A behind-the-scenes look at building Product Tours

That’s why we recently launched Product Tours, a dead-simple onboarding tool featuring a code-free tour builder, error notifications and quick-start templates.

As our Design Lead Gustavs Cirulis shared recently, the development of the product did not come without its own challenges. So I hosted our Director of Product Management Brian Donohue and Senior Product Manager Patrick Andrews on the podcast today to get the behind-the-scenes look at how this product came to life. What were the critical decisions? Where were we confident, and where were things a little bit shaky?

Short on time? Here are five quick takeaways:

1. Think big, start small. It’s easy to dream up a sophisticated solution to a problem, but first you need to validate your assumptions to ensure you’re on the right track before going nuts with features.
2. To differentiate yourself in a crowded market, ease of use is only part of the equation. Good design and even workflows are easy to imitate (and if you build something good, it will be copied).
3. Compound product interest is a competitive advantage that’s hard to copy. It allows you to maximize a new feature’s value with minimal effort because the feature is part of this system or platform.
4. Carefully consider adding bells and whistles. Sometimes the uncomplicated product is the best product, because it’s not boring to the customers who are feeling pain. Other times, you may want to maximize reach with attention-grabbing features (especially when you’re making a big marketing push).
5. There’s a debate about dates: some feel they’re too constraining and that the team should be able to launch the product whenever it’s ready. But working toward a launch date can be a powerful tool for making progress and decisions quickly.

If you enjoy our conversation, check out more episodes of our podcast. You can subscribe on iTunes, stream on Spotify or grab the RSS feed in your player of choice. What follows is a lightly edited transcript of the episode.

Phil: Being part of the Customer Engagement team here, I think a lot about product onboarding and expansion, so you can imagine how excited I was when we launched Product Tours last month. Patrick, can you describe exactly what Product Tours is?

Patrick: It’s a tool that helps you create interactive product onboarding guides for online businesses or online products. You can think about it like a sequence of tooltips or tours that you can stitch together to highlight parts of your products or to drive users to take simple actions.

Multi-step pointer messages in a Product Tour

There are two primary use cases you should think about. One is the first-time onboarding, so this means helping new customers coming to a product for the first time. Show them around, show them the ropes, and hope that it gets them to see value quickly. Second is feature announcements, which means helping your existing customers discover, understand and get going with new features. That’s basically what it is in a nutshell.

Phil: It’s like this lifecycle – everybody should be onboarded to everything all the time. Brian, why do we think this problem space matters so much?

Brian: I think you do have to zoom out a bit. It starts with so many businesses, which are subscription businesses. For them, conversion extends far past just handing over that credit card. You’ve got to convince your customers to stick around, and to do that, you have to actually ensure they see value, and you want them to see it fast. For that, you need really good onboarding. But this is not a surprise; everyone knows this. No one’s going to disagree with that point, but so many teams struggle to make their onboarding experience anywhere near as good as their core product, so the results of this is poor retention, and basically that’s poisonous for a subscription business.

Patrick: So it’s like exercise and dieting: everyone agrees it’s a great idea, and we all want to be healthier, but it’s really difficult to actually do it.

First-class onboarding is within reach

Phil: I find that really interesting. If you’re working on the onboarding, it means you’re not working on the product and that’s a really tough decision to make. Right?

Patrick: Exactly, it’s a huge tension.

Brian: It’s a huge challenge to get teams to do this and to do it well. Few teams put nearly enough effort into onboarding. Honestly, it tends to be an afterthought, or you do it once, and then you forget it. We know firsthand how hard it is. If I look back at my couple years here as a PM, I think it’s embarrassing. I gave the absolute minimum attention. Probably not nothing, but it feels like scratching the surface of that. So I totally fell into that trap myself.

To do this well, you need to figure out your aha moments. Our researcher, Lynsey Duncan, wrote a great post about that recently. Then, you need to create onboarding experiences that are actually relevant and targeted. Unless you have a super simple product, generic onboarding isn’t going to cut it. You have to tailor it to the jobs your customers are actually using your product for.

Brian: A core part of the pitch for Product Tours is that you don’t need to use up your precious engineering and design time to make first-class onboarding. And that means not only can you do it once, you can revisit it, you can optimize it, change it and build more of it without stopping your product teams from building new stuff. That’s critical.

Tours are extremely easy to build

Patrick: So it’s making this super easy to do, and then more people are actually going to do it. It’s like net new people doing this job when they weren’t doing it before.

Phil: That really resonates to me because, as someone who’s involved with onboarding, before we had a product like this, getting something hard coded into the product to explain a particularly tricky feature could take a lot of time, and it underwent a lot of scrutiny. And as you guys know, our products are getting improved all the time. When the product changes, the onboarding changes, and you go right back to the start again. Having that timeframe reduced to practically zero is an absolute game changer for me.

This was a big investment for your team and for a bunch of other teams in our go-to-market all around the business. I think you guys started on this all the way back in July last year. How confident were you that it was worth that kind of investment?

Brian: One thing that’s interesting here is that Product Tours certainly turned into a big project and a big release, and we tried to make a lot of noise about it, but it didn’t start out that way. We weren’t actually sure if this was going to be a big, new product, and in that sense it was different from a lot of the big releases we had last year. Last year, we had a new Messenger release, we had Custom Bots, we had Resolution Bot. We knew these were big, ambitious projects from the outset. But not Product Tours. It started from a long-standing feature request – as in a years-old request. There was always clear demand of it, but it was really up to the team to figure out how big this actually should be.

“So many teams struggle to make their onboarding experience anywhere near as good as their core product, so the results of this is poor retention”

Patrick: We really knew customers wanted something in this space, but the question was what our play was going to be. Is this just a new simple message type, or are we going to do something much more sophisticated? A whole new product? And we really went on quite the journey to figure this out.

Our starting position was actually very big and ambitious. We had this whole advanced learn-by-doing pitch that enabled you to build very sophisticated tutorials. It felt very powerful, innovative and exciting, and we were pretty pumped as a team. But then when we reviewed the proposal with leadership, we actually got pushback. And the feedback was classic product-market fit feedback: “This is a great product, but how big is the market? Are we building something that’s just going to appeal to a small corner, or does the majority want something far simpler?”

Think big, start small

Phil: You were obviously pretty invested and had this great, fully featured product in mind, so how does that pushback feel when you’re that invested emotionally into something?

Patrick: Well, the feedback was great. It was really balanced. But I have to be honest, it was tough for me and for the team. We were pretty psyched and motivated about the direction we had taken. To get that course correction, even though it was definitely the right thing, was tough to take.

Brian: I wasn’t involved with the project at that point, but I remember seeing you shortly afterward, and I knew the meeting was happening. I didn’t know any of the details of the project at that point but I could see it in your face – it was clear the bubble was burst after that.

Patrick: So we decided to stop with that, and in true Intercom fashion, we picked a small iteration on an existing message type that we had. We definitely knew at this point that it wasn’t going to capture the full opportunity, but we really knew that this would help us learn more. We felt like if we dipped our toe into the water, it would really help outline some of the feature gaps and really solidify whether there was a demand for the solution and what form that solution should take. That’s exactly what we did.

“We need to start small and get feedback from customers to then be confident that the think-big was really worth investing in”

And you know, after running the beta for 10 days, it was crystal clear the list of the feature gaps that customers wanted. Even more than that, we had a huge heap of excitement for this possible product that we were going to build. It was really at that stage – which was probably 10 to 12 weeks after we initially started on this project – that it elevated to being a big project and it got the attention of the company and the big go-to-market plans.

Brian: In retrospect, this was actually a really good example of our principle of “think big, start small.” We need to start small and get feedback from customers to then be confident that the think-big was really worth investing in. Just reflecting on this, it’s surprising how much the default mindset for product folks naturally leans toward “think big, start big.” You get excited by that ambitious, hairy project, and you really have to push back against your natural tendencies here. It actually takes real discipline particularly when you think you’re onto something to build it that way.

Patrick: You get really excited, and then you can over-invest.

Phil: Exactly. It’s easy when you’re working on solving these great big problems to just get super excited and it’s such an important principle, and obviously it was great feedback from leadership.

Standing out in a crowded field

Phil: At this point, let’s take a time check. It’s the end of 2018, there’s clear confidence that the customers want this and value it. We know it’s going to be a great, meaty product release. On the flip side, there’s already competition, and they’re selling this type of software. How did that existing landscape affect how we built this?

Patrick: From early on, we felt there were two key areas that we could differentiate on with Product Tours. The first one was ease of use. With ubiquitous design patterns now established, it’s actually pretty rare that a really good user experience can be a standalone differentiator. When we looked at the competitive market, we felt there was huge scope for improvement. Trying to use these other tools to build Product Tours, it really felt I’d been dropped into Photoshop in the early 2000s. It was like being smashed in the face with configuration controls everywhere. There were Chrome extensions, complex workflows. You know, it was tough to use.

“It’s actually pretty rare that a really good user experience can be a standalone differentiator”

We really felt that some good product design could really, really improve things. We felt that we could really differentiate ourselves by making the experience way simpler and faster – and more importantly we could do it without requiring any technical skills by design and engineering. Don’t get me wrong, there’s loads of stuff we still need to improve on the experience. But the early customer feedback has proven that this was a good bet.

Brian: I think what’s interesting here is that good design as the differentiator is not actually defensible. It’s easy to copy good design and even more complex things like workflows. That’s what competitors are going to do. So copying is just a reality we all have to accept in building software. If you build something good, it will be copied. And if you aren’t taking inspiration from elsewhere, you’re probably either lying or being dumb.

But we think we have another more defensible differentiator centered around the power of our platform — centered around the whole Intercom system and what it can give you. Product Tours fits into our whole system, and that’s where we think we can change the conversation. For example, with Intercom, we tend to take things like targeting and personalization for granted. We have fantastic capabilities here and we can quickly bring them to Product Tours.

We’ve got Resolution Bot, which automatically answers simple questions, right? And now you can add a tour to an answer in this really nice design. Think of it: a customer asks you a question in your messenger like, “How do I change my payment detail?” Instead of just telling them to click a link, you can actually show them, bring them there, walk them through the process. It’s like a little bit of magic when something like that can happen.

Resolution Bot can walk your users through tours

Phil: Those walkthroughs are the most expensive kind of customer support interactions, right?

Patrick: Yeah, they’re the ones that support teams get frustrated about because they’re wasting time explaining simple how-to’s. We all wish we could build products that require no how-to’s, but I don’t think any of us have quite reached that yet. So this is a little bit of magic that we can bring with Product Tours.

Another place we can bring Product Tours is with Custom Bots. If you don’t know what use case your customer is coming to your product for, you can send a Custom Bot and ask them, “Hey, what are you coming here to do?” and you then set up a tree flow to send them the right product tour that’s actually relevant. So with what was relatively a small amount of effort, we’re able to really open the door to conversational onboarding.

Custom Bots trigger specific tours depending on the context

Another thing: you can send Product Tours directly from the Intercom Inbox. You can add them to the Messenger home screen.

Showcase relevant tours right in the Messenger

And then all the new stuff we’re going to build in the future, a lot of that’s going to benefit Product Tours as well, sometimes literally for free or sometimes with minimal additional investment. So all this stuff you can build on top of the core use case – we think of this as compound interest, and it’s a critical differentiator. You can’t actually build this idea of compound interest at the outset – this idea that with each new product feature you build, you add incremental value across the entire platform beyond the feature’s core use case. This compound interest comes from years of building the product incrementally and thinking and building in systems, and then you start to reach a point where you can start to stitch things together in your product really fast. It’s incredibly exciting from a product development prospective, and we think it’s really exciting for our customers as well. This is an example of a product differentiator that’s way harder to copy.

“This compound interest comes from years of building the product incrementally and thinking and building in systems”

Phil: Yeah. Totally. It’s something that I’ve got experience with since the launch of Product Tours, and it’s something I don’t think really hit home to me until I started using it.

We made some Product Tours to onboard people to our individual core products and features. Then we found our support team could leverage that as well, and then you’re saving hours of their time. Our sales team are using them to walk people through the product and demo them. It requires very little outlay from us to create Product Tours, and all of a sudden there are three or four different teams that can use it as well.

That’s the compound interest you’re talking about, where you’re maximizing a new feature’s value with minimal effort because the feature is part of this system or platform. It’s a personally resonant example of that compound interest and that maximization of the platform you’ve got at your disposal.

The time and place for video

Phil: So, we’ve talked about differentiators. We’ve talked about the ease of use. We’ve talked about the way it’s stitched into this all-encompassing platform. But something else which really hit home for a lot of people was the inclusion of video. How did that fit into the the overall story? Was it part of the plan to differentiate Product Tours with something like video? Because obviously I’m involved with video; we use it in Intercom’s onboarding experience, we use it in our webinars, we use it in demos. Tell me all about that.

Patrick: The truth is, it wasn’t in the pitch at all until very, very late in the game. I think it was mid to late February when we were doing one of our end-to-end product reviews ahead of launching Product Tours into beta. The general sense in the room was: “This is a solid product. But is it boring? Have we built a boring product?” Honestly, I think even Brian stood up energetically in classic Brian style and did a huge spiel about how we’re missing a spark or a flare.

“First-to-market rarely guarantees best-in-market, but as humans, our first question is usually, ‘What’s actually new here?'”

Brian: What’s important is that sometimes a boring product is exactly what you need. Sometimes a boring product is actually the best product, because it’s not going to be boring to your customers who actually are feeling that pain. But when you’re doing a big release – when you’re putting a big push from your go-to-market team — you really want to ensure you’re giving it its best chance to make a splash, because when you release it, you’ve got that brief period of attention, and you want to maximize that. We all know that first-to-market rarely guarantees best-in-market, but as humans, our first question is usually, “What’s actually new here?”

Phil: This is also interesting because our customers who are going to use this need something that’s attention-grabbing because it’s their product onboarding. They need to spark that kind of interest with their customers to get them to see value quickly in the product and give them momentum to get to that “aha” moment. So it works on a lot of levels. Patrick, Brian has now just crushed you by telling you you’ve made a boring product. What happened?

Patrick: A couple of us in the team huddled together after that meeting and literally started whiteboarding a lot of ideas that could have this “Brian spark” or give it that extra dimension. And one of the ideas that was thrown on the board was video. At the end of last year, we launched a bot where you could send video messages. So the question we were asking ourselves was: “Could we do that for Product Tours? Would that be cool? Would that work?” Really, that was the level of resolution we had at this stage.

It was interesting enough to help us poach an engineer from the original team that built video messages. We basically said: “You’ve got a week. Scope this out. Let’s get a sense of what this feels like.” We pulled together this video tour, and we all sat down, and it was one of those moments I’ll always remember at Intercom. We were like, “This could be a game changer.” We had been on a campaign for the last few years to find a way to use video at Intercom, and we’d had success with that but we hadn’t found that real killer use case. Time will still tell, but we really feel like this could be really special. The really amazing thing was because this had been built already, it was actually pretty cheap to pull this over and use it in tours. It was super quick to build, and that’s how it got in.

So it actually ends up being another example of compound product interest, right? It was not a big bet, and it was really compelling. So all of us were really confident in making this bet, because it was actually pretty cheap and because we could build on top of what another team had already built. That was really exciting.

And the feedback has been awesome from customers as well regarding the video. It’s super engaging. They’re loving it, and it speaks to their product as well. It’s something that doubles down Intercom’s commitment to make internet business personal.

Brian: Yeah, that was actually another reason why it such a no-brainer. Once we saw that it was compelling, it checked the box for so many angles that we were like, “Oh yeah, for sure.”

The constraints of dates

Phil: I’ve just got one more question before we wrap it up. We talked about this product being kicked off last July. It got delivered in April and we went to market. Like all of our other big launches, we set a date for pretty far out for this release and we stuck to it. I know there are a lot of arguments either way, but dates are usually the contentious thing to work toward for product teams, right? Some people feel that they’re a bit too constraining. You should be able to produce the best thing ever and release it whenever it’s ready. But Patrick, I’ve heard you actually love working to dates. Is this actually true?

Patrick: Yes, absolutely. I think just one clarifying point is when we do work to dates at Intercom, we don’t blindly put a stake in and say, “Now it’s March 21st; we have to build the product in that time.” We make an informed guess based on the rough idea of what we’re trying to build and come to consensus that way. But yes, I do think dates are fantastic because they are probably the most powerful forcing function for making progress and decisions quickly.

Ultimately, if you can’t change time or the resources of your team by magically summoning new engineers and designers or the bar on quality, then the only real dimension for the trade-off debate is scope. What features are in, and what features are out? Having a time constraint really forces you to make super hard scoping calls and really hone in on building the most important features. It’s the constant debate that everyone on the team is having every day. Should this be in scope? Is it really, really needed? Is it critical for that launch? And it really, really works.

“Having a time constraint really forces you to make super hard scoping calls and really hone in on building the most important features”

Phil: Okay that begs the question, what’s a good example of what you had to de-scope?

Patrick: With Product Tours, a really good example would be reporting. I don’t know if you remember, Brian – we had this big scoping debate on whether we should launch with some reporting functionality, and the conclusion we came to was no. The question we asked was, should we focus on reporting over some of the core building or workflow features that would enable you to build the type of onboarding you wanted to build? In our head, it was more important to focus on helping customers build the tour that they wanted to build because if they couldn’t build the tour they wanted, then reporting has no place. They didn’t care about reporting on something they can’t build, right? That really helped us, and I would say that was a good example of scoping.

Brian: We knew we needed reporting, but the question was whether we needed it for launch or if we could have minimal reporting until then and be committed to building it after. That was unexpected. I think you and I were like, “We have to have it in, for sure.” It was actually go-to-market pushing back, and we had this really healthy scoping debate. Man! I did not expect to come out with that result, but it was a great example of where we had a trade-off: “Here’s what we can do at this time. What’s more valuable to impact the launch?” And reporting dropped out.

Phil: So hard dates make difficult decisions easier to make.

Patrick: The other thing is that this doesn’t have to translate into a stressful experience for the team. I mean, it can and often does, but I think one of the things I’m proud of is that we’ve matured to the point where we can build an ambitious product, deliver it pretty damn fast to a high bar of quality, and do this without people actually having to pull their hair out or work crazy hours. Once teams understand and everyone has bought into making small trade-offs, you can have a pretty intense but incredibly motivating experience that is not a crazy, hectic one. For me, it was particularly gratifying to feel like we’ve reached that point.

Phil: There it is: your window into how we create and launch a product in Intercom. And if you’re interested in working with product teams like this, we’re hiring in our Dublin, London and San Francisco offices, so get your applications in at intercom.com/careers. Thanks for listening, folks.