BTP: Building a New Product Line with Gui Schmitt
Behind the Pixels is a conversational blog series that aims to give a glimpse into the work & life of being a designer at Klaviyo
AH: Today I’m talking with one of our longest-tenure designers who’s done quite a few projects in his time here and was actually an early Behind the Pixels guest. Why don’t you start by introducing yourself again? How long have you been in UX? And how long have you been at Klaviyo?
GS: I am Gui Schmitt. Been a designer here at Klaviyo for almost five years at this point, which is crazy. Throughout my journey here I’ve worked on several projects that include the SMS builder, the email editor redesign project and now more recently I’ve been working on the reviews project which is a major new initiative that is going to be released soon or, maybe by the time you’re reading, this already out, which is exciting. Yay, future me!
AH: [Laughs] Yay future Gui! Amazing, and reviews was a different type of team than we previously had at Klaviyo. Can you give us some brief context around the team and how you all work?
GS: I believe reviews is the first product that we’re releasing that comes out of this initiative called Klaviyo Ventures. In it, teams are free to experiment with new ideas, validate them for product market fit, and then take them to scale and bring it over to our main product or do an independent product. In this case, reviews was released on a limited scale for Shopify last year, similar to a soft launch, where we validated that market fit. This was before I joined the team and once we identified that people were resonating with this idea, we decided “hey, let’s take this and bring it over to the main platform and build all the features in order to enter this new area which has established players”. Obviously for people to switch, they need to have a good set of value unlocks and we want to ensure that we’re meeting their expectations entering this brand-new market.
AH: Yes, that’s interesting. You’re definitely hitting a new market and there are some pros and cons there. You know what’s expected, but also you’re trying to make something that’s competitive. Which kind of leads into my next question — you said there was some research early on to validate the opportunity. Did you do more research later? Was there an opportunity to talk to customers about concepts?
GS: Yeah, this project is fairly different from others in that we already had a product that was very basic. So we did talk a lot with users of that existing product to see what they liked, what they didn’t like, and what they wished to see in a more full solution. And because we were entering this new market, and we had those competitors that were trying to match. We talked to customers of those competitors to see what they like about their current platform, what they don’t like, and what would make them switch. So that gave us an idea of what was the perfect quadrant of things that we should focus our attention on. Ideally, you want the easy-to-build, high-value unlock — not everything is perfect like that, but we were trying to find things in that direction. And I think we’re going to learn a lot as we launch this, and our idea is to have a much quicker cycle of improvement and iteration upon that first release to respond to how people are reacting which is very different and exciting.
AH: That’s a really fun way to build and cool that your team was able to pick out what could be a competitive advantage early on. Anything interesting that you did during the creative process to come up with solutions, maybe a particular brainstorm or something?
GS: Well, I think what was a particular challenge around this project was how it operated within our design system, and the components in it. For those who are not familiar, components are a series of Lego blocks that we can assemble to make any sort of interface that feels, behaves and looks like what people expect from the platform. So I think whenever you’re trying to solve a brand new problem, you are faced with a decision of what can we do with your existing Lego blocks? And should we build any new Lego blocks to better suit this problem space?
For reviews, there was a lot of back and forth around like, ‘Hey, what should we do here?’ Because you can always do something with the existing Lego blocks, but it’s just a measure of compromise where you have to consider things like — how often are people doing this? Is this a critical path in the overall journey that requires an extra level of investment? I mean, if designers had their way, everything would be custom, right? [Laughs] But obviously, the reality of a business is that you have to make decisions and considerations and choose the best solution for this given moment in time. And a lot of times like you might get a feel of something, but again, going back to that original point, you always want to check and validate that feeling, and putting it out there sometimes is the best way to validate something that you might not be sure about, to see if the investment is worth it. Every project goes through some of that in terms of what is the best solution, but I think when you’re entering a new area like this, it tends to happen a bit more often and so I think that was a considerable, unique characteristic of reviews.
AH: Makes sense — it’s all about collaboration and give and take to make the right decision. So, here at Klaviyo one of our design values is trustworthy. I’m curious, how did you all work that into reviews?
GS: So to me reviews are all about trust. Reviews are taking the customer’s voice and opinion and using that to add trust to your brand and add social proof and validation that this is a product that real people like me and you are using and are saying its great. Throughout my research into this area, I found certain interesting insights about, you know, human psychology when it comes to trust. So, for example, a lot of people might think that having a perfect, 5 stars out of 5 product is the most trustworthy, but research shows that people actually trust products that are around 4.8 or 4.7 because everyone understands that not everything can be perfect. So when something looks too good to be true, people are skeptical, especially when you think about how now with technologies like AI and generative creation of content it’s easy for someone to whip up text. For us, it was key to try to find ways to take the data that we have and that we’re collecting from reviews and make it very clear that this is real and try to use different signifiers to do that, like promoting people to take pictures, so that it’s more realistic. We even consider things like the presentation of a review — if you go to a product’s page, everything is studio-level quality and that’s great, but people also want to see the real lighting, the flashlight on the cell phone look, and things like that.
AH: Totally. That’s super interesting, the psychology piece and making it more human, even though it’s technology. We want that to come through in the design. Did you run into any technical limitations as you started to build?
GS: It’s the piece around the design system, but ultimately, I think we are shipping with the right set of compromises where we feel like we got to do some new stuff as well. It’s not as if we compromise on everything. But in terms of reaching an MVP, you gotta sometimes cut some features — it’s not that we won’t do those but about what do you do first? Not everything is technically feasible for day one.
On the design side, I had to make several different states of the product — this is the initial release. This is subsequent — slicing the long-term goal into little pieces. I think that helps to drive the conversation because when a stakeholder is seeing a screen that has a lot of stuff in it, depending on the stakeholder, that might be either exciting or anxiety-inducing. [Laughs] Like if you’re talking to an engineer and they see all those features they might start asking ‘Oh, what is this? And this?’ and you’d have to stop and explain ‘No, no, no, don’t worry about that for now.’ So making sure that we’re always focusing on what’s the right thing to show to the right set of stakeholders. That wasn’t quite the question, but that was just something that came to my mind.
AH: And incrementality leaves room for more exciting things to come which is always good. All right, what’s the biggest challenge you had to overcome while designing and building reviews?
GS: So one interesting technical challenge that we had is we looked into how people are using review content in their emails before we released this product and the reality is that people just copy and paste quotes. The thing about that is you can copy and paste anything and it’s also laborious to do that because you have to do all the work by yourself. So we’re trying to find ways to take all the data from things that people have said and use some sort of AI technology to help us parse through all that data and find the strongest sentiment from each review and help bring that over to their emails so that they can easily take the strongest phrases that people are saying and put that into their messages. So we built this AI tech that analyzes each review and finds the short quote or phrase that you can use in your message. That’s an interesting problem because we wanted to do something that was better than the existing copy and paste and we also wanted to do something that would save time. In the future, we want to allow people to also choose other parts that are not just the AI selection because it doesn’t get right every time. But as it gets better, it should become more correct each iteration cycle.
AH: That’s really cool. Always fun when we can toss in a little AI magic in there and then connecting back to the Klaviyo app makes the whole thing feel like a really good package. On the flip side, what are you the most proud of thinking back over the process of building reviews?
GS: I think we’re fairly confident at this time, but again we’ll see [laughs], that what we are putting out there in terms of helping people get the most out of a reviews program meets their needs and can help them save time. The AI quote is an example of a saving time feature, but another thing that we did was with moderation.
Moderation plays a huge part in the time that you spend with reviews. There’s a bit of setup, but once you’re set up, most of your time with reviews is spent moderating — making sure that the things that you’re putting on your site are okay. And we built reviews so that the task of moderation has a lot of different ways that you can save time. For example, all the things that you have pending are presented on the first page, so you don’t have to go into multiple pages to see any questions or reviews that are pending. And this is something that came directly from feedback that we received from people using the original app.
So I think I want to be proud, but it’s hard because it’s not out yet. But I think there’s a lot of potential for some of these ideas. We have more features that we want to do in the future that are going to be aligned with that same ethos of saving time — like doing sentiment analysis, for example, or using AI technology to get the sentiment of [reviews] and then also potentially flag ones if someone puts a curse word in your review, for example. People get mad sometimes on the internet, as we know, so having a filter for that could be helpful. We’re building an inbox of tasks that people have to get through, so being able to sift through those tasks pretty quickly and efficiently is our goal.
AH: I like that — something we hear from customers all the time is that they don’t have enough time. So very smart. This next question is kind of similar, but do you have a favorite memory working with a team on reviews?
GS: We got to meet in person a few times. We’ve got to go to the office, which is great — it’s a different dynamic. I’m glad that we were able to do that early on in the project as we were kicking things off. That’s the moment that you’re trying to get a sense of who you’re working with, how they’re like, and what their way of working is and I think you get a sense of that quickly when you’re in person. And of course, I’m looking forward to the release of it so we can also celebrate. So I guess a favorite future memory!
AH: Favorite future memory, lock it in! Cool, and you’ve been alluding to this throughout but what happens next for you and reviews?
GS: I’m looking forward to continuing working on those features and iterating and learning from the initial release so we can continue to deliver value to this new initiative.
AH: So you’re looking forward to getting reviews on your new product, reviews. Sounds good.
GS: Yeah! [Laughs] We’re always open to feedback!
AH: And last question, do you have any general advice for designers who are working on a net new product line like this?
GS: When you’re entering a new area or vertical, it’s a little overwhelming at times because you’re going against people that have been doing this for years. And I think you just gotta do your best to absorb that and then understand that, ‘All right [our first version] is not going to have all these things that these competitors have’, so how can you find ways to set yourself apart? In our case, a lot of the strength comes from consolidation — being all in one platform and trying to find ways to connect with other areas and unlock value that way. If you’re starting something new at a new company, you might not have that, but it’s finding the little things. Little inefficiencies in these other competitors that you can potentially innovate on and do differently. There might be some gravitation to do things in a certain way, but maybe that’s also an opportunity to do it in a different way and try to stand out through that. I think reviews is an interesting area because there are so many surfaces. For example, you have people filling in the review as a customer that bought something — we didn’t even talk about that on this interview! — but you also have the people they are moderating and using that review to promote their brand. So there are all these different areas where you can find ways to stand out and do things in a different way.
⭐️✍️— This is the first big release from our ventures team — props to Gui & team on a job well done. Can’t wait to see whats next with reviews and the ventures team as a whole!
Does taking on the big challenges sound up your alley? We’re always looking for great people to join our team.
