<< All Episodes

Season 1, Episode 6:

Getting Practical with Segmenting by Intent


Apple Google Spotify Overcast RSS

Episode Summary

Ok, enough with the high-minded design philosophy: let's get into how Value Paths can be IMPLEMENTED.

The gist: When someone arrives at your offering, it's because they're seeking change in their life. There are patterns there, and you can design for the good ones.

Which life-changes do you want your product to be REALLY GOOD AT? How can you DELIVER on them? And how can you tell that your efforts are PAYING OFF?

Today, we explore all of the nooks and crannies of that topic with the practical example of Segmenting by Intent.

📚 References

🔤 Transcript

Samuel (00:10):
Hi, I'm Samuel from UserOnboard.

Yohann (00:13):
And I'm Yohann, also from UserOnboard.

Samuel (00:16):
And today as we promised, we are getting more practical with our recommendations. We're taking things out of the realm of philosophical pillars, and we are now getting real by talking about how to put this into action on a tactical level with one of our personal go-to tactics, which is segmenting by intent.

Samuel (00:39):
But, before we dive into things, we have mentioned in the past that we are open to feedback. And I did see that there is an Apple Podcast review by someone named brainslikepickles, who mentioned that "premises," the plural form of "premise" does not rhyme with "please." And my initial response was, "yes, it does." Or at least, "it does to me." But then I looked it up and it turns out that any plural version of a Greek word like "thesis," becomes "theses" (rhymes with "please"). But premise is a Latin word, and it don't work that way for Latin words. So it's premises (rhymes with "nemesis"). So my UserOnboard promise is that I will attempt at every turn to now use the word premises instead of premises, and shout out implicit to brainslikepickles for the feedback.

Yohann (01:36):
I didn't know that. Thank you so much brainslikepickles.

Samuel (01:41):
Yep. So this is the power of listener feedback in action. But anyway, we promised to get more practical. And I think that we should get practical. So, where we last left off was walking through the three pillars of healthy growth as we see them. Which are roughly pillar one, Path Design. Get better at getting people from point A to point B. Pillar two, Performance Valuation. Measure to make sure that you actually are getting better at getting people from point A to point B. And then pillar three, Beneficial Outcomes, is all about picking the right point B's to get people to.

Samuel (02:16):
So let's say that you've gone through...especially that last pillar of identifying beneficial outcomes and your team has some outcomes that do two things. One, they produce value for the user. Two, they produce value for the business, or at least they're strongly correlated with producing value for the business.

Identifying which user outcomes correlate with revenue

Samuel (02:35):
So if you have some beneficial outcomes in mind and you don't necessarily have the ability to test how strongly correlated they are with revenue because you don't know which users desire which outcomes or not, you're not really clear on how they would differ from a performance standpoint because you don't know how to separate them to begin with, what we recommend doing is segmenting by intent early. So the idea is to insert a question or series of questions into your activation process that tries to triangulate what it is — which goals or outcomes — that that particular user is there to achieve.

Yohann (03:22):
I agree. And before we go any further, I just want to make a quick point about something you mentioned before: Identifying point B's.

Samuel (03:30):
AKA, beneficial outcomes.

Yohann (03:33):
Yes. You know how dictionary writers — what are they called? Is there specific term for people who write dictionaries? I'm sure there is, but I'm forgetting what it is at the moment.

Samuel (03:48):
It would be in the dictionary, I guess. Right?

Yohann (03:50):
Right. But let's just call them dictionary writers for now. Or dictionary editors is a better word. I was watching an interview with a dictionary editor recently. And she was like, "When we add words to the dictionary, we're not trying to be prescriptive. We're not trying to tell people what words mean. We're trying to be descriptive. We're trying to capture how a word is being used."

Yohann (04:15):
And I saw a parallel there with Beneficial Outcomes. When you're sitting in a room internally, just people in the company talking about what outcomes the product supports, you are being prescriptive. Because it's not really informed by what users are actually doing and how they're actually using the product. And that's a good place to start. But the idea is to get more descriptive and less prescriptive with the Beneficial Outcomes that your product supports. What can you actually notice about how people are using the product, about how they're speaking about the goals that they're pursuing with the product, and start to work that language, and that thinking? And that situation, that target situation. How can you start working all of these things into your everyday operation?

Samuel (05:12):
I completely agree. And I love the analogy. The takeaway for me there is that we place more of an emphasis on the reasons that people already have, that are driving them to engage with your product, rather than the reasons that you impose upon the users by telling them that they can now do X, Y, or Z. It's really up to whether they want to do that or not. If they don't want to, if there's no user innovation there, it's not going to power the engagement. And if your revenue is dependent on sustained engagement, then that's not a good mix. So the idea here is to identify the outcomes that are already taking place or that your users are already coming to you for help with, and get really good at the ones that also strongly correlate with powering your business model.

Yohann (06:06):
And it's not a subtle point to make. Because in our industry, we often talk about manufacturing desire. We talk about 'convincing' people that they need our product. And that is the antithesis of selling umbrellas in the rain. When you're standing out in the rain, you don't need to convince someone that they need an umbrella. They are coming to you for an umbrella. And those are the kind of outcomes you want to uncover.

Samuel (06:34):
Yes. As we've said, it's "location, location, location," but for time instead of space. Although I fear we're getting a little abstract at this point. So we want to make sure that we're keeping things nice and tactical. So let's try to stay on track.

Samuel (06:47):
So the basis that we're working from is if you have identified beneficial outcomes that you believe users genuinely want, and that you believe will power your business model or correlate with revenue production essentially, then it's really a question of digging in, and investigating, and putting it through its paces, and seeing if that truly is a lever that you can invest in. Invest in helping your succeed more at that particular outcome, and be able to reap the business correlated rewards that come with it. Fair?

Yohann (07:25):
Yeah. Well put.

Samuel (07:27):
Okay. So when we talk about getting tactical, what specifically we're suggesting is that if you can come up with maybe let's say four beneficial outcomes, three to four beneficial outcomes that you think would be representative of maybe not necessarily the majority of your user base, but at least a big chunk of your user base is probably coming for outcome X, outcome Y, or outcome Z. And if you can come up with some pretty good guesses through the three pillars that we've already outlined, you can start putting them to the test right away by segmenting by intent, which is the theme of this episode.

Samuel (08:11):
The first step that we recommend here is to insert a question or a series of questions into your activation process, asking people which goals they themselves would like to pursue. And even if you're not doing anything else — there's no personalization afterward or no even recognition of them answering the question or anything else — all kinds of Beneficial Outcome things can happen from there. But even if they didn't, you would still be benefiting massively as a business for two main reasons, or two main areas of measurement and analysis.

Samuel (08:55):
Now I'm wondering if plural of analysis is analyses. Is that right?

Yohann (09:01):
Latin or Greek? I don't know.

Samuel (09:08):
All of a sudden my brain was just like, "Wait, is it analy-sees? Or is it analy-siz? I think it's "sees." Anyway.

Yohann (09:14):
Yeah. I think it's "sees." It's so funny how our guts are more tuned to Greek and Latin than our brains are.

Samuel (09:23):
That's a really astute observation, #embodied-cognition.

Yohann (09:29):

How your business benefits from segmenting by intent

Samuel (09:30):
But anyway, we're talking about two main measurements or areas of analysis, one of which being: being able to see the distribution across the three different outcomes. To be able to say for the people who came in and said that they were seeking X outcome, how representative is that of our user base? Is it 80%, or is it 8%? Because those tell very different tales and give you some early clues around what is actually driving engagement rather than what you think is driving engagement, or even worse as Yohann pointed out what you have, "decided" is driving engagement.

Yohann (10:16):
And it's a completely separate thing to correlate which of those outcomes is most contributing to your revenue as well. Because it might be the case that 80% of your users are pursuing outcome X, but that the 20% of users who are pursuing outcome Y are contributing to the majority of your revenue.

Samuel (10:35):
Absolutely. And that is point of analysis number two. Which is not only what is the distribution across the different outcomes like, but how performant are each of those different outcomes. Because you could very well find something where you've created something that your traffic happens to desire at an 80% or eight times out of 10. But that it's only really good for the people who are the two times out of 10, exactly like you said.

Yohann (11:05):
Right. So thinking about practicalities here Samuel, I have a question.

Samuel (11:10):

Yohann (11:12):
What if a user is pursuing a combination of outcomes, or they encounter these intents and can't separate one from the other?

Samuel (11:24):
I should have addressed this earlier. Yes. I appreciate you getting us back on track in that regard. So if we went back in time a little bit and we were talking about inserting a question or a series of questions into the activation process, my recommendation would be to make those the options that you present, the point B's or Beneficial Outcomes that you think users are coming for, present those as optional check boxes where you say, "Which of the following brings you here?" essentially. That way, you're not only supporting the user's sense of autonomy in the sense that they are the one who is in control. But you're also letting them opt in rather than having a real strong effect of, "Well, I have to choose one of these three. So I guess I'm just going to go with this one," or whatever would skew your results in that way.

Yohann (12:20):
Right. And does this affect the process downstream?

Samuel (12:26):
How do you mean?

Yohann (12:27):
I mean if users choose, if they opt into multiple beneficial outcomes, are you still able to sift through the data later on?

Samuel (12:39):
I would absolutely say so. I mean, the way that we are effectively recommending to measure this is to insert the detector to figure out what Beneficial Outcomes are actually being sought. And to measure the part about how strongly they correlate with revenue is, you want to not only see how well those people go on and convert. From an onboarding standpoint, you might be thinking of your trial to paid conversion rate or something like that. But you also want to be looking at the kinds of plans that they're opting into, whether they're paying monthly or on an annual basis. And if it is a monthly basis, how many of those people are making it to month two, and how many of those people are making it to month three and month four. And if you're measuring things on a user by user level where you can detect A) which users which want which outcomes. And B) what the revenue performance of those users are, then you can very quickly identify a big delta at times in performance.

Samuel (13:42):
Like, in our example that we always use of Invoicer, the software for sending invoices. I would imagine that in this hypothetical company if it had a free trial, the people who successfully get an invoice paid during the 14 days of the free trial would be worth much more money to the company than a user who did not even create an invoice. Somebody who has made that much progress with the product would be much, much more likely to be a good, steady, solid customer, even just based off of little data point regarding a strong indication of them receiving value.

Samuel (14:20):
So if we can identify what kind of value people are seeking on a broad level and be able to see how the performance of people who seek those different kinds of value plays out in terms of revenue, that gives you a very powerful indication of which beneficial outcomes again in a low-fi sense, probably have a strong correlation of revenue. And therefore, if you can increase your "free throw percentage" of being able to deliver on those outcomes, you will likely result in a significant uptick in revenue to go along with it.

Yohann (15:01):
Can I put on my evil mustache and ask a few questions here?

Samuel (15:05):
I love it when you break out the evil mustache. I'm ready for all of your contrarian questions.

Yohann (15:11):
Okay. So in the example you just used of Invoicer, if you know, if the data tells you that people who have an invoice paid go on to become better customers, why not just focus on that action and skip the segmenting altogether?

Samuel (15:34):
I completely agree that you can pay attention to post-acquisition user behavior, and come up with things that strongly correlate with revenue production. That is not really new territory. There's the famous examples of Facebook having seven friends in three days, or Twitter following 30 people, or different user milestones that you can coach people to. And in the Invoicer case, we're kind of stacking the deck in our favor because the value of having an invoice paid is pretty inherently defined. Whereas the value of following 30 people, or getting seven friends connected, or whatever those proxies of the past might have been don't have an inherent motivational aspect that you can explore, and better understand, and therefore better serve.

Samuel (16:29):
So what we're talking about is saying that even in the Invoicer example, even if we have an activity like getting an invoice paid that strongly correlates with revenue, we can't currently in our imaginary company, we can't tell the biggest reasons why somebody is trying to get an invoice paid to begin with. And if we can identify the patterns that take place in a bigger context than just getting the invoice paid, we can then present ourselves with a lot of opportunities to lay out a path for people to succeed in the thing that they're coming to you for help with in a way that transcends a one-size-fits-all user experience for a single product. Is that answering your question enough?

Yohann (17:18):
It is. It is. Let me just summarize what I'm hearing from you. Getting an invoice paid is a sub-outcome. And right off the bat, you're distinguishing between sub-outcomes that are just purely engagement based, like get to 30 friends. And sub-outcomes that are inherently motivational to a user. So getting an invoice paid for example — it's very likely that that sub-outcome features on the critical pathway of that user's value path.

Samuel (17:52):
And it's likely to be viewed as a beneficial milestone on that pathway for the user, but for the user, it's never about what happens in the app. It's about what happens in their life. And the more in tune you can be with the different kinds for life reasons that drive people to go to Invoicer to get an invoice paid, the more likely you can perform at not only helping them with the specific thing that they came for, but also doing other things that strongly correlate with being good customers, like getting invoices paid and things like that too.

Yohann (18:26):
Right. So if you optimize for a particular sub-outcome without looking at that bigger reason why, you have no guarantee that that optimization is going to result in a better situation for the user later on. You're still leaving some things up to them by optimizing for a sub-outcome rather than the Beneficial Outcome?

Samuel (18:48):
Yeah. And I don't think that we're against optimizing for in-app behaviors at all. I would just say that if that's the limit of the attention that you're paying to what is driving those behaviors, then you're just leaving a lot of opportunity on the table.

Yohann (19:04):
Right. I think that's a great distinction to make. That designing for Beneficial Outcomes is not about stopping whatever it is that you're doing and doing something new. But it's about addressing all of the points of support that you can provide. And addressing that providing that support will result in more business value.

Samuel (19:24):
Exactly. So from a practicality standpoint, I imagine some people are saying, "Well yeah, but I can measure user behavior correlations with user engagement, or revenue production," or whatever it is that you're trying to generate. "But I can't measure the life-reasons that are bringing people to the product." And that's exactly why we recommend inserting the segmenting question early in the signup process is so that even if you don't know how well you're necessarily even delivering on the different Beneficial Outcomes that you're advertising, you can still draw a correlation between the desiring of those outcomes, and the performance that people have as customers. And that is a really, really valuable insight and piece of data to be able to work with and provides you with a growth lever that essentially nobody is currently using right now.

Yohann (20:21):
There's another small point that I want to make. So far, we've talked about segmenting by intent as being one step that you insert into your activation flow. And in this step, you have beneficial outcomes listed out for users, and you have them opt into a combination of them depending on their goals. But what if your users aren't clear on their own intents? It can happen. And it happened to us when we ran a project like this with a company recently. And what we did in that case was instead of having just one screen ask one question, we had a series of screens ask a series of questions. The first starting at a high level, and then slowly boiling down what users really want to do.

Yohann (21:14):
So it started with for example, "Are you here to do X or Y?" And these were very thematic, marketing value prop level outcomes. They weren't specific or discrete points of value, like the ones that we are trying to design for. And when users said that they were interested in X, we gave them an X specific further set of questions. Like if you're interested in X, then do you want to do one, two, or three? Through that process, we helped them think through exactly what they were expecting the app to provide for them.

Yohann (21:56):
Afterwards when we were looking through the data, we discovered situations where someone wanted X, but the discrete outcome that they were pursuing was not something we would've tied to X to begin with. And those kind of insights are very eye-opening. Because sometimes, people say one thing and mean another. And the more vague your outcomes are, the more the chance that this can happen. Does that make sense?

Samuel (22:30):
I think it was very well put. In fact, I don't have a whole lot to add on top of that, other than the fact that I agree with you. And that the name of the game is always to find the granular distinct outcomes where users can recognize the fact that they have received value, and you can measure whether they have actually received it or not. If a discrete point in time is what we are ultimately, those are the juiciest Beneficial Outcomes. But as far as starting at a more thematic level and then working your way down to specifics, I completely agree. And I think that that is generally speaking, the way to do it.

Yohann (23:12):
It's very useful to think about which discrete Beneficial Outcomes you can put on one screen. But just in case you can't do that, just in case you can't reach consensus internally, to think about not being prescriptive and to be descriptive running with a series of questions to help users be discrete. And "discrete" to mean point out individual moments in time that they are pursuing, can be really helpful.

Samuel (23:44):
Not trusty with a secret.

Yohann (23:48):
Yes. The language can get confusing sometimes.

How USERS benefit from segmenting by intent

Samuel (23:54):
Yeah. And one other thing that I would mention while we're talking about the sequence of questions where we're helping people drill down into more discrete outcomes that they're actually seeking is that every single time that somebody is opting into an outcome, whether it's broad and thematic or really granular and concrete, every time that you are just demonstrating to them that this is something that this product is good for, that is sending them a really encouraging signal. You are letting people know, "Hey, the thing that you are here to do right now is something that we recognize as important, and that we have focused on providing at a high level." So it's reinforcing the anticipatory value that somebody is going to be getting by getting really specific about how exactly it is that you plan on serving them in the thing that they're coming to you for help with.

Yohann (24:49):
Nothing gives me more confidence as a user than that feeling that the thing that I'm using was made for me.

Samuel (24:56):
But not even made for you. That's the distinction I would make is that it's not made for you as a person. Because that would weird me out very much if somebody made an app specifically for me, but it's for people in my situation. If you understand the situation that I'm in, that is what really brings me calm as a user.

Yohann (25:14):
Right, right. Fantastically put.

Samuel (25:18):
Thanks. So we've already said that even if it doesn't benefit the user, it's already providing a wealth of value to the business in terms of giving them information about one, what the distribution of the outcomes across your user populous is. And two, how well each of those different segments converts and produces revenue on an ongoing basis.

Samuel (25:42):
However, it is also better for the users in a number of different ways. That's kind of the whole essence of value paths as a concept is to find ways to align your outcomes so that you can both be pulling in the same direction as a user and as a software company.

Samuel (26:01):
So for example, ways that it can benefit the user. We just touched on the reinforcing of the value proposition by getting more and more specific as people drill through the different screens. But even if you go with just one screen and you just present a high level of let's say four different beneficial outcomes that you think somebody might be there to accomplish, then it also is already giving the users the benefit of signaling to them this is what's important here. These are what we're really good at. And if the software is a little cagey about it and it's like, "Yeah, let's just focus on how cool our interface is. And let's not talk too much about actual real world outcomes. It's kind of a drag." That's not what you want to be getting as a user. You want to be receiving an experience that is tailored to your situation and helping you actively get to the situation that you desire, rather than presenting you with an interface and leaving it up to you.

Yohann (27:02):
Right. Apart from the interface, another common onboarding tactic that really doesn't sit well with me — and stop me if I start to rant about this, but —

Samuel (27:15):
Spill that tea Yohann.

Yohann (27:18):
The sales questionnaire. The series of questions that usually are part of a signup process, that have absolutely no impact on the user experience, but exist solely to give the sales team more material that they can use to sell to me later on.

Samuel (27:37):
Yeah. Completely agreed. Because I also imagine people thinking, "Well there's already so much friction in our activation flow. Should we be adding more questions? Is that really what the user wants?" I would say yes. Because having looked at as many onboarding experiences as we have, we have seen so many times that there's the sales screen exactly like you're talking about where people have to snitch on themselves and let you know how many employees are at their company, and all kinds of things that have nothing to do with what they came there to do. It's just this disempowering, one-sided, extractive experience where they have to comply with your demands or leave. And ideally, that's not an ultimatum you want to be giving users, because a fair amount of them are going to choose to leave.

Samuel (28:27):
And so instead if you're asking them questions, let's talk about you. Let's talk about your situation. Let's talk about where you want to go. Those are the kind of things that users are motivated to engage with, because they are a breadcrumb trail of them getting to where they hope to get to when they came to the app in the first place.

Samuel (28:46):
Users aren't dumb. It's not like you can just force them to rat on themselves so that your sales team can decide on whether they're worth their time or not. Instead, you want to be demonstrating that you are a good servant to them. You're not the one in charge. The user is in charge. We've known this for 20 years. But we still design in such a way where we say, "Then we make the users do this," "Then we make the users do that." And I don't think that that is right at all. I think that you help users do things. And any time that you come in and try to act as the dominant enforcer, you are going to get met with a black screen. So, I mean that's my personal take. Maybe I ranted the rant that you were about to.

Yohann (29:32):
You did the rant for me. I was just going to say.

Samuel (29:37):
But yeah, anything to add to that? Should we pile on the fire while we're here?

Yohann (29:43):
No, I think that covered the point adequately.

Samuel (29:48):
So sticking with the tactical theme of this episode and focusing on practical recommendations that we can make, you can also improve the user experience for the users significantly beyond point that we are talking about here. Where we're not just offering to them the fact that we can help with these different things and then saying, "Okay, thanks for the information. This is very valuable for the company." But you can also then segment the experience that that person has with the product, in whichever way you think best suits their pursuit of the particular outcome that they have told you that they're seeking.

Samuel (30:27):
Like if we imagine a persona segmented product where you say I'm a marketer, or a developer, or a designer, and then you get maybe a slightly personalized experience afterward. You can do the exact same thing, but crank it up to 11 when it comes to helping them pursue outcomes. You can get much more granular about which specific outcomes they are looking for, like Yohann mentioned a moment ago. But you can also, regardless of how granular it is, design a different layer of experience on top of your product for whoever raises their hand when you're asking, "Are you here for X, or Y, or Z?" You can say, "Okay, you're here for X and Y. Therefore let's come up with..."

Samuel (31:14):
Like for example going back to user onboarding again, a lot of times companies will come to us and they'll say, "We've heard good things about using a to-do list in our onboarding. So we would like to do that." And the most important question is not, "Well, should you pre present the to-do list as a little toaster in the bottom right? Or should people be able to activate it from a menu?" None of that is as important as: what does the to-do list tell people to do. What is the sequence of actions that you recommend people take? And how do you guide them to a meaningful outcome, preferably the meaningful outcome that they just stated that they were there for? And there's a lot that you can do for somebody who says that I'm here for outcome X, that people who say I'm here for outcome Y never see.

Samuel (32:06):
And vice versa. And then the same for outcome Z, where you can almost think of it like you are creating, it's almost like you're building on your own product layer of the experience. But using it in such a way to guide people through the process that they have currently at hand. And if you can get really good at those, it can inform your acquisition by helping sharpen your marketing and positioning and signaling to your users. It can help in the form of having well matched testimonials. It can help in the regard of having more fitting copy. Where instead of having button copy that relates to outcome Y, you have button copy that relates to outcome X. That's presented to only when people say that they're in search of outcome X.

Samuel (32:54):
There's a lot of different things that you can do to customize the experience as small as copy changes to as big as creating courses or entire different guided experiences through your product that help people arrive at a meaningful, real-world outcome that is bringing them to your product to begin with. And if you do that to enough of an extent, you can almost think of what currently exists as your product experience. Again, I use the term one-size-fits-all user experience of your product. And you can think of that as like the fallback version. That's like the sensible default, where you're presenting them with the version of your product that isn't augmented toward a particular outcome or another. And instead, is just sort of the vanilla version of your product.

Samuel (33:43):
I think that right now, we're trying to stuff all of this value and competing outcomes from different user types or situations into a one-size-fits-all experience. And I think that the users are worse off for it, and it creates more confusing interface and a number of other things on top of that.

Samuel (34:07):
Especially when that one-size-fits-all experience is designed in a way where you have executives issuing mandates like, "Here's our 12 month roadmap or different things along those lines." Or you're creating features with the hope that it will unlock some segment of your market. But without thinking about ... like again, we use the example of pancakes all the time. There's a time when you need the whisk, and there's a time when you need the spatula, and there's a time when you need the fork. And those are very different times in the pancake making process. And in a similar way, you want to be handing the right tool to the user at the right time in their process that they are already engaged in before your product even came on the scene. That's the ideal user experience from the user's perspective is, "This meets me where I am and takes me to where I want to go. This is perfect." So does that resonate with you Yohann?

Yohann (35:03):
Absolutely. We've covered a lot of ground, but let me just take a moment to summarize. What we are trying to say in summary is that if you have the resources and the organizational capacity to design very personal experience based on the situations of your users, go ahead and do that. But if you can't, that doesn't mean you have to abandon segmenting by intent. There are some easy ways in which you can segment the user experience based on situation.

Samuel (35:40):
What kind of ways Yohann?

Tactic #1: Re-ordering the steps in the path

Yohann (35:42):
Way number one. Like Samuel said, you can reorder the steps in the process. So based on what someone is seeking, you can make the nine step be the second, or the second step be the fourth. We're talking about a choreography of actions here. Path Design is very much about figuring out the timing of the different things that users need to do so that they can get to the places they need to get faster and easier. And a quick way to segment by intent is to reorder that sequence of steps.

Samuel (36:18):
Or to remove steps entirely. If you think you have a hand of playing cards as different steps that the user can take that you can help them with. And they say, "I'm here for diamonds. I don't care about spades." You can be like, "All right. And we are not including spades." You can use it to remove complexity from the user experience as well. I mean, I don't think either of us after studying user onboarding for as long as we have would recommend just taking people through this rapid click next 20 times tool tip tour that just makes people look at a bunch of different features. You want to take them to the feature that they need right now.

Yohann (37:01):
Right. On this subject, before we get to more tactics, on this subject, Luke Wroblewski has a fascinating design pattern called gradual engagement. And gradual engagement is about revealing the app to your users in stages and not all at once. Tool tip tours that have 20 steps one after the other, it puts a lot of pressure on the user to remember all of that information in order to be successful. It's almost like a memory test. And users are not coming to your app to take a memory test. And gradually introducing them to the things they need in each moment is a very autonomy friendly way to teach them how to get to where they want to be, which that's something you're an expert in, and they are not.

Samuel (37:57):
Exactly. That's one of the other aspects that we should definitely hammer on here is whatever the Beneficial Outcomes that you present to the user, you should be good at being able to deliver those. You should know those inside and out, and preferably at a world class level. Because that's ultimately what your differentiator is, whether you realize it or not. It's not because you have a cool logo, or because you have a nice color aesthetic, or you pick the right font, or things like that.

Samuel (38:25):
So ideally, you should know the ins and outs of this process significantly better than the user does, because it could be that it's the user's very first time doing this, which is what's causing them to find a relationship with an app like yours to begin with. So the idea of actually getting people to the outcomes that they desire through simple means, I completely agree with you on and shout out Luke W. for sure. I've learned so much from Luke Wroblewski's work, that I definitely appreciate the fact that he has come up in this episode. And there's plenty more there where that insight came from.

Tactic #2: Using situation-driven in-app copy

Yohann (39:02):
Okay. So easy ways to segment by intent number two: use situation-driven copy more. So copy is a great way to have an impact on the user experience without putting too much effort in. You don't need to change the product. You don't need to design a whole new screen. It's just about changing a sentence. And microcopy can come in particularly handy here, because Samuel mentioned button copy. But I want to bring in this added element of microcopy, the names of features, the phrasing in the to-do list. These are all opportunities that are just sitting there that you can customize to a particular situation.

Samuel (39:50):
Headlines, absolutely. Call-to-action copy for sure. Error copy. These are all areas where you don't need to go in and have your development team build out a new feature over the quarter. But that you can say, "Okay, we're going to ask people empathetically what it is that they're here for, hoping we can help them with." And then we're going to at least give lip service to that, just by reinforcing the fact that what we're asking them to do comports to the thing that they're trying to do, and that there's some sort of relationship there by including it in the keywords, and the helper copy, and things along those lines. I completely agree.

Tactic #3: Segmenting onboarding emails

Samuel (40:35):
So this might be tactic number three, but to have separate onboarding email sequences for the different segments as well. A lot of times-

Yohann (40:46):
It was. It was tactic three. Ding ding ding.

Samuel (40:54):
But yeah, there are a lot of times where we work with companies in a growth capacity. And we got to be pretty squirrely with the kind of changes that we can make to the user experience. There's a lot of times a 500 pound gorilla of 'product' or capital P product that has a roadmap of its own. And you got to be resourceful, guerilla warfare kind of in going in and trying to strategically and surgically improve the user experience with whatever levers you have available.

Samuel (41:26):
And one of the most commonly available levers that we have seen over, and over, and over again is that companies, even though product might be a total turf war, nobody cares about lifecycle emails. Or at least they don't to the degree that they should, even close. And generally the instrumentation is there. And sometimes we've even heard, "Oh yeah, our lifecycle email is set up by an intern two years ago." If you have the machinery in place, you have your hand on a powerful lever for growth. And a lot of times, it's just covered in cobwebs and dust at many, many companies. So not only do we recommend investing in lifecycle emails as a growth lever period, but from a tactical segmenting by intent standpoint, you can also have either different variations of the same email sequence go out, or you can have totally different email sequences going out depending on what kind of goals people say that they're there to achieve.

Yohann (42:27):
Right. And thinking even about related copy like in-app messages, we worked with a company recently that it was the first time that either Samuel or I had seen this in an onboarding experience. They had a little box on the dashboard that just had tips in them. You remember this Samuel?

Samuel (42:53):
I do.

Yohann (42:54):
And it was great because the tips, they were making good use of space first of all. And second, they were tips that we really needed. It wasn't an intercom button. It wasn't a tool tip it. It didn't hijack the experience. It was just there in case we needed it.

Samuel (43:13):
Yeah. Kind of like you would imagine a to-do list, but in this case it was less instructive and more just FYI. Yeah. But I think that what we liked about it in particular was the fact that the items inside it were relevant to us. If it was this rotating area of tips that provided no value and had no relevance, then we wouldn't be fawning over it as a design pattern. And users wouldn't say, "This is such a cool part of the user interface. This is bringing value to me." They would think, "This is just a thing that. Okay, I guess I just don't pay attention to that because that's irrelevant." And that's it. That would be the end of the experience with that particular element for them.

Samuel (43:58):
So the key thing that we keep coming back to is you want to be leaning into relevance at every opportunity. The way you can position yourself to do that is to segment by intent to begin with. But then when you have started being able to identify which users are currently desiring which outcomes, you can also massively influence the experience that ensues in a way that doesn't impact the other users whatsoever.

Yohann (44:29):
An off the wall example that I can give here, I'm playing a video game at the moment. And usually when I'm playing video games – I'm very new to video games, but usually already when I'm playing them — I ignore the little messages that pop up on the loading screen that give you a tip. Like usually, the thing loads so fast that you don't have time to read the message, or it's a message that is completely irrelevant to you or what you're trying to figure out, or it's too advanced for where you are in the game at the moment.

Yohann (45:05):
But the game that I'm playing right now, every time you're about to enter into say a battle or a puzzle area, they give you a battle or a puzzle related tip in the loading screen. And I find myself reading all of these, because they're going to help me do the next thing that I have to do better. It just came to mind when we were talking about tips and relevance. And if someone has told you about their situation, you can make use of loading screens or empty areas in your dashboard to include some low weight copy there. Just to tell them, just to prepare them better and align them better around what they're seeking to do.

Samuel (45:52):
Absolutely agreed. So let's say in your activation process, you ask a question or a series of questions that help you identify the beneficial outcome that is driving people to your product to begin with. On the very next screen that you might be asking for them to give you their credit card information. And if you have a testimonial that happens to strongly relate to the thing that they just told you that they're there to do, it's probably more likely that the people who are there seeking that outcome who see a testimonial with somebody who says that they were able to achieve the outcome with staggering ease and proficiency thanks to the product, are going to be more likely to go through the credit card data entering process than they would be if they were just given a generic testimonial, or especially a testimonial that spoke to an outcome that they said that they weren't even here to do.

Yohann (46:51):
Absolutely. I think that covers tactic number three perfectly. So, so far we've covered how inserting a question into your activation flow can help you segment by intent. We have covered the ways that this can help you get clearer on the beneficial outcomes that your users are seeking and help you correlate those outcomes with revenue. We've covered easy ways that you can influence the user experience, even if you're not redesigning the product to cater to a particular situation better.

Yohann (47:30):
And there's one thing left to cover, which is how you can use these segments to research your own understanding of these Beneficial Outcomes. Like for example in the previous episodes, we talked about the thematic and practical distinction and how the value prop is very thematic, but you're trying to identify particular discrete moments in the user timeline to design for.

Yohann (47:56):
If you're finding it difficult to come up with those moments and you want to do some user research, the user segments can help you there as well. Use that information as a starting point to figure out what questions you're going to ask to users, which users you're going to ask these questions to. Your successful or healthy users, how have they attained the beneficial outcomes that you're trying to get new users to?

Yohann (48:27):
Depending on how you want to run your user research, the segments of Beneficial Outcomes can help you in different ways. We would love to come this in more detail, actually. As I'm talking about it, I'm realizing that it's too much to go into at this point in this episode. What do you think Samuel?

Samuel (48:48):
Don't hurt their brains Yohann. Yeah, so we've covered enough territory already. But I would just echo what you laid out there in saying that in the same way our belief is that if you can speak to what is relevant in that person's life/situation, that you can therefore make your product experience more relevant. And I would say the same goes for your research process as well. I would say that we generally find better open rates, better response rates, better attendance rates when we're talking about having people lined up for interviews and things along those lines. If it's something that if we reflect back to them, that we understand what's motivational to them upfront and reinforce throughout that experience, rather than just saying, "Hey, we're a company and we are doing user research. And you're a user. So you want to get on the phone or whatever?"

Samuel (49:44):
The example that, I think I've given this in a previous episode. But if you're giving a dog a pill, where if you get a prescription medicine for your dog, you can try to hold its mouth open like an alligator wrestler, and then whip the pill into the back of their throat and hold their mouth closed until hopefully they swallow it. But then sometimes they spit it out, and it's all slimy, and you got to do it all over again. Or, you can just put it into a meatball and the dog eats it right up. And in that same way, you're always trying to think of what is the meatball of context of life happening stuff that is driving this person to take the pill of giving us revenue and giving us their time inside this product? What's the thing that is really, truly motivational to them? And how can we harness the energy that that brings to the table at every step of the process?

Yohann (50:38):
I just want to say it's weird. Switching mindsets from thinking that you're doing people a favor by giving them access to a product, to seeing it as, "I'm giving them a pill that they have to swallow using this product is not something they want to do, but they have to."

Samuel (50:59):
Yeah. And getting clearer on the have-to-do's is really what drives your business. It's the same as the Mario image of it's helping people throw fireballs. It's not the attributes of the fire plant itself. But, everybody's got a different definition of what fireballs means to them. And if you design for everybody, you're going to be designing for nobody. But instead, you can design a fallback option for everybody, and then design an enhanced or augmented version for people based off of different major patterns that you've observed of fireball pursuing. And especially the ones that you've also demonstrated to be worthy of your time as a company, because you can draw a correlation between users succeeding with that and them succeeding as customers.

Yohann (51:53):

Samuel (51:54):
Anything else on this Yohann?

Yohann (51:56):
No, I think that's a wrap for now. I would love to cover the research side of segmentation in more detail in a future episode.

Samuel (52:06):
I sense we will have to dedicate an entire episode to that topic my friend.

Yohann (52:11):
Let us know if you'd like to hear it sooner, rather than later.

Samuel (52:15):
We need more feedback than just "brainslikepickles". Give us a shout on the reviews, and also reach out to us with any questions, concerns, critiques, or other comments to growth@useronboard.com. We would love to hear what you have to say, and we read everything that you send.