what you’re really good at, such a great developer has build open source to the company your product and try to meet their people and maintain them. Know, just build to the last to get up and forget about them. Give them a nice site. They do what you can and actually support it. If you’re a great writer, do a lot of writing. If you’re great visually, do a lot of like give a lot of design examples of readings. And that’s how you basically attract, you mean to such a limited space for attention underneath that, like, you know, there’s just no market for suboptimal stuff. So a shitty blog gets no traffic basically like or you know, it’s like, well, I’m not going to tropical for different reasons. Like they’ll have like scandalous or sensational posts. And, but I think in general, like, you know, every I’ve never met a good product that doesn’t have people behind that you can market. They just don’t realize that what they could do with marketing. So like it might be like, well, why don’t you interview everyone else in your industry and you know, okay, well, I guess I would try whatever. But what you need first and foremost is attention and amount of attention hopefully to move into like activation, acquisition, etc.. And internally, I mean like in terms of growing teams, I would be careful to make sure and be thoughtful. I think there’s a limit. I mean, kind of gets an edge of sort of stuff there like but you can actually advise people. But I think like most startups, their concerns aren’t in terms of hiring. It’s usually in terms of revenue, which is I opposed to a blog called for principles of pricing. You should never forget that kind of that goes true to. Points are always the same. It’s like, Charles, you’re a good and you’re comfortable with Charles Board. You’re comfortable with planet changing your prices. Like they’re just, you know, the things people don’t realize about pricing. Mm hmm. So that’s the money talk to, like, get your pricing right and find your most effective market channel and go nuts.

Bronson: Yeah, that’s great. Well, thank you so much again for coming on growth after TV.

Des: No problem. Thanks a lot from on the.