Published: 7 May 2021
Onboarding is not a narrative
Gerry Gaffney
This is Gerry Gaffney with the User Experience podcast.
My guest today has worked on a range of hardware and software products across a range of industries. Currently at Google, she’s also worked at eBay and Nvidia. She has a specific interest in user guidance and onboarding, and in fact, she’s written a book called Better Onboarding, and it’s a very nice and concise exploration of this topic.
Krystal Higgins, welcome to the User Experience podcast.
Krystal Higgins
Thank you for having me, excited to be here.
Gerry
Now I’ll remind listeners that as always a transcript of this episode is available at uxpod.com.
Krystal, can you start by telling us what we mean by onboarding?
Krystal
Yeah. When I talk about onboarding, I mean the process during which people are sort of acclimating to our products for the first time, and it’s a process that connects many activities that bridge the gap between someone just checking out something new and getting into the state of being a core user of it.
Gerry
And you use that term core user throughout the book. Is a core user somebody who has completed onboarding or who has reached a certain percentage of onboarding, or how do you think of it?
Krystal
Core user is a term I use to sort of represent when the person who comes to your product gets to a state where they’re sort of sustaining their own needs, as well as the needs of your business as well. So for different products, there will be a different definition, but it’s definitely the mark of a successful onboarding experience when people get to that state,
Gerry
How did you get interested in this area?
Krystal
Oh, well it, it started maybe in a more superficial way. I was working on setup wizards for some hardware devices like printers and stuff. And at first I was very excited about creating finely crafted up flows. And along the way I realized you can’t really teach people through that process. There’s actually much more interesting things you can do along the user journey to help people learn. And it turned into a wonderful blend of UX, teaching, educational sort of industries, and kind of went from there across all the other projects I’ve worked on.
Gerry
In the hardware business. In some ways it’s a bit easier. I know printer manufacturers a few years ago started putting a big sheet of paper, right at the top of the packaging material saying "Read Me First." And of course people would pick it up and throw it away immediately, but it’s a bit harder in software, isn’t it?
Krystal
It is. It is. I mean, in software you have many things sort of working against you if you want to do this, sort of read me approach, right? People aren’t necessarily purchasing a tangible object that has a hassle of returning later, right? It’s really easy to delete an app. And people honestly don’t think as much about needing to learn a new app and sit down with content to learn it. So you have to work with that as well when you do it.
Gerry
One of the things that you say that many people have and I quote, “an instinct to treat onboarding as a story,” and you talk about products that kick off with “Let’s get started” and end with “You’re all set.” But as you point out that approach may not be effective.
Krystal
Yeah. So you know, these sorts of things are narrative approaches and, you know, they can appear in apps, especially like a video or a carousel of screens, or even a super long web page describing all of your features. And the problem is that these aren’t super successful because it’s, it’s often us telling users what we want them to know, because we don’t know what they need yet. And so it risks everything feeling very irrelevant and overloaded to those new users. And it also assumes that people are coming to products to be schooled, but they’re really not, right? They’re coming to address a specific need. And they’re not really expecting to have to sit down with content and you know, a myriad of other things can be negatively impacted by this kind of approach, right? It’s out of the context of use, it’s a lot of work to maintain and scale these things, and it can even just seem, make a product seem overly complex. If you have to see all this content of front, it kind of gives you this perception that this thing might require me to get a lot of help to use it successfully.
Gerry
You mentioned videos there. Videos are very popular for onboarding. How do you know if they’re effective for your particular product? How do you know if they’re right for your audience?
Krystal
Videos are one of those things that you have to be exceptionally sure that people need that content upfront. Videos are great on demand for some people, right? If you have decided you want to dig into a concept more and see someone actually demonstrating it, then videos are great, but I find them better in the context of sort of later support or help content or supplemental content. Videos, just, you know, they assume again that maybe you have headphones to listen to them with or something. And they can be very passive as you sit there and you wait for it to get to the part that maybe you actually care about.
Gerry
Yeah. And it can be very disappointing when you fire up a video and you find out that it’s somebody reading a text that you could have read yourself by now.
Krystal
Exactly.
Gerry
You mentioned George Fan using the first level of Plants vs Zombies II as a way to present simple challenges to familiarize players with key components of gameplay before taking on more complex levels. And you say that for him, “it’s a compliment if a player says they didn’t notice the game had a tutorial.” Do you think this is a good approach, but do you also think that it’s suitable for non-game products?
Krystal
Absolutely. The underlying idea behind what George Fan is talking about, he’s, he’s kind of talking about gradual engagement, right? So in Plants vs Zombies, there are, it’s a distinct to levels games. So having the first level sort of be a gradual introduction into the gameplay space makes sense in that context, it’s just level zero and you can skip level zero or go through it quickly, however you want, but it doesn’t feel like it’s not a part of the experience. There’s another game, Breath of the Wild, where that’s more a map based game. And in the first kind of part of the experience when you’re getting onboarded, you’re restricted to a portion of the map, but it doesn’t feel restrictive because you have no reason to expand into the later world, but you’re using this time before the map opens up to learn the concepts that get you out of your current map area.
So in products, this is, this really relates because it’s about just kind of gradually exposing the different parts of your product as people get more immersed.
Let’s take a document writing app, for example. The app might start with an almost empty space, but there’s one sample document in there and the document can be fully editable, but the content already in that document can sort of be showing you what the product can do. And it’s great because it doesn’t get in the way of people who are very knowledgeable about these types of products. They can ignore it, they can delete it. They could create a different document from scratch, but those who might want to play around could even just use this document as a way to sort of grow off of, and turn into one of their own. So thinking about these ways we can naturally grow users from a small portion of our products to getting into the more complex stuff over time is really what that game concept is about.
Gerry
Now you’ve already alluded to this dilemma, but to read another quote from the book: “To effectively onboard new users, we can’t rely on all of them to read or watch a video about our features settings and UI right away and expect them to be successful when they’re in the thick of things. And we can’t drop them into a new space without any support.” So what should we do?
Krystal
Yeah. So what, what we need to do, and I recognize it’s kind of a trick, right? Because the two options you describe are on opposite ends of a spectrum and like anything in the design world, we kind of need to be doing something that’s sort of in the middle. So the best approaches will really be trying to guide people while they interact with our product. That means anchoring. That means anchoring information that we’re providing to the action that someone’s in the middle of and structuring that in such a way that you sort of prompt someone to get into these new activities that they might be doing, guide them through providing information or examples as they’re doing the kind of meat of the activity. And then following up with them at the end to sort of guide them forward and point out the other areas of the product they might explore. And I call this term guided interaction to kind of give it a framing that will help you maybe explain it to your teams, how to take a better approach.
Gerry
Tell us about the paradox of the active user.
Krystal
Yeah. This is something researchers from IBM a while ago defined when they were watching people set up computers and use new computer software. And they were noting that the people in the study, instead of diving into these really great training manuals that had been created for them just rather wanted to jump in and start doing the task they cared about. And because they were just jumping in without training first, they ran into many errors which impacted their ability to be successful. And so the paradox was that to be successful, people needed to read these kinds of materials, but because you can’t rely on a people to read those sorts of materials upfront, they needed to find another approach. And they called that a paradox. So obviously that approach can include anything from a better interface to actually putting this content in the context of people using it, as they dive in and start doing their specific task.
Gerry
You reminded me there, I remember one of my kids opening some new device. I can’t recall what it is and opening up the package and taking the instructions and literally throwing them on the floor and saying, “I’m a boy, I don’t read instructions.” Without wanting to get into murky water, do you think there’s any gender differences between the way, the way that men and women approach onboarding? I know that’s a question out of the blue for you.
Krystal
Maybe not gender differences. There, there is an organization, a group of people and they called their, their research GenderMag. And they were going through to see how different genders or groups of people approached learning new software. What they found wasn’t necessarily gender based, but it’s really about different people will take different approaches to new products. And so there are tinkerers, which is, sounds like your son just tossing the manuals on the ground and diving in I’m one of those too. But then there are people who do prefer to get a little bit of a holistic look at things beforehand. I used to be friends with somebody who would, you know, spread the manuals out on the floor and look at everything before they started building something. So what that means for us and is, is that we have to find, you know, flexible ways of guiding people, given their situation. And that’s another reason why upfront stuff that is given in one form like a video isn’t successful is because while some people might view that stuff, many other people will not. And so you need these sorts of in-context options for people who take these different approaches.
Gerry
It’s interesting your using the term upfront there. One of the, I guess, frustrations at times of using a new app or a piece of software is when you do jump in as a tinkerer and get to a certain point, and then you think, Oh, there was something in that, you know, in that intro that perhaps I should have looked at, and then you can’t find it because the developers have assumed that by that stage you were already onboarded.
Krystal
That’s absolutely the other issue. There’s, you know, those assumptions that you only need this content once upfront are really wrong. Sometimes you don’t know what you don’t know, right? And so once you jump in, you start playing around, you’re assuming as a user that you’ll be able to figure it out along the way. And that assumption is based also on assuming that the people who built the product you’re using have also planned for that. So if you’re suddenly in the middle of a process or a higher risk kind of activity, like maybe you’re trying to delete a bunch of files or something like that, that’s not, when you want to find out you don’t actually have enough information to know what to do next
Gerry
Now to change topics slightly. There’s another quote I’ll read out. “A fair number of the challenges I have experienced in user onboarding have stemmed from unclear or misguided goals that the team should have been aligned on before the project kicked off.” Do you have any advice for this situation?
Krystal
Yeah, absolutely. So usually these mix-matched goals come from different people wanting to get different things out of onboarding. So first you want to try and align your team on a good definition for what onboarding means for your product. And that often means a good definition of what state of core use as, as I call it in the book, do we want people to end up with that we can say, hey, they’ve been successfully onboarded and they’ve reached a state where they don’t need to be continually kind of guided in the product. That also includes a lot of user research and building this new user mindset. So you want to make sure that you’ve got a healthy mix in any research you do of existing users and new users. Don’t just save the new users for one study on onboarding. And you can also build a new user mindset in your team by just having everyone go through the first experience themselves. It may require resetting a device or clearing, you know, all of your history or something if you’re working on a website. But it’s very funny how often we just forget that experience and don’t go back to it as we’re building a product. So getting your team to revisit that occasionally and feeling what the new users might feel is going to really help that.
Gerry
Yeah. And I think particularly when you get new team members, you can rather than explaining things to them, you know, let them dive in and even if possible observe or get them to describe or document that experience.
Krystal
Totally. That’s a great option. You can also have your team constantly go out and try new products that aren’t yours and report back on that as well.
Gerry
You know, it occurred to me that there’s an inbuilt trap in the use of the term onboarding at all. I mean maybe if we didn’t have a concept for onboarding, it would be just supporting users throughout their journey.
Krystal
Yeah, that’s definitely one part of it. I think the only issue with that is that sometimes it risks us losing sight of new users altogether. That’s probably why it’s still important to keep them in the loop when you do later ongoing research studies, so that you’re not losing that perspective and seeing if you actually do need a bit of heavier guidance when people are new.
Gerry
Yeah. in the book in a section called “Mapping onboarding journeys,” you have a really neat and simple way of planning out the onboarding needs. I know it’s a little bit difficult in an audio medium, but can you describe it?
Krystal
Yeah. So it’s kind of a circular journey map. Now circular or orbital maps, as they’re sometimes called aren’t really new to the design world, they really have, you know, many uses. But the way I try to encourage teams in the book do this is to see onboarding as a constellation of activities that someone might do. I’m trying to get people out of the idea that onboarding is a linear path that everyone follows in the same order. And more that it’s about kind of people coming in from different edges of this kind of circular universe. And at some point they’re going to meet in the middle at this definition of core use or established use that you want them to end up at. And along the way, they may hit a few of the dots that live in the outside as they’re moving towards this unified sort of center definition. Hopefully that will make some sense to your users over the over the audio here.
Gerry
Okay. Well they could always go out and buy the book.
One of the dilemmas that you also mentioned that you’ve kind of alluded to it as well but I think we should discuss it specifically and I’ll quote, “Interruptions, push people away.” I mean, that really must be a key dilemma in onboarding or providing hints, prompts or assistance in general. How do you, how do you know what the right amount of interruption is? How do you test to see whether the amount of interruption that you’re doing is pushing people away or whether it’s appropriate, and how do you tweak it?
Krystal
Great question. Yeah. For onboarding, because people always see it as this separate process. You were mentioning that earlier overlays, interruptions, these kinds of things seem very tempting, right? Because in your mind, you’re saying we only need to show this information once. So we’ll just pop it up once and then it goes away. But the problem with that is timing. If you don’t have the timing right then people have a knee-jerk reaction typically to just closing overlays without reading them because they’re interrupting their flow. So the first thing to do is make sure that if you’re going to show information in this format that you’ve got the timing right. And that’s usually going to happen after someone has completed a full task or they’re in a waiting period, maybe they’re waiting for a file to finish uploading. If you’re interrupting someone mid flow, even if you’re there to tell them about a security alert or something like that, they still have a tendency not to read it and to try and dismiss it as quickly as possible. So that’s one way to do it. I also just encourage, really save these for the critical stuff, the critical things that have backup solutions in your product, because you have to be constantly thinking if someone dismisses this without reading it, what is the risk of them not having read that? And so constantly be thinking about that as you plan out and consider overlays in your design options.
Gerry
I guess it comes back to the principle of stripping stuff out as much as possible rather than adding it in.
Krystal
Absolutely. You want to try and integrate guidance and information as much as possible into the core design of your product.
Gerry
It occurred to me when I was reading the book that a lot of the advice you give is not limited to onboarding, but is really about general design principles. For example you know, “guiding people through this means we have to create continuity, provide support in context, offer alternatives and make errors actionable and informative.” That’s just a quote from one particular part of the book, but it really is not just about onboarding. It’s much more broad, isn’t it?
Krystal
100%. Thank you for taking that away. Not many of the concepts I show in the book will be ones that UX veterans haven’t heard of before. These are all really important concepts, but sometimes we just lose sight of them when we’re designing for new users. Because onboarding has all this pressure on it to get people set up right away and get them knowing everything in our product. We somehow have tended to throw all of these standard design best practices out the window, as we try and, you know, address this one moment. But the reality is that our products are constantly needing to guide people, whether it’s through a new task, whether it’s through changes, anything of that nature and just remembering the core concepts of helping people work with new software or new hardware will really help us to design a product that scales to more audiences.
Gerry
To highlight another dilemma. I know we were talking about a lot of dilemmas, but you said in the book “Whenever possible, defer prompts for creating an account until later in the onboarding experience, when the user is more likely to trust you and can understand the personal benefits,” which, you know, is very sensible advice. But on the other hand, you said, “don’t let a user create a lot of content during free sample experience only to tell them they lose it all if they don’t make a commitment to an account, because there goes your positive first impression.” They’re two very contrary requirements, aren’t, they?
Krystal
They are definitely ones that could be contrary if sort of taking each on their own without the broader context of what makes a good onboarding experience. So it’s really about deferring signup prompts until you can be sure people have enough context to make a decision. Now, surely if you put a, and I talk about this in the book, if you put a signup prompt upfront, you will get sign-ups. However, they may not be the most informed ones, or they might be people who still want to try your product, but haven’t quite decided that they should give you their accurate information. So they give you fake information to move forward with. And you don’t want that just as much as you don’t want fewer signups, right? So it’s really about putting things in context. There are some great products out there that do this well.
Canva does this on their website. You can go in, you can start the design of an initial flyer or a greeting card, and you will be able to download it as just a static PDF. But after you’ve gone through and download it as a PDF, you get prompted then to sign up for an account, if you want to actually save these things, to edit, tweak and have more robust sharing options for later. And that’s a great way to sort of put things in context and help you make that decision that you want to commit to the product.
Gerry
It can come across as extremely dishonest if you, if you do hide limitations, can’t it?
Krystal
Absolutely.
Gerry
Lest listeners think that the book is sort of all about principles and so on. It is quite, it’s full of very detailed advice as well. Now I know this might be a little bit of effort to talk about at this point, but you say we can break onboarding actions down into three parts - prompt work involved on follow-up. Would you mind digging into each of those three parts for us?
Krystal
Absolutely. So this structure that that section is talking about is really to help you understand the, as a designer, how guidance can actually be structured, even if it’s still kind of part of your overall experience. So you need a way to set expectations and get people to take an action, right? And so that is the idea behind designing a prompt. A prompt doesn’t have to be an overlay that pops up, a prompt can be a button that sits in a web page, but it’s really about structuring some guidelines for how you think about what types of things will help someone make the decision to take action at that point.
Then the part of the work is kind of what you normally might think of is guiding someone, it’s about giving information. If they’re on different steps of a flow. It’s about supporting them if there are any errors that come up in that process of going through the flow.
And then the most important part of that structure is the follow up state. So you’ve successfully completed a file or you have saved something or shared something. You need a confirmation to reinforce the work you’ve just done and kind of put things in context of the broader product, but you also need a hint of where you can go next. And we don’t necessarily think about these next steps as much when we’re designing these flows for new users. And so this is how you can see, cool, the follow up state of an action I just guided someone through can actually link up to the prompts for other future states. And that’s the way I kind of want people to think about these almost little modules that can be flexibly encountered at different times in the user’s onboarding journey, but can still link together to take them forward.
Gerry
We’ve talked about onboarding for hardware and primarily software, but it’s also a term used in an organizational sense for bringing new people into an organization. Have you any thoughts on the relevance of your book to that particular topic or is it out of scope?
Krystal
It’s not out of scope, but I’d say that more than anything my book has been influenced a bit by changes from that industry and touches on lightly how some of our perceptions of what onboarding needs to do in products are influenced by the older ways that employers would onboard their employees, where they would define it as just sitting in the room in a room, getting a couple presentations or having to read a couple of booklets about the company. And some of us have transferred that in a literal sense to what we should be doing for new users. So if anything, it’s more about taking, yes, this thing exists in the workplace and it’s very people oriented, but at the same time saying we can’t take away just the orientation part of that industry. We have to look at what they’re trying to do, which is really grow people into an organization over time.
Gerry
Okay. I’ll remind listeners that Krystal’s book is called Better Onboarding. It’s nice and short which I always like in a book and it’s got really solid practical advice on this topic. I think any team that needs to onboard users would gain a lot of valuable insight from it, even if they’re highly experienced in UX in general.
Okay Krystal Higgins. Thanks for joining me today on the user experience podcast.
Krystal
Thank you so much for having me. This has been great.