Enterprise vs Startups: Established Products vs Greenfields Products

Audrey Tsang, Benoit Terpereau, Jenn Dearth, Parul Goel at UXDX Europe. Video: https://www.youtube.com/watch?v=tNNKSt86PXg

Readable transcript: edited from the recording's captions for readability (fillers and false starts removed, punctuation and section headings added). Wording is the speaker's own. Timestamps are positions in the video. Names marked [?] could not be verified against the audio.

Introductions: startup or enterprise

[00:00:00] Rory: We've all seen your talks. I hope everybody has seen your talks. But what I want to do is a very quick introduction, and in the context of your introduction, can you explain whether you're coming more from a startup or greenfield environment, or more from an established enterprise environment? I'll start with yourself. Audrey, please.

[00:00:19] Audrey: I'm Audrey Tsang. I'm the chief product officer at Clue. We've been around for 10 years. We've got about 12 million monthly active users. We're a B2C female health app based in Berlin. So it's not exactly greenfield. We've definitely got users that love our features or hate them. But it's not enterprise.

[00:00:43] Rory: Great. Jenn.

[00:00:43] Jenn: Yup, great. Hi, I'm Jenn Dearth. I work in an enterprise. Workday has over 12,000 employees, just reported a billion-dollar quarter, and has about 42 million active users. And within that, I work within the productivity suite, which is kind of an entrepreneurship within the larger Workday. So I'm in an enterprise, but I would say we have more of a startup environment inside of the enterprise.

[00:01:13] Rory: Great. And over to yourself, Parul.

[00:01:13] Parul: Hello everyone. My name is Parul Goel and I head product for channel partners at PayPal. So, similar to Jenn, also an enterprise, a large company. Not only that, I also build products for enterprise. But I like to think that it is a mini startup where we build up products from scratch and take them to market.

[00:01:37] Rory: Great. And finally, Benoit.

[00:01:37] Benoit: Hey guys, I'm Benoit Terpereau. I'm VP Product at Deezer. Deezer is a worldwide music streaming service, and we have something like 14 million monthly active users. So it's not a greenfield, it's more, I (inaudible) ICT and an IPS[?] product.

Can you run an MVP on an established product?

[00:01:55] Rory: Great. Excellent. Thank you for introducing yourselves. I'm going to jump right in with an established product. We always hear about the MVP, build, measure, learn approach. But can you do that kind of an MVP approach when you already have an established product with established users who have expectations for how things are going to work? I'll start with yourself, Benoit.

[00:02:23] Benoit: Yeah. I do think that you can do MVP, even if you have an established product and users. What strikes me the most with MVP is that, at the end of the day, a lot of product managers don't really understand what an MVP is, and we have all our assumptions. We want to put in more features because we love our product. I always say to them, okay, you want to add this feature, but are you sure that users want it? Why not just not deliver it, and wait for people to ask for this feature? If they don't, then the feature was useless. If they do, then you will be able to deliver it later and get some user satisfaction.

[00:03:02] Rory: Great. Okay, thank you. So yes, you can use that kind of approach. Parul, you just mentioned there that you're an enterprise delivering to enterprises. Do you have enough volume of customers to be asking the questions?

[00:03:22] Parul: Yeah, I agree with Benoit. Usually, if you have an established product, it's customer problems that drive innovation and drive change. In the enterprise space, we have more than 200 customers who have integrated with our product. Compared to a B2C product, that's very small, but for a B2B product it's substantial. So in this case, there are two things that matter. One is how many of these customers are asking for the capability, and number two is who those customers are. Because, let's be honest, if it's a large strategic partner, them asking for something carries more weight. So those are the two things we take into account.

[00:04:05] But something else that's different is that if we are building something without them asking for it, ideally we would build it in such a way that it wouldn't require them to be (inaudible). They wouldn't have to do anything to consume this new product. That's really the challenge in the B2B environment. How do you innovate without impacting the existing integration, without breaking the existing integration?

[00:04:32] Rory: And how often are you successful? Because that is quite a difficult challenge. Often with companies and APIs, you do have breaking changes because of what people are asking for. So how successful are you in the goal of not doing breaking changes?

[00:04:49] Parul: It is tricky. I think it depends quite a bit on the nature of the change. With an API, you can add stuff without breaking previous integrations. But changing existing capabilities, that's hard. What that means is we do a lot of validation, a lot of vetting upfront before we roll out an API. Because once it's out there, to some extent it's installed. A few years ago, we upgraded our API from V1 to V2, and to this day we have customers still on V1. Because it's just hard to get customers to move up, to upgrade to a new capability. And it makes sense. They have different priorities. So really my advice would be, if you have an API-based product, just make sure that that's the right product you want to roll out, because making changes is very tricky.

Getting feedback in B2C and B2B

[00:05:39] Rory: Great. I'm just going to touch on what you said there around the upfront validation before you start making those changes. I might go to yourself, Audrey. How do you go about it? Do you have any challenges? In the B2C world, is it different, or is it easier or more challenging than the B2B world, to get that feedback and validation before you start building something?

[00:06:02] Audrey: Whether it's easier or harder, it's probably just slightly different. Our feedback comes from our users. Usually it's in the form of either they write in through support, or app store reviews. We also do a lot of user research. For instance, we have a new product that we actually just shipped today. The team would definitely call it a minimum viable product or a minimum lovable product, around pregnancy. What we've done is gotten a lot of feedback leading up to this, from users and interviews, et cetera, before actually rolling it out as an experiment. So I'd say that feedback comes in lots of forms. It can be qualitative or quantitative through surveys, or we look at literally how a particular feature performs, or we hear about it in support messages or reviews.

[00:07:00] Rory: And do you find it challenging getting that feedback? Do you get enough of it, or are your customers willing to share, or is it quite difficult? I guess health could be a touchy subject.

[00:07:14] Audrey: Yeah. It's a very touchy subject. One of the things that we at Clue care a lot about is respecting the privacy of our users. I understand that this is some of your most personal data and personal content that we have. Surprisingly, our users are actually willing to share. They're willing to talk. Sometimes we even get a lot of health questions through support. We by no means give health advice, but it's our bodies. We do want to talk about it.

[00:07:48] Rory: Great. And Jenn, I guess it's the opposite. It's an enterprise environment. Or am I wrong? Is it something that people want to talk to you about? Or is it just something that they want to forget about, and it's just in the background?

[00:08:07] Jenn: For us, for all of our new products, we run early adopter and fast follower programs. We do a selection across Workday's 3,000 global customers, but we ask folks to apply. Then within that, we look for a diversity of high volume, high transactions, geography, industry, size, just to make sure that we're getting a really good mix whenever we run these programs. Because obviously everything on Workday, if you think about it, is really mission-critical transactions. We don't want payroll to fall over if we're trying to build a new product that sends out emails. So there's definitely a pretty rigorous fast follower, early adopter program that leads before we become generally available.

[00:08:54] And whenever we select them, we also make it really very clear up front, to say we're expecting you all to run through these use cases, and we're expecting you to spend at least four hours a week or a month. I was listening to Jeff Patton's conversation right before this, and we do very, very similar practices, where we always have engineers in the meetings, at least one in every feedback session. We release, and then we ask them to hop on a call and do feedback with us.

[00:09:24] Rory: And with that program, sorry, that sounds really interesting. Do you have to incentivize people? Are they jumping to join? How do you get people in?

[00:09:34] Jenn: I think we're super lucky, because we're talking to folks who understand that what we're trying to do is make their lives better. They want to have input and they want to have a say in how the product gets developed. So the incentive is: are you going to make my life better? Are you going to save me time? Then yes, I want to give you feedback. I want to have a voice at the table. So we usually have the opposite problem, where it's hard to say no. To say, hey, we can't actually handle more than X number of customers, but we'll include you next time.

[00:10:08] Rory: Great. That's a good problem to have. Benoit, your background: you're currently with Deezer, but you have a lot of B2B experience as well. Have you ever struggled with trying to get that feedback from customers?

[00:10:23] Benoit: Not really. To me, the main (inaudible) is when you work on B2B, and especially in enterprise software, which is what I did, you can have 10 customers on the phone and you have your roadmap, because you don't have that many customers and they are really vocal and they know what they want. So you can do that. It's really straightforward product management. What is more different with Deezer is, it's a bit like Audrey: you don't have trouble getting user feedback, but you can be overwhelmed by it. I say to a lot of my product managers, if you're not talking to one user per day, you're not doing your job well. But the issue with that is sometimes, once the user (inaudible) for a product manager, it can be really crazy, because he wants crazy features and stuff like that.

[00:11:16] To me, in B2B you work with your customers and you have your roadmap done, and then you have a slow pace of execution. Usually when you do enterprise software, you have a one-year roadmap. It's different when you're doing B2C. The main difference is, if one person or even 10 give you some feedback, you cannot make it a reality, because probably it's only those 10 people who want something, out of 14 million. So this is where user research helps. You have insights when you talk with users as a product manager, and then you pair with user research to try to get the truth. And the truth is probably something more statistical. Do we have a majority of users that will like this feature or not? It really helps to find the truth wherever it's possible.

Avoiding going in a thousand directions at scale

[00:12:12] Rory: That's a great point. As a startup, you're almost just thinking, my first customer, I'm going to really hone in on what they're asking for. My first five, 10 customers, and really build a solution for them. But now when you're at scale, how do you avoid going in a hundred or a thousand different directions? Because you have so many different users with so many different needs. (inaudible)

[00:12:40] Benoit: It's really important, what you said, because I think, a bit like Audrey, we have a lot of community and we have (inaudible) on the app stores and stuff like that. But you have to be careful to not listen only to vocal users. We have really power users that are very demanding in our communities, for example on Reddit and on Twitter. And you cannot always say yes, because those guys are power users and it won't make the difference. It's maybe one or 2% of your user base, and you cannot do that. So I like to work with tools like prioritization metrics based on impact, making sure that when we want to deliver something, it's going to fit the best mental model for users, or at least it will have an impact on our numbers: new users, or intensity of usage, or something like that. But we cannot just say yes to the power users. Sometimes we do it because they are vocal, and it can be good publicity for us, but we are really careful with that.

[00:13:39] Rory: Great. I'm going to ask Parul the same thing there, because I guess it's different ends of the scale. There's a hundred million plus, 140 million, I can't remember the exact number, sorry, versus the 200 customers. So are you still very much listening to individual customers? It's still a challenge. You don't want to divide your product in lots of different ways. So how do you manage that?

[00:14:05] Parul: Yeah. Rory, that's definitely a challenge, especially when your customers are large platforms like Facebook and Etsy and the likes of them. They usually do have custom needs. What I have found through this journey is that you have to have an ability to evolve over time, but you have to build an identity for your product. Who do you want to be? Do you want to be somebody who provides custom solutions, white-label[?] solutions? Because then your product is going to look very different from a product that has self-service onboarding, where you have this set of capabilities and that's what you do.

[00:14:48] I think this is something that we had to work on over time. We realized that, although we said we would do a self-service platform, a generic platform that anybody could use, over time the reality of the segment we serve is that it's not going to be completely automated. We are always going to have large strategic partners that require something custom. And for the sake of business enablement, we have to be open to that. We have to have some capacity to serve them, for opening those doors. But at the same time, we also need volume. So we do have two tracks: one is custom, and one is self-service, where our smaller customers are able to utilize what we have by just taking care of themselves. This is something that we have learned to do over time.

[00:15:37] Rory: And how do you balance it? Is it just experience and judgment, or what do you use?

[00:15:40] Parul: This is where what Benoit was saying comes in: making it a very objective, data-driven decision. For example, the impact, obviously the revenue impact that we will get from it, the severity, the long-term relationship with the partner. A lot of things go into deciding which bucket a partner or a customer would fall in.

Listening to the customers you are losing

[00:16:05] Rory: Great. Audrey, you were saying you've got a large audience base. You've got a lot of vocal support. Is there any risk, or do you look at who's not your customer, or which customers you're losing? Because there could be a risk that if you're just listening to your customers, you're limiting your marketplace.

[00:16:31] Audrey: Well, I think this is what both Benoit and Parul talked about. It's what goal are we focused on? Right now, we are focused on subscription revenue. So when we look at the prioritization of potential features, do we do this or this? Obviously we all have limited resources. What we specifically looked at was what unsubscribers were saying, because we wanted to specifically look at their needs. And the reason that we launched pregnancy today was because that was one of our most common reasons for why people were unsubscribing from our premium features. They were potentially using Clue to conceive, and then they got pregnant, and then it was like, well, now what? So this was really viewed as a retention feature.

[00:17:21] There were a couple of other things that helped us determine to focus on that. It wasn't just user impact. It was also, hey, what's the proper sequencing? Were there dependencies on other features, or other technical dependencies? Or do we have something like, let's say, (inaudible) that might prevent us from launching this as quickly? So we were looking at a basket of things, but fundamentally focused on how we best grow revenue by delivering things that subscribers actually want.

Managing customer expectations while experimenting

[00:17:51] Rory: That's a great example, actually, the offboarding stuff: figuring out why people are offboarding and solving for that. And I'm going to touch on the customers' expectations, because I mentioned it at the start. Jenn, I'll give this one to you. With a product like Workday, people are expecting stability. They don't want to log in every day and see that things are different and changed, and that you're running another experiment and it will be rolled back tomorrow. So how do you manage those kinds of customer expectations and still enable you to do the experimentation that you need?

[00:18:29] Jenn: That's a great one, Rory. When I first started at Workday, I was pretty surprised, because Workday releases biannually. And again, to your point, there are over 40 product areas. People don't want to be surprised. They actually asked Workday to start releasing less frequently. However, within our organization there's also a productivity suite, and our tool is still released every week. We're able to do that because our products are very, very close to our user base. We're able to segment. The idea behind it is that people are opting in to think through and use our productivity tools. They already have that in their mental model, because of how things have been marketed and designed and positioned. They understand that these are tools you use as an augmentation, something that gives you a superpower when it comes to using normal Workday. These are productivity tools that sit on top of the rest of Workday.

[00:19:38] Rory: Okay. Great. Does anybody else have a way that they manage that conflict that they want to share? Any experience with the fact that your customers might not be so willing, in certain environments, to do those experiments and see things change? I'm sorry, go ahead.

[00:19:59] Benoit: Just in B2C, it's something that is business as usual[?]. Two things. First of all, you try to do experiments mostly on new users, so they don't really know what's going on with your product. And sometimes you do it with your regular users, but in this case you involve your power users, and you make it so it's kind of a reward that they are testing a new feature. It's kind of a reward because they are really faithful to us, and you can do it that way. It's a bit hard to manage if the experiment doesn't work and you have to decommission your experiment. But it's part of the game. And I think for power users of digital products, especially in a subscription model, people are kind of used to it.

Overcoming bureaucracy in the enterprise

[00:20:46] Rory: Great. I'm going to come back to the core premise here: enterprise versus startups. In startups, they move fast, break things, just get it working. The runway is short. You need to get there before you run out of funding. In enterprise, it's the stereotypical bureaucracy. So how do you try to overcome those challenges with bureaucracy, so that you can keep the pace of change and innovation strong? I'll start with Benoit.

[00:21:24] Benoit: By nature, these are digital products, so we are going pretty fast. We have bureaucracy, but it's not our issue. I did a lot of enterprise software, though, and that was more complex, especially because you have a lot of contractual obligations. Your customers don't want to see features go, and you have to maintain a lot of versions of your product. It's much more complicated.

[00:21:59] One tip, which I did in the past, is to put customers around the table and make sure that they talk with each other. When they all want something, you try to find the minimum that can make everyone happy around the table, so they can have it faster. You start a meeting with various customers and you say, I want to develop this feature, what do you want? And they will all say, I want feature A, feature B, feature C. And you say, if you want all of that, you will have everything in one or two years. So do you want to sit down and find the MVP, or minimum lovable product, for this feature? Then let's start talking. And if you can promise something in three to six months, then they'll agree. So it's (inaudible) this kind of thing: pairing with your customers, a single voice of truth about delivery and our complexities.

[00:22:55] Rory: That's a great example. I hadn't actually thought about it from the customer side of things, but it makes a lot of sense. Is there anybody, maybe Parul or Jenn, in the bigger (inaudible)? Jenn, from an organizational perspective, is there bureaucracy within the organization that you need to jump through?

[00:23:13] Jenn: Yeah. I actually watched Hendrik's[?] talk yesterday, and when he said one plus one plus one equals ten, I thought it really is true. Sometimes we think of large enterprises as slow moving. Bureaucracy almost has a negative connotation to it. But every feature has a positive and a negative. Even though bureaucracy sounds like a negative, the positive is you have leveraged impact. What you can take from the startup world and apply into enterprise is that you're still making those connections. Instead of a startup going out and saying, who do I want to partner with, what market should I enter into, in an enterprise you are looking internally. You're saying, which app teams should I be partnering with? You still have to think of it. Systems thinking still applies. You're still thinking through which connections you should be making, but you're networking internally as opposed to externally. So I would say, just keep in mind that being an enterprise has its positives as well.

[00:24:18] Parul: And if I can add to what Jenn is saying: we are a payments company. So if we roll out a change, it is super important that it works and that it's compliant. This is where bureaucracy actually helps us. It helps us make sure that we are on the right path. And there are ways to cut out the unnecessary bureaucracy: by empowering the teams, by making it about the customer, by making sure the leadership is aligned before you go into more detailed discussions. Those are skills we have honed over time. But I also do agree that, being in payments, I feel there are a lot of upsides of having these large workflows and forces behind you to support you.

[00:24:57] Audrey: I actually love this reframing of bureaucracy. It's not something that's set up. And I miss being in a startup; it's fun. We get to move fast and break things. I would say that there's definitely a downside to moving too fast and breaking things, because it can feel like whiplash. With an MVP, you may not invest long enough to actually see, is this really going to work or not? Because it was so minimal that you might just ship it and then go on to something else. And bureaucracy, or at least leveraged resources, or more feedback, or more focus and commitment to something, can actually help really refine a product through its process.

[00:25:43] Rory: Great. Audrey, you said at the start that you're past the startup phase, but you're not an enterprise. You're in the middle. What is the culture like in Clue? How would you describe it?

[00:26:00] Audrey: I think we're definitely a startup. We're just not struggling for product-market fit. So, depending on your definition, we're definitely still a startup. We're about 70 people. If I were to describe the culture, the thing that stands out is that people are just truly passionate about our purpose, which is to make female health more accessible and clearer, supporting women and people with cycles to make the decisions in health that they need to make. Everybody is very focused on that mission and that purpose, and that really makes the culture unique. But it also means that there are lots of passionate ideas and passionate thoughts, and wanting to go in a lot of different directions. So bringing that all together is definitely one of the challenges, but also one of the joys, I'd say.

Closing messages of positivity for 2020

[00:27:00] Rory: Great. I think we're just about at time. But I'm going to ask everybody here, since this is the last few minutes of UXDX 2020. 2020 has been a weird year, not going to lie. It's a bit strange. We're not all sitting together, as I would have hoped that we could have. But do you have a message that you just want to leave, of positivity and inspiration, to finish off the day? I'll just go across my screen. Audrey, you're the first one.

[00:27:25] Audrey: Lean into those challenges. Lean into the weirdness of 2020. We had to do a lot of things differently, and we've learned a lot from it. So the silver lining is it forced us to do stuff differently, and that was probably a good thing.

[00:27:43] Rory: Great. Jenn?

[00:27:48] Jenn: I'd say I have three thoughts around what to be optimistic about. The first is really around innovation. I feel like there have been a lot of really interesting things popping up. I don't know if you all have heard of Lunchclub AI, but it's this idea of using machine learning to pair people to have a coffee. As a result of these constraints, I feel like I've seen a lot of innovation. The second thing is just how companies are responding. I think there's been an outpouring. I loved the first panel that kicked off UXDX, where you saw these leaders of large corporations like LinkedIn and Shopify saying how they're supporting their employees and their people. It's really powerful to see that corporations really have the stewardship, and they're thinking about social responsibility differently.

[00:28:34] And the third one is just the shift in how we work. I have two young kids at home, and we're able to cram things in. People are opening their eyes to how work can still be done and be very productive, even if it's a little bit different than traditional ways. So the shift is getting us to think more creatively, more broadly about what can be possible. Those are my three points.

[00:29:00] Rory: Excellent. Parul?

[00:29:00] Parul: I have two, kind of similar to Jenn. One is innovation, because it's usually constraints that force us to think differently, and I'm specifically looking for innovation in travel. Because that's an industry that's been severely impacted by the pandemic, and I'm sure there's going to be a big comeback and it's going to look very different. So that's something I'm looking forward to. And the other thing that the pandemic showed us is that we can work from anywhere. We don't really have to be in the office. As a product manager, I always assumed I have to be in the office, and now I see, oh, I could live in Bali and still be a product manager. So that's what I'm looking forward to: living in Bali and still being a product manager.

[00:29:42] Rory: Excellent. I'd love to join you there. We're currently getting winter. Winter is definitely approaching.

[00:29:53] Benoit: Let's do UXDX in Bali next year.

[00:29:53] Rory: As you say, we are doing UXDX APAC in March. But Benoit, what message do you have?

[00:30:00] Benoit: I think it's going to sound pretty much the same. I think we all need to embrace the challenges that we face, in at least two areas from my perspective. From a business or startup standpoint, there are so many new opportunities that open up. Digital is booming. Online retail is booming. You can now sell pretty much everything and provide any kind of service. Companies need new online services every day to boost collaboration. So in terms of business in the digital world, I don't know if goldmine is the right word, but there are still good opportunities.

[00:30:46] And from a personal standpoint, you can imagine new ways to work, to be much more efficient. Look at how many meetings we're doing by Zoom. Because we have Zoom fatigue, I think a lot of us just realized that meetings are actually like a vampire and suck blood. You need to work better, more in an asynchronous way. At the end of the day, work won't be the same after, and I hope for good.

[00:31:15] Rory: Excellent. I just thought that'd be a nice way to end UXDX 2020, a bit of positivity going out there. From my personal perspective, I said it earlier on when I was talking to Jeff Gothelf: my daughter was born in April, and had I been going out to work every day, I would have just left in the morning and come back in the evening, not seeing her. And I've really enjoyed watching her discover new things as she's growing. So that's my positivity for 2020. Thank you to everybody. Thank you, Audrey, Jenn, Parul and Benoit. I really appreciated your insights. And we'll just hand back now to John, who's going to wrap up the day.