In this episode we talked to user onboarding and UX expert Samuel Hulick. Samuel is best known for running useronboard.com, producing pithy onboarding teardowns, and writing “The Elements of User Onboarding.
- The importance of understanding customers desired outcomes beyond your product.
- What it means to think of products as value delivery systems.
- How to get started improving your customers path to success with your product
Connecting with Samuel
When it comes to onboarding, many companies focus on specific elements like tooltips, videos, and guided tours. But these companies miss the forest for the trees. Users don’t care about tooltips, they care about outcomes—outcomes that give them superpowers.
And building a valuable product is about effectively delivering those superpowers to users. In this interview, Hulick explores a better paradigm for product design, company-centric vs. customer-centric metrics, investing onboarding dollars in areas of highest impact, and how customer research extends far beyond the onboarding experience.
Stuart Balcombe: Your Twitter feed encapsulates a lot of things going on in the UX and product world. And recently, you’ve been very succinct about your viewpoints: don’t hack growth, hack user value; software is not inherently valuable; great UX is not necessarily valuable...I’d love to hear how you got there.
Samuel Hulick: I have a background in user experience design. Over the last decade plus, I was working in-house at agencies, software companies, etc—as well as consulting and contracting.
When I started useronboard.com about five years ago, that created a big influx in professional opportunities. People were saying, “Hey, I like how you make fun of these other people's products, and we want you to do that for us. But don't make fun of it as much, and don't publish it.”
I was like, oh, there's a business here. And I've been constantly asked from strangers or clients (or anywhere in between): “We're thinking about adding a tooltip tour. Do you think that would be better than an introductory video?” or “I've heard good things about in-app messaging. What product should I use?”
Whereas the really relevant question isn’t, “How do you show people your product?” The really relevant question is, “How do you actually help them do what they want to do?” And presumably, “How do you identify what they're even supposed to be doing—or want to be doing—to begin with so you can help them?”
...the really relevant question isn’t, “How do you show people your product?”
For me, good onboarding vs. bad onboarding has never been a question of, “do you use tooltips?” or “do you use hot spots?” or things along those lines. Instead, you want to think of onboarding as a scripted interaction with someone signing up. What do you want their activity to consist of, and how do you generate that activity?
And the activity really has to start within the context of what the user is wanting to do and trying to do, rather than what you want the user to do. Because you're always going to lose that fight.
That totally resonates with me. As I got into Jobs to be Done, desired outcomes, and progress people are trying to make, it became clear to me that progress has nothing to do with your product. Your product is just a vehicle to get people from where they currently are, to that progress. It’s really a mindset shift.
One thing I'm curious about is: how do you think about defining success? Product teams, and companies in general, are always going to want to measure things. How do you think about defining what success looks like in product?
To be clear, I am pro measurement. I would be relieved to see the design community embracing measurement more. There are a lot of ways to identify what the business impact of design changes are, and most of them are just sitting out there right now. Nobody's really paying a lot of attention to them.
Your product knows what your users have been doing. Why doesn't your reporting know?
The term “seat at the table” gets thrown around all the time. And businesses would do well to incorporate design into their high level strategic decision making more.
What I have found is companies either focus on design as sort of a loose concept, almost like humanitarian values they carry within their organization. That's the school of user experience design focused on—good, bad, or otherwise—things like personas, user journeys, scenarios, and things like that. The point is to inculcate an appreciation of the user's perspective into the decisions, but without a strong way to tie those decisions to the actual impact they have on the user base.
I would be relieved to see the design community embracing measurement more.
Alternatively, you have giant corporations like Facebook, or any corporate level software service at scale, that is incredibly numbers focused around (hopefully) user retention and user engagement. But they very quickly blur the line between what's good for a user vs. what's good for that person as a cog in the overall system. For example, how much viral worthiness someone has, or how much ad revenue they're generating by being sucked in.
There is a really viable middle ground between those two where you identify goals aligned with your users’ goals. And once you align your incentives, track your mutual progress.
There is a really viable middle ground between those two where you identify goals aligned with your users’ goals.
If your question is specifically what numbers should you look at for user success, and how has my appreciation for that changed over the years, then I would definitely say I skew away from the normal funnel-type metrics. Things like acquisition, activation, revenue, referral, etc. Those are all very company-centric. You want to keep an eye on those, but you don’t want to design for them.
Things like acquisition, activation, revenue, referral...You want to keep an eye on those, but you don’t want to design for them.
You want to design for assisting people, in whatever context of progress they're bringing your product into. And ideally, you want to make your product or service as self-aware as possible, so you can identify what change somebody is looking to make.
And when you pay attention to what people are actually doing with your product, the activities they’re engaging in are inherently measurable. Because that's all user activity being saved to the database. Your product knows what your users have been doing. Why doesn't your reporting know?
You brought up my next question, which is around understanding the context users bring with them and the context in which they’re trying to achieve their goals.
How do you go about identifying that context? How do you understand context enough to design for it?
That's a million dollar question for sure.
Most of my experience is working with companies who already have a product and already have a user base. And I provide an alternative perspective. The nature of the challenge you’re describing is very different there vs. coming up with a thing to focus on from scratch, in the sense of starting a whole new product.
If you already have a product with people organically signing up for it, and you brought me in to look at the onboarding part of that experience, this is what I would be paying attention to: can we identify through qualitative research and quantitative research what is on people's minds when they're signing up for the product?
Let’s say we work at AirBnB and, for some totally unexpected reason, it turns out a lot of people are renting AirBnB places to celebrate their anniversary. (I don't think that is very likely at all.) But let's say there is some sort of non-obvious insight we gain from user research, which is hopefully what you get from research. We found out a lot of people were signing up within the context of wanting to celebrate their anniversary.
Can we identify through qualitative research and quantitative research what is on people's minds when they're signing up for the product?
That to me would be a meaty thing to design around. It's not just, “How do we get more people to do to book places to stay?” or “How do we improve conversion rate and optimize our check-in funnel?” although I think those would certainly be worthwhile, too.
But specifically, how do we situate this whole service within the context of what someone is trying to do? In a way that lets them know they’re making progress on what they care about, in a Jobs to be Done sort of way.
Kathy Sierra has been a major influence in my thinking in this regard with her notion your software gives people superpowers or capabilities.
...how do we situate this whole service within the context of what someone is trying to do?
If you think of onboarding as somebody going through the process of attaining those super powers, you have to wonder, why do they want to fly or why do they want to be invisible? What is the thing they’re trying to do with the invisibility they’re acquiring from us?
I love that that concept. Say someone gets a cape. I mean, maybe you just want to wear the cape. But probably not. You want to use the cape; you want to be able to fly. If companies asked that question—what do you want to do with the superpower?—more often, they’d have greater success identifying the end goal. And ultimately, build a more valuable experience.
How would you advise teams define onboarding and think about it on an ongoing basis?
On a paradigm level, I would encourage people not to think of their software as a product. Period. Because it leads you to questions like, “what is a good introduction to a product?” Or, “what should our onboarding on-ramp experience be?” and “what kind of feature tour should we put into it?”
Those are all understandable questions when you’re operating within the product paradigm. But those aren’t useful for top-line business-level outcomes your company is looking to generate.
In economics (which I will certainly not claim to have any expertise in) there's a concept of goods vs. services. In the goods section, you’re manufacturing tangible objects for sale. Whereas services is more inherently nebulous and more outcome-focused. Designing a product as a material good leads you to ask all sorts of the wrong questions. Instead, you want to be a designer of outcomes for a user, your user base, or people in general.
It's much healthier to think of your product as an intangible value delivery system. For example, you just have this code sitting on a server. It doesn't exist in any genuine capacity, outside of when a real user is requesting it onto an actual device and actually looking at it and engaging with it.
It's much healthier to think of your product as an intangible value delivery system.
Instead of asking, “how many intro videos did they watch?, focus on saying, “what is an outcome that's good for the company? What is a user activity that generates value for the company?” Start there. We don't have to be Mother Teresa selfless or bend over backwards for users all the time. We want there to be a mutual exchange, so we can even start with what's most important to us.
Let’s say there's a company that does invoice management, and they have a 14 day trial. I’d immediately focus on not only how to change the onboarding, but what the definition of performance for onboarding even is. Because what you want to do is convert brand new signups into people who have just upgraded their trial.
...you want to be a designer of outcomes for a user, your user base, or people in general.
There are two ways to think of the space between those two states of existence (new and paid) The first is as a tour someone should take, and then you just hope for the best. The second is as creating a whole system of transformation from one state to the other. The questions associated with the second are a lot more interesting.
I would be trying to figure out what actually has to happen, in reality, to go from brand new user to upgrading out of a trial. Because those are what your growth levers really are. And then it's a question of, “how do you create supporting interfaces to facilitate that?” And those could be blog posts, or certainly something like a lifecycle email campaign. There are all kinds of different options. It doesn't only have to be solved within the product and certainly not all at once.
That’s interesting you bring up things outside of the app experience. In the same way outcomes can exist outside of your app, and your app plays a small role in helping somebody achieve an outcome, helping users get close to an outcome can be outside of the product too.
How do you think about onboarding for products that aren't used often? How do you think about ensuring they see success and value as quickly as possible?
Well, boy, that introduces maybe more questions than answers.
But where where my mind immediately goes is, first of all, is the business model built in a way that acknowledges and aligns with that cadence of activity?
...is the business model built in a way that acknowledges and aligns with that cadence of activity?
If you're trying to have a hockey stick growth model, and your retention is good but your engagement is low because it's not something that comes up often in someone's life, you're going to want to build a business model that accounts for that in a major way.
You're not going to sweat what day three retention looks like, because it's something people don't use every three days. If you do sweat that, you're just going to stress yourself and your users out, trying to manufacturer relevance where relevance doesn't already exist in the users’ lives. So one thing I would say is try to identify, as much as you can, the inherent value somebody is looking for.
...try to identify, as much as you can, the inherent value somebody is looking for.
If you're looking at something that isn't used very often, I think a much more relevant question is: every time that situation arises in someone's life, are you the choice they make for resolution? But that doesn't mean you need to introduce that situation into their lives on a daily basis either.
A concrete example for me is, whenever I'm trying to figure out where to go eat, especially in an unfamiliar area, I turn to Yelp. Yelp owns the number one SEO ranking in my head for “Where do I go get food?”
...every time that situation arises in someone’s life, are you the choice they make for resolution?
And there are times where I might go months without using Yelp, but it doesn't mean they've slipped off my radar. It might just be because I've gone a long time without needing to think hard about where I'm going to eat. Or there are times when I'm traveling to Europe. Yelp doesn't necessarily help me because I'm getting, you know, reviews from fellow American travelers who want french fries that taste more like McDonald's. Yelp turns out to not be valuable then.
But it's not a question of brand affinity, or user loyalty, or anything like that. It's a question of: are they the preferred way of addressing that particular situational concern within the context of that situation?
Let’s assume someone has a product with some users, and they're looking to improve how the product delivers value to those users. Where would you suggest they start?
I'm biased toward user onboarding because user onboarding is one of the best places to do focused customer research. The concrete is still wet, so to speak, and users are in the moment.
They're not a customer that's been on autopilot with you for eight months, and you're asking them to speculate retrospectively what their value probably was. With user onboarding, you can ask people right in the heat of the moment.
From a customer development or user research standpoint, I love asking people a really simple question after they sign up for anything. I even do it off and on myself with a newsletter I have. Every day I get a list of people who signed up the previous day. And it's not hard to shoot an email to each person individually and say, “Hey, thanks a lot for signing up. Why is user onboarding on your mind? What is causing this to be relevant?” or “Why is this of interest to you right now?”
Pay attention to the words they use and how they describe their situations.
These might sound like icebreaker type questions, but it's the biggest thing I want to know. Not just on a one-to-one basis with each individual user, but also to see patterns across the user base overall.
And so I would say—as much as I can generically extend my advice—there are few barriers, at any decent software company, to getting a running list of people who are signing up. Send them a personal outreach email, and ask them why X is on their mind.
The sort of answers you get are incredibly valuable, as far as passive aggregate customer development is concerned. But also pay attention to the words they use and how they describe their situations. Use those as clues for areas to conduct further research in. Or take those words, terms, or turns of phrase, and use that as marketing copy, etc.
Not to mention, of course, you're kicking things off in a really warm and hospitable way for each user you reach out to. And a lot of times those individual relationships can also lead to a real dynamic of abundance.
If you do proceed, you can begin picking up on patterns and saying, “Hmm. We have this invoicing software, and it seems like whenever we welcome people and ask them why invoices are on their mind, it's almost always because they just got their first client or the person who handled invoices left the company.” You will probably find patterns to the contextual situational desires people have.
And then my recommendation would be to really think about how comprehensively helpful you can be. Pick the most frequently identified pattern, and think about how you can curate an experience to be more in line with the particular aspiration people have.
...think about how you can curate an experience more in line with the particular aspiration people have.
If I were investing my onboarding dollars in the highest impact, highest leverage, most growth-oriented capacity that I could, it would along those lines.
This interview has been edited and condensed for clarity.