All posts

UX Research Topics


Field Guide


Thank you! You are all signed up.
Oops! Something went wrong while submitting the form.


BlogAwkward Silences

Creating Better Help Content with Andrew Sandler of Adobe

We chat with Adobe's Director of Innovation and Experience Strategy about what it takes to create great help content and the ROI it provides

Carrie Boyd

When Andrew Sandler joined Adobe back in November, he faced an interesting design challenge. How do you manage a vast library of help content, spanning tons of different products in different industries, so that users can easily find what they need to know to fix their problems? Andrew has learned a lot about leveraging the power of community to problem solve, experimenting with different formats to make technical explanations more accessible, and proving the value of great help content.

Erin and JH chatted with him about how he positions help content to stakeholders, tackles creating content for products that have evolved from box software to the cloud, and uses research to focus on the right things at the right time.


[1:43] Great product doesn't need as much help content, but people will always need help, and the more powerful your product is, the more help they will need.

[8:53] By connecting customers with the right information through communities, they're 3x less likely to reach out to support for help.

[10:59] Early engagement = better retention, so Adobe segments out its customers to focus on what gets them started.

[13:35] Adobe has segments and chapters to keep everyone on the same page and communicating well.

[25:20] People who get value from help content actually end up having a higher lifetime value. It's all about trust and mutual respect.

[27:41] How Adobe creates help content for different languages and cultures.

[32:09] How Andrew is thinking about simplifying solutions, and making help content work smarter, not harder

[37:58] Help content and product can work together to create even better solutions for users.

[39:32] Quantitative information can tell you what some of the issues are, while qualitative can help your team dig deeper into why they're there

[43:03] Building out recommendations for other things you may be looking for helps your help content build a story for the user.

The best stories about user research

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.


Erin: [00:00:30] hello everybody. And welcome back to awkward silences. We're thrilled to be here with Andrew Sandler. He is the director of innovation and experience strategy at Adobe.

Welcome Andrew.

Andrew: [00:00:44] Thank you. Thank you for having me today.

Erin: [00:00:46] We're going to talk about the UX of help content today and content as an experience and providing a better customer experience through better help content and all sorts of good stuff like that. We've got Jade here too.

JH: [00:01:02] Yeah, this seems like a fascinating one. I feel like it's often an afterthought and it seems like it probably shouldn't be so excited to, to dig into it.

Andrew: [00:01:10] Yeah, I'm really glad to talk about with y'all.

Erin: [00:01:13] Fantastic. So let's, let's dive into DHS observation. There is help content often an afterthought. And how do you think about it?

Andrew: [00:01:20] Yeah, so it's really interesting because my background is in product design night. Most recently, it was at a healthcare startup for two years as head of design and research. And before that, you know, I've spent time at Walmart and Groupon and eBay and all those places, you kind of think about the impact of the product and how we solve it.

And what you realize is two things. When you build great product, Often people don't need a help, but the more complicated it gets, the more people depend on you to not only get assistance, but really learn the thing that you're doing. So when we at Adobe think about our health ecosystem, it's a lot of learning content because the tools have gotten really complex over the years and it's, you know, just basic assistance.

How do I download? How do I pay my bill? What is this charge on my account? Things like, you know, The software that I bought eight years ago for a previous operating system, isn't running on the new one. What should I do? So I think that there's a piece of it where we can partner with our product teams to build great products and reduce the dependence on help.

But there's also the piece where, you know, people just really need assistance. And I don't know that it's an afterthought, but I will tell you that when my boss showed up in our org four years ago, we didn't have a well thought out. The customer experience strategy. And she was really instrumental in putting that in place.

And I think we've actually seen a sea change in how people can engage with us online to get the information that they need.

Erin: [00:02:43] I like that. Yeah. So the, is, is the goal to make help content unnecessary? It sounds like no. Right? Cause I think there's this to go along with that, after that idea, this idea that, well, if the product's good, then you don't need help content. You're saying obviously the more complex the product, the more there is to know about it, but that, but that people are always going to need help.

Andrew: [00:03:07] Well, I think, I think that's true, right? I mean, it could be something as simple as look, you know, hacking happens and all of a sudden your credit card gets compromised. And the thing that you had for automatic billing doesn't work. So it might be as simple as like, where do I go to change my payment options?

That's, you know, a type of help we can do. It could be, you know, Photoshop has this really amazing AI system called sensei, and it can help you do auto edge detection and that's brand new and nobody's seen it. So how do you teach somebody how to use that thoughtfully? So you could make the product really, really easy that the more powerful it gets and the more different things you want to do with it, the more you probably going to want to learn how to do that.

So there is a piece of which is like, teach me how to use it. And there's a piece which is like this thing just isn't working. 

And so I think that both of your statements are true. Yes. You want to make the product super intuitive, although that's such a loaded term, but you also want to make it so that for the people that are coming up to speed or learning it or learning something new about it, they have a place to go to get that information really, really easily.

JH: [00:04:11] Hmm, what comes to mind for me is in the agile manifesto, right? They have working software over comprehensive documentation. But I think the way they phrase that is really smart. Right. Because it says it doesn't say working software and no documentation. Right. Like it, like they, I think they just say like, prioritize these things first, but there's these other things are still part of it.

Right. the other ones, you know, individuals and interactions over processes and tools, it's not saying no process, no tools. It's saying just prioritizing the people you're working with and stuff. And it feels very similar in this regard. Right? Like it's, you want the software to be intuitive and easy to use, but like, There's a role for documentation and you can play that role poorly, or you can play that role.

Well, I think is what you're getting at. Right.

Andrew: [00:04:50] Yeah, I think that's right. And then the really interesting thing with Adobe, and I've only been there since November full transparency. is that. We started in a world where we were shipping box software. Right. You go down to your local, you know, and in our case, on the West coast fries and you pick up.

creative suite or Photoshop off the shelf and you'd install it. And they come with a manual. So a lot of the people that had grown up kind of writing for Adobe had been technical writers because we shipped boxes of software with instructions. And then when we pivoted to cloud, all of a sudden, this is a whole new set of skills that you got to develop.

And now we're one of my areas of responsibilities is. With automation and thinking about that from an experiential standpoint, and how do we think about using thoughts both to route you to the right place, but potentially to even solve your problems online. And those are bots both on our hope platform, as well as social buzz that you might find on Facebook or Twitter.

And so the skill set that you said that technical writing skills set is often just a little bit I don't want to say off the Mark from what we consider to be great writing today, but it's, it's definitely a relearning experience for a lot of the people on my team to think about. I used to write for box software and now I write for the web.

And I used to write for a really mature professional audience. And now, you know, when you think about things like Adobe fresco, which is on the iPad that you can use pens on, it's a whole different demographic, the whole different set of expectations, right? This isn't the desktop publishing world. This is really creativity for everyone.

  Erin: [00:06:26] What strikes me as you're talking about your. Users, right. Your readers and people who are needing help and needing to learn. you talking about a lot of different kinds of people. I mean, given Adobe scale, By definition, in different States in their life cycle, right. I'm trying to accomplish this.

I'm in this state, I'm this kind of person I used to buy the box software. I never bought the box software. I'm on this device, you know, et cetera, et cetera, et cetera. how do you go about sort of identifying all of that and do you use existing sort of frameworks from the Adobe product? Or how do you figure out who you're talking to, how to talk to them and what they care about?

Andrew: [00:07:10] That's a great question. You know, I mean, one of the really fascinating things, and we finally started digging in, we had, well over 450 million visits to our health properties last year, and it was almost 300 million people that came. And about, you know, over 6 million then started some online conversation with us.

And that's a massive number. I mean, that's almost a half a billion visits in a year, right. So our guess is we're probably talking to just about everybody. it turns out that, you know, when we looked at our data, it looks like over 50% of new users will get in contact with us in the course of a year.

That might be a bad sign, but we think it's actually a pretty good one because they're consuming to a lot of. To a large extent, a lot of our, what we call our learn content. It's the tips. It's those tutorials. It's the how tos. A lot of them are really, really well produced videos internally from one of my sister organizations that happened to be hosted on the platform that we have responsibility for.

We also know that, you know, we built a whole community platform last year, which internally we've codenamed Alfred, but it's really a platform where people can come and meet. other creatives that are going through similar processes or ask questions. And the fascinating thing is when people don't get in touch with us at all, through any of our web properties, they tend to.

Contact us at almost twice, the rate of those that do, and almost three times the rate of those that engage with our community. So we know when we can bring people the right information, they're less likely to need to reach out. And when we can bring them in contact with other people going through a similar thing, they're much less likely to need us.

And so when we think about it from an experiential standpoint, it's how do we bring people the right information, wherever they happen to be in their Adobe journey? So this would be somebody that might be brand new to design software or, you know, imagery tools. This might be somebody picking up a new skill.

We know that we have about 12 different internal personas that we think about in terms of creative skillset and need. And then we've actually also started now doing some qualitative usability testing on the actual help surfaces. So how do we think about our health platform as a product in and of itself? And our goal is to sell you an answer for free, basically.

JH: [00:09:26] Yeah. It's just the, you mentioned the 12% is there, like, it just seems so varied in this case. And like, even at the simplest level, right. Of like the person who gets their Ikea furniture home and opens a box and like throws a manual on the. Aside and get started and the person who reads it first, like those just seem like such different use cases.

I'm very curious how you guys try to solve for that and how you try to learn about what each person wants, versus the person who wants to get started versus the person who like, you know, has a very specific question that they want answered right now. And they don't want anything else.

Andrew: [00:09:58] Yeah. I mean, I think that, you know, that that's kind of the perennial tug that we feel so are.  EVP of product for the individual solutions is a guy named Scott Bell scan. He had founded B hands before he took on his product role and he wrote a really great medium article about kind of the, the user's first mile journey.

So how do you get them started was something that might be as complex as the creative ecosystem that we operate in. And so that's influenced a lot of our thinking, how do we get you the right information to get you going? Because I think every business in the world will tell you that if you can get that early engagement, the likelihood of you remaining with the service goes up exponentially.

And the longer you go on without using it or using it meaningfully, the less likely you are to stay. So part of the thinking is just how do we get you going? I mean, Adobe faces a really interesting set of challenges in that, you know, we've got. What we think about as very discreet product segments, and then we're also global.

So we have over 20 different languages that we support. You know, we've got really three major product areas. There's creative cloud, which is what we all have probably experienced with Photoshop and illustrator and Adobe XD. Then you've got the experience cloud, which is, you know, All of the work that we're doing with Marquetto and Magento and empowering e-commerce Adobe analytics.

And then you've got document cloud, which has a lot to do with how we think about Acrobat and PDF, and really working for big businesses. And those are all different segments with different users and different needs. And then when you dive in a little bit deeper, the way that, that our team is structured within CX customer experience to help it is we have six different segments and each one of those segments has a support lead.

So I am personally responsible for support for the design and web. And I have a manager who works with me that runs a team of both content strategists and social writers that are online to support our community. And then we have a specific segment for the really high impact high need customers, which we have nicknamed internally ABCD.

So that's account information, billing information, cancellations, and download issues because we know that those drive 85% of the context that we'll get through our active support queues. So really thinking about user intent when they come in and with kind of cross to the product suite that you're working with.

Erin: [00:12:23] Hmm. I know, it's so funny for JJ and I, where we're, where we joke that we're the largest company in the world, which we almost are at 35 people. So, you know, it's always, it's fun talking to people in larger organizations about how you manage, you know, the sheer, breadth and scale of, of the different folks in different stages you're trying to work with.

And it sounds like you have, you know, divided up in a way that seems to make sense. How do you stitch together that journey across these different units and segments?

Andrew: [00:12:57] So I love that question because I think, you know, it goes to kind of the heart of, of what. My boss, his vision for how he thought about it. So, just came up with this idea that we'd have both segments and chapters kind of riffing off. I think a little bit of the Spotify model of, of chapters and guilds and so horizontally, we will have content strategy as a chapter.

And so there will be an embedded content strategists for all of those segments that report into a single leader. And so we can stitch together that kind of horizontal vision of excellence for each of the vertical needs that we have. Right? So you get the consistency around, you know, style guide, tone, and voice, all of the things that you would assume good product to do well.

Well being really responsive to like what we do with Adobe rush and video creation on the iPad is going to be very different than what we need to do with, you know, government and enterprise context for, you know, the content cloud. We need to write with an Adobe voice. We need to speak customer, not speak Adobe.

And, you know, I think our biggest failure up until now has really been to an extent shipping our org chart. So we put very specific types of content on very specific platforms. And then you kind of have to rely on the user to go find them. And I think the big turning point this year has been to rethink that and make sure the right information is available at the right time with the right context for our users.

So it seems super obvious, but you know, we just, we hadn't gotten there up until 


JH: [00:14:37] How do you do user research for health content? Is it just like any other user research? Can you schedule sessions and understand pain points and stuff or, or do you take a different approach when trying to do this type of discovery and digging in.

Andrew: [00:14:50] Yeah. So I think that there's a couple of different angles. So we do a data driven, bottoms up approach, which is for everybody that gets in contact with our. Customer service or active support. We log it in a system and that system has embedded NLP, natural language processing. So it can start kind of classifying contact issues.

And from that we've built this model, which is our tack model or toasted total adjustable context model. So we look at how do these things bucket up at the highest level, and then almost work backwards to get to a leaf level like, Oh, people want to talk to us about billing. Okay. Let's, let's dig in one level deeper.

Oh, people want to talk to us about changing their payment method. Oh, let's go. One level deeper. Oh. People are really having this issue with, they can't find the link to update their credit card. And so you can kind of see the reasons why people call and from that, you really get a sense of where the hotspots are.

And then you can kind of do the top down. Like once you write the content, then you can go back out and you can say, okay, so we think we have a solution. Let's either put it through some level of qualitative testing or you can say, all right, let's launch an internal AP and see which one of these deflects, the most calls from active support.

Because at the end of the day, a lot of times service is considered a cost center, right? Every single phone call cost you, I don't know, less pick a number $6. So every time we can get somebody, the answer that they need without having to contact us, they're happier and it's costing us less money.

JH: [00:16:24] Is that, is that a difficult one to, Like measure in a quantitative basis. I'm thinking of a former employer that I won't name, but at one point we AB tested, you know, having the phone number easily accessible from the homepage. and you know, the results look great. Hey, causal, and laid down when we don't have the number on the page, but then when you talk to some people afterwards, they're like, you know, some are fucking pissed or like, I just want to talk to somebody and I don't know how, So, how do you, how do you balance that kind of stuff in this, in this part of the tier point?

Cause it, cause it is viewed sort of sometimes as a cost center. So it seems like a tough balance to figure out

Andrew: [00:16:56] Totally. And I mean, this was I, one of our senior directors of strategy in the company, and I had a very long conversation about this. Cause I'm like, there's a way to get to a hundred percent of flection and that's you just hide the number, but it's a hundred percent, you know, the, the thing that I keep coming back to in mind is, you know, people call you when they're having a problem and it's a brand inflecting moment, right.

Because if you delight them and you surprise them, they will go out and rave about you. And if you piss them off, they will get on Twitter and light you on fire. I mean, I think if you've ever bought from Zappos, you know, you send it back and they've already shipped you your new size before you even completed the return.

And it's just delightful and you're willing to pay the premium to get the service and people love it. And you think about other services where. They're just, you know, and I'm not going to pick on any of the cable companies, but like they are some of the least loved businesses in the world. So to answer your question, I think we think about it two ways.

We look at the overall financial impact to us, but then we also really, really sensitive to have metrics and check metrics that are very much about the customer's experience and journey with us. So looking at things like a SEASET score, Can we send them a post context survey to figure out if we solve their problem and what they think about their interaction and then train people on what went well and what didn't.

So we have a very close partnership with the actual call center team to figure out ways that we are answering questions. Well, both in person, either online or through, I mean, voice or through chat, and then how we might feed that forward into the content we're writing and vice versa.

So, you know, I think the goal is always to get to, nobody needs to get a hold of us, but we need to make sure that we're solving their problem first and foremost, and then being thoughtful stewards of the company's investment in us.

Erin: [00:18:47] you talked about there sort of, if I am saying the spec correctly, there's this sort of hierarchy of, you know, figure it out within the community, like UDC answers, then there's like the learning help content answers that you all are creating. And then there's, you know, get on the phone and sort of in order of most preferred, do those also.

I, it seemed like those were the most sufficient order, right? Like, scales with other people, creating it, scales with us, creating it, you know, one to many, and then these one-to-one calls. And then you also find that that is the same border in terms of experience, right. That, that, users are happiest getting their answers that way.

Or it's the most efficient or whatever metric users might care about when they're trying to find answers.

Andrew: [00:20:13] Yeah. So I'd love to say that was true. I think what we're seeing is when we connect them to the community, it is the most effective for them. So people that engage with the communities seem to need our help a whole lot less than people that don't in terms of where we've had the most success over time.

It's actually. With our health platform, which internally is called help ex like I said, it's that half a billion page views a year. So by traffic and volume, it torques community right now, but there's so much promise and community. So actually one of the things that we're in the process of going into qualitative testing on.

Because the new product support hub page for a couple of our major products that we're going to AB test that has an embedded community widget in there that routes you to thoughtful conversations about the thing that you're looking for, really with an eye to saying, Hey, we have this available to you and you know, you should 


JH: [00:21:10] When, when determining like who to talk to, Do, is it important that you're trying to talk to users who have like recently interacted with some of these help resources or recently encountered a problem and seeing if they were able to get it resolved? like, is that like kind of the primary persona you'd be screening for like in discovery type research or, can you kind of talk to anybody and just be like, Hey, would you be able to figure out how to do X, you know, through this experience?

Andrew: [00:21:35] Right. I don't know. I mean, I think that that's definitely a very, we're in a very early stage of the process up until, A few months ago, we didn't even have a concept of how we might usability tests some of our new experiences. And then, I was lucky enough to land the role and they reorganized our team really and supportive this.

So right now, I think we're more concerned by your level of experience with the tools, because our hypothesis is you're going to be looking for different things, depending on how familiar you are with the tools in the ecosystem. I think one level of sophistication further, if we can get a hold of those people then is to think about people that have gone through a previous experience and ask them if they're willing to come on board and help us, you know, give their perspective on whether this newer experience might have worked better for them in that situation.

So aspirationally, I want to go exactly where you were saying, but right now we're still super early days. If I can talk to the right people that have familiarity with the tools, some may have needed help recently, that would be a huge step forward.

 Erin: [00:22:42] How do you run your AB tests? You, you know, 50, 50 random what 

Andrew: [00:22:47] Yeah. so good news. You know, we worked for an amazing company that builds piece of software called Adobe test and target. So we're able to actually do some fairly sophisticated segmentation. On the backend. We tend not to go out the doors 50 50, so like best case, you know, we might run a 10% sample again with a half a billion pages a year.

You can kind of like get a statistically significant sample in a relatively short amount of time just given the overall traffic and volume. I think the long tail for the help ecosystem is very long though. So there might be pages in there that might see a handful of people a year. so, you know, really being thoughtful about where, where to place the experiment and then how to segment to best support it for things like our chat bots, where we're rolling out into new markets, we're launching a bunch in Japan right now.

You know, we'll, we'll tend to do like a graduated rollout monitor metrics for a little while kind of step function up to the next level and then go again. So I think we'll start it at 5%, 10%, 25 50, and then a hundred percent.

JH: [00:23:55] Hmm, the Zappos example you, throughout earlier made me remember, you know, a data point I've heard from them before that was, they actually saw higher retention from people who had interaction with our customer support team on their first order then than not, which is this really fast. Like, you know, you don't want to go in the business of like messing up first orders intentionally.

but is there any like similar dynamics that, that you all have observed within like the Adobe universe of people who actually are spending. A ton of time in these parts of the site are doing a lot of page visits or actually like some of the loyalist users, because they, they really trust and rely on these resources.

Andrew: [00:24:28] Yeah, no, that's a great insight. we actually did. We looked at, overall, lifetime value LTV, and we showed that for people that had reached out to us and, gotten an answer, I'm not counting whether or not it was the right answer, but their lifetime value almost doubled over somebody that never reached out.

So I think like, like we think about it's those brand building experiences, right? It's I had a moment of pain or a moment of need, you were helping helpful in resolving that need for me, I believe in, you know, it's the company, the product, the solution, and that loyalty absolutely comes through.

JH: [00:25:07] Yeah, it's almost like it creates this like optimism in the product or service. You know what I mean? I'm trying to think of an example for myself and, you know, two years ago when I'm starting to play around with the web flow and, and looking into that as a solution for our marketing pages, hadn't used the tool before played around with it a fair amount.

And just kind of discovered that they had this amazing, like they called it university with like all these different topics and videos and help things. I couldn't tell you, you know, what I looked at or what problems that helped me solve. I don't remember any of that, but I do just have this very strong belief that like, Oh, if I'm in there and I get stuck somewhere, I can find an answer really quickly.

And I know that. And so like it, it just kind of creates confidence in the tool almost if that makes sense.

Andrew: [00:25:42] It really does. And I mean, it makes you feel like they're in it for your success. You know, we had a very similar experience in my time at Groupon, we built our merchant marketing pages initially on Webflow and my designer had an issue and reached out to the support and the CEO ended up writing him back and they had this whole long back and forth conversation and you're like, wow, like the guys taking that much time to make sure that my little marketing page is a thing.

Right. And so you just, it's this kind of, it almost feels like mutual respect. Like you're willing to invest in the product. They're willing to invest in you learning the product. And of course you want to stick with that.

Erin: [00:26:19] So 


JH: [00:26:21] Yeah, good Twitter follow, shout out, lead.

Erin: [00:26:24] Yeah. Very good. On Twitter. Awesome. Okay. So Andrew, you're talking about Japan in 20 different languages. So just to pile on this topic of, Oh my God, you have to talk to so many different people. You have the opportunity to talk to so many different people. When I think about content and obviously different languages, but not only in different languages, right?

Different cultures, different, et cetera. 

Andrew: [00:26:50] So Japan is a really interesting case because you know, we, we have these, this content body that was written by people. You know, over time that had historically come out of technical writing. and you know, as we've moved forward and evolved, we've hired more and more people that are content strategists and UI writers.

But Japan is really funny because right, the help system, I'm going to talk around this circle, but I'm going to get to you, your question, the help system that we have today, isn't nearly as smart as it needs to be. And by that, I mean, when you log in. We aren't sure of what products you actually own. So we consider that to be entitlement.

So because of that, I have to tell you all the things. So I joke with my team that we are always complete at the expense of being concise. I don't know which version you, I know which version you're running. Cause that's what you searched for. But I don't know if you got version a or B what account you're on, what you bought, what you've had historically.

So I tell you all the things. And it's usually a lot of texts and not a lot of imagery. So we actually started to see bugs getting filed and, you know, unhelpful ratings from our Japanese customers, because they were used to much more of a visual, almost comic book style, where you have visual 

panels with very few words.

And it kind of prompted this, this thought experiment of like, what would it look like if we redid it as almost, you know, two by two graphic, few text graphic, little bit of words, graphic, you know, small description. And so that was one of the first AB tests we ran and it turns out it's wildly, wildly successful in the U S as well as Japan.

So where we're used to writing pages and pages of information for you. And in hindsight, it all seemed very, very, you know, obvious and intuitive. Like of course people want to look at pictures, but when you're talking about things that are really kind of complicated, like edge detection and how you use selectors and all of that, there's no really clean, straightforward way to do it visually.

So you ended up doing it descriptively. And when we pushed ourselves to simplify this story, It just made the content that much more consumable. And so in that case, you know, our Japanese market actually pushed an improvement that we are now in the process of rolling out more globally. That was really, really 

Erin: [00:29:22] okay. Yeah, awesome. I was going to ask, I mean, doesn't everyone official, picture's worth a thousand words, you know? so, and I know, right. Japan is a, the, obviously the language, the, signs more sort of, abstract visual, in a lot of ways, but yeah, I would, I would think if you could communicate visually, that would translate in a lot of cultures,

Andrew: [00:29:43] But I think that goes back to your original question, which was like, when you think about the team and the staffing. We've actually built a dedicated team of like visual communicators that are, you know, designers and illustrators that really think about how to display that information through, you know, illustration, infographics, wire frames that we can use 

Erin: [00:30:06] Great. Right. And it also reminds me of, what is the, you know, the accessibility sort of truism that when you, design for what seemed like edge cases or, you know, marginal, situations you end up designing for everybody better. you know, have you found that from your kind of research into other languages and cultures?

Andrew: [00:30:28] Okay. I haven't seen that specifically yet. And I think that's because again, we're trying to be exhaustive with our explanations. one of the things that we're finding is have historically been excessively redundant. So like we will restate the same three steps for every scenario. That you might possibly go through.

And what we found is that people tend to tune out again, like if you're looking at this wall of text, even though there might be that piece in the middle, that is totally relevant to the thing that you're trying to answer, you're just going to get discouraged. So we're trying it also, again, it feels very intuitive, but it, it took us a while to get there.

How do we simplify and group solutions, common solutions to multiple issues together. So, we have a new test that we're running around error messaging and having a bot. So you just kind of type in your air message into the bot and it will route you to the correct resolution. It turns out the same resolution might work for 10 or 15 different error codes. And so, you know, we've always done a one to one mapping and now we're like, Hey, we can be smarter than that. We can do the heavy lifting the customer. Doesn't have to do the heavy lifting and we can use automation to make sure that people are getting to the right outcomes.

JH: [00:31:50] Nice. The biggest question I always have, the mind that comes to mind for me is when you're talking about this much help content in various forms, written and illustrative. how do you keep it up to date? Right. Like the, I assume the product teams are working hard to work in fast iteration cycles and make improvements.

And sometimes you make a small change to something and you didn't know that it was referenced in some support page somewhere. so you didn't maybe think to tell anyone, like how do, how do you deal with that challenge?

Andrew: [00:32:16] So, like every good product, you know, we've got a really good bug tracking system. so a lot of times as the PMs are going through the notes, like we will pay right before releases, we'll review it with them. we're also going through and constantly grooming and pruning to make sure that articles are wildly out of date, for each product there's kind of a new release set, but I think it is just, it is, you know, It's the garden aspect of what we do.

It's how do we keep everything working? We are not yet perfect at it. We're constantly striving to get better. we're in the process now of kind of looking at the hundred, most consumed URLs kind of per product or per segment, and going through each of those to make sure that those pages have a disproportionate.

The amount of our attention, because we know that that's where people are going. And then, like I said, that long tail is super long. So, you know, we start looking at quarterly at metrics or pages that have less than X number of page views. So we still need them up. Can we archive them? Can we store them off as PDFs with links to them?

Can we put some of those into the community and just route people for that particular topic? I mean, we've got some really interesting conversations going on where people have gotten in touch with support for the campaign downloader for creative suite two or three, and that hasn't been done. Those were built for operating systems that have been not supported by their manufacturers for the better part of maybe a decade.

Now. Right. And so like, when you think about, we've got these perpetual licenses out in the world for software that doesn't even run on modern computers, it's just a really interesting set of challenges to face. 

JH: [00:34:02] I'd be fascinated. 

Andrew: [00:34:03] And I mean, the cool thing is like, we're like, Hey, we'll give you a brand new free license for this other, you know, do version of this software. And some of the times we're like, Hey, I've got this, you know, Mac book, white version. That's it won't run.

JH: [00:34:17] The polycarbonate one. That was a good

Andrew: [00:34:18] Right, right. I mean, my wife, when she took her first advertising job, got the yellow, I call it the toilet seat one.

It was the original clamshell was the bright orange.

Erin: [00:34:27] Yeah. 

Good. So Andrew you've been 90 days at the, in the gig. what, what, what are you excited about for the future? I think 90 days you're supposed to be like, you're on the positive side. Right? You learn a little something you're contributing a little more than you're consuming. What's what's what's 

Andrew: [00:34:53] Well, I mean, one of the things, and this is a total shout out to my peers. It is a spectacular company. Like I didn't expect. Love it as much as I did. In fact, I had to embarrassingly tell my boss at the previous three design gigs that I'd drawn with, basically steered clear of everything, but illustrator and Photoshop occasionally because some of our competitor tools, especially the online ones were so powerful and so easy to use.

and then I really got into some of the stuff that we've been doing and taking a second look at things like Adobe XD and really understanding how it works for teams and multiplayer is pretty cool. But I like that the company is as inclusive as it is. I like that it's as invested in the community as it is.

And then for me, it's really kind of trying to shift the company mindset and push it along to make sure that we're of greatest service to our customers wherever they happen to be. And so I think there's really this groundswell now of helping people understand their tools and get really, really competent with them pretty quickly.

And so part of that is just making the right information available at the right time. So rather than make you come to us and have to hunt for it, how might we put it in front of you in a way that helps you best? And that's the thing that I'm really excited about. So working really closely with the team, that's working on Photoshop to think about how we might embed help in community content inside the app in a way that we haven't really done successfully in a very long time, thinking about working with our learn team and all the crazy cool videos that they produced.

Give you that really easy to consume information, working with our social teams, to think about how we might build bots that help you on Facebook or Twitter that answered your question or take care of your issue without forcing you to come to one of our sites and sign in and go to a specific search and find a specific page and click a specific button.

So really doing everything that we can to remove the barriers to helping people be successful. and you know, I think. Bringing prototyping and research into this capability is something that my leadership is really excited about and they were willing to invest in me to help that happen.

JH: [00:37:02] Awesome. This all reminds me, about like a thing we've explored before of like how quantitative data can support qualitative data and vice versa. And they're kind of like friends, not either, you know, not an either or, And it feels very similar to me about it. Like when you think about like improving the product and improving the intuitiveness of the app versus, you know, investing in health content, it's, they're very like complimentary, right?

And I think sometimes maybe there's almost an arrogance within some product managers or whoever it may be of like, let's just fix the thing and make it easier to use, but like there's a role for this stuff and they can help each other and you can take some of the same principles and techniques, and share across.

Andrew: [00:37:36] Yeah. So one of the really cool things is we get to do. A monthly readout with the actual product teams for each of these segments. And we sit down with them when we say, Hey, here's some of the rising topics and conversations. And here's where a lot of the heat is with issues that people are facing.

you know, one of our. Love more longer lived imagery products was having a performance issue and we were able to identify it early and then work with them to triage it and work with the act of support, to gather more information. When people called in and they pushed out the fix in the last release, they seemed to fix the memory issue that they were facing for.

It seems like probably a pretty long time. Like I said, I mean, I keep bragging about Adobe XD partially because you know, in my design, Background, but it's also part of the portfolio of things that I support and not leadership team all the way from the VP of product all the way down is a hundred percent committed to partnering with us.

So they appreciate the kind of quantitative information we can bring about where issues lie, as well as some of the qualitative insights that we can get through testing in the conversations. and so if we can bring them a list of like, here's all the chat logs, where people have had an issue with this particular feature, they can then go dig in and figure out and test different ways to solve that problem within the product.

So I think what we see is that the qualitative gives us the quantitative gives us a sense of what the issues are, but not the rationale or the reasoning behind it. And then the qualitative really gives you that nuance and that texture and the intent of the user to really dig deeper.

JH: [00:39:08] the piece too. I think it's always helped for me as a thought exercise is, you know, there's a spectrum here, right? And the ends of each, the, like, if you think about the ends of the spectrum, neither seems that appealing right. Of we're never going to improve the product and we're a hundred percent gonna rely on help and support content for people to learn how to use it.

And vice versa of. We're going to improve the product constantly, and we're gonna invest zero in any sort of help or support documentation. Like, it feels very unlikely that an organization would pick either of those extremes. And so like, once you've done that thought exercise, you're like, all right, we're somewhere in the middle.

And it's just a matter of what balance is it? 50 50? Is it 70, 30, whatever it is. but there's, you know, you're going to do both. And so how do you figure out the right blend and how do you make them be supportive of each other? It feels like a really good exercise.

Andrew: [00:39:49] Yeah. I mean the good news, bad news is I think we've, we've decided on both and we're doing that, but really funding two almost parallel efforts, one to build a great product and you know, one to build great support and the crossover and the matrix is a thing, you know, D do we wish we had tighter communication?

Absolutely. Do we wish that. You know, things, I don't know. I mean, it's a really interesting challenge. I like the way you stated it, right? Because as part of the world where all, everybody just works together to solve the problem, or maybe you, you embed the writers in the product team, but then you miss the continuity of tone and voice and solution, or you have two separate orgs, but then there's kind of.

Communication overhead and information sharing. One of the things that I'm noticing about Adobe is the transformation from being a box software to a cloud software companies is real. And there's, you know, habits that we have to build in things that we have to relearn. Right. When Photoshop back in the day, used to, I've heard internally and I have no way to just.

To, to kind of justify your answer it, but there was this sense that at one point Photoshop and illustrator felt like maybe they were at odds with each other a little bit, that they were competing for very similar users. It turns out they're really complimentary set of tools and capabilities. And now that we're in the cloud together, you know, we're trying to think about like, how do we make the same thing behaves the same way across different tools.

And I think that that becomes true of. But that learning and support piece as well. Like how do we support the complexity of each individual product and make it easier to use while ensuring that we're speaking to our customers in a way that is recognizably, Adobe.

JH: [00:41:36] that makes 

Erin: [00:41:37] Yeah. And back to, I was thinking the one, two, three, you said, you know, you always have the one, two, three, and the, every, you know, support article and people gloss over it. It's like, how do you communicate the same thing? Over and over consistently without 

Andrew: [00:41:53] Yeah. 

Yeah. No totally. And it's a real thing. I mean, and even there, I mean, there's things like just template optimizations for that stuff, right? Like we put a table of contents on the side of each article in support of these really kind of long form written pieces when you're only a page long, it stops making a whole lot of sense.

So how might we think about this space differently? Might it be other articles that you're looking for? I mean, the analogy that I bring in from my background in eCommerce is that if you do a Google search and you draw upon to any major e-commerce retailer today at the top of that page will probably be other things you might be looking for because they know the odds of you hitting it.

The first time are relatively low. Do we want to do the same thing with health articles? We know that of our inbound traffic comes through Google searches directly. Are there ways that we might want to share with you? Other things you might be looking for, you know, are there ways that we might want to connect you to actually being able to purchase the software from that?

Are there ways that we might want to show you community inspiration and videos and solutions that people have built with this stuff? one of the things that we've been exploring recently is we know with video games, people go through three or four different tools in order to get the animation.

Element that they need the power of video game. How do we start stitching that together as a more comprehensive solution? Not just here's how you draw character, you know, Photoshop or illustrator. And here's how I use three D animator. And here's how you import it. And here's how you export it. But like really build that story of like, here's how you build something for it, for you and take you through each tool.

And each step that you might use to give you a sense that this is something that you can do as well.

Erin: [00:43:38] Well, good luck with all that. I was just thinking we should, we really need a, where are they now? you know, follow up. Series so many cliffhangers. but fair, very exciting stuff.

Andrew: [00:43:54] No, it's absolutely been a pleasure. Thank you for letting me talk at length about all the cool stuff we're doing and, I just want to give a shout out to you guys. He's there. Interviews has been amazing for the research that we've done before and that we're going to do. I appreciate it. Having access to all the tools that you guys help us

learn from. 

JH: [00:44:13] Cool appreciate that.

Carrie Boyd

Content Creator

Carrie Boyd is a Content Creator at User Interviews. She loves writing, traveling, and learning new things. You can typically find her hunched over her computer with a cup of coffee the size of her face.

More from this author