Ep. 348: Kurt Muehmel is the Chief Customer Officer @ Dataiku, the platform democratizing access to data and enabling enterprises to build their own path to AI in a human-centric way. To date, the company has raised over $146M in funding from some of the best in the business including ICONIQ, Firstmark, Battery Ventures and CapitalG to name a few. As for Kurt, he joined the company over 5 years ago and has risen from AE to VP EMEA to VP Sales Engineering to today as Chief Customer Officer. Before Dataiku, Kurt spent 5 years at Deloitte as a Manager advising primarily European public authorities on sustainable development policies.
In Today’s Episode We Discuss:
* How Kurt made his way into the world of enterprise SaaS with Dataiku having started his career at Deloitte in Paris?
* What does it take to go from 0-1 in implementing both AI and data science disciplines in 20th-century companies? Where do many go wrong with their first steps? How can one assist them in the right way? How does Kurt feel about services revenue? At what stage or ratio does it become too much?
* How does Kurt approach the challenge of change management? What does great change management look like? Where do so many go wrong? How can content be used to efficiently scale change management practices? How does one need to engage different teams for effective change management?
* How does Kurt think about the right pricing mechanism for the customer today? How does one find a mechanism that does not disincentivize the customer with usage? How does Kurt feel about discounting? To what extent is Kurt and Dataiku willing to engage with pilots and POCs? Where do many go wrong here?
If you would like to find out more about the show and the guests presented, you can follow us on Twitter here:
Below, we’ve shared the transcript of Harry’s interview with Kurt.
Harry Stebbings: We are back, and this is the official SaaStr Podcast with me, Harry Stebbings. And it struck me the other day that in over 345 episodes, we’ve never had a chief customer officer on the show. Appalling, I know, but I’m pleased to say that changes today. And I’m thrilled to welcome … and I’m going to butcher this name, so I’m so sorry for this surname pronunciation … Kurt Muehmel, Chief Customer Officer at Dataiku, the platform democratizing access to data and enabling enterprises to build their own path to AI in a very human-centric way.
Harry Stebbings: To date, the company’s raised over $146 million in funding from some of the best in the business, including ICONIQ, Firstmark, Battery Ventures, and CapitalG, to name a few. As for Kurt, he joined the company over five years ago and has risen from AAE to VP of EMEA to VP of sales engineering to today as chief customer officer. And before Dataiku, Kurt spent five years at Deloitte in Paris as a manager advising primarily European public authorities on sustainable development policies.
Harry Stebbings: But enough from me, so now I’m thrilled to welcome our first ever chief customer officer, Kurt Muehmel at Dataiku.
Harry Stebbings: Kurt, it’s such a pleasure to have you on the show today, our first ever chief customer officer, so thank you so much for joining me today.
Kurt Muehmel: No, thank you as well, Harry. This is really great. I’ve always been a fan of SaaStr, and so it’s really great to be speaking with you.
Harry Stebbings: Well, that is so kind, and charm will get you everywhere. But let’s start with, how did you make your way into the world of SaaS and come to be at Dataiku today?
Kurt Muehmel: I’m originally from the US, but I found myself living in France for a long time, which sounds very romantic. And I guess in some ways it was. It was a personal connection that brought me there and still persists to today. But basically, I bounced around from a number of different jobs, did some consulting for quite a while, working on actually some other topics, sustainable development, more environmental topics, things like that, doing good for the world, but I’d always been enamored with a business model of the fast growing venture-backed startup often with a SaaS business model.
Kurt Muehmel: And so I’d gone through, I’d done my stint in consulting, ended up at a big four consultancy, felt that I needed to get away from that for many of the reasons that people get away from that, just the work life balance and so on. And you know what I did is I actually did something I’d never done before. I went to a job fair, actually. And so I was in Paris and back then there was a group running a startup job fair. And I went there with a stack of CVs and I passed them out. And thankfully, I met this 15-person company, Dataiku at that time, and they were willing to take a bet on me.
Kurt Muehmel: I joined, started interviewing with them. In the second interview with Dataiku, I think I picked up that they were interviewing me for a sales position. And I said, “Hey, I’m sorry, guys. I think there might be a misunderstanding. I studied philosophy in university. I’ve been consulting on environmental and sustainability topics. I think you’ve got it wrong.” Thankfully, they convinced me otherwise. And I actually got started on the sales side of the organization at Dataiku, which was great because I realized that it was actually all about understanding customer needs, understanding our product, understanding the technology, understanding the market and mixing it all up and finding a way to make it all match, to match the needs to some sort of solution and then to take care of the business considerations to make that work.
Kurt Muehmel: And so, yeah, so that’s where I got started at Dataiku, and over time progressed through a number of different roles, managing a sales team, managing the sales engineering team, until I was able to step into my current role, chief customer officer, overseeing everything customer success, professional services, as well as our teams supporting our online user community, our online training platform, which I stepped into last September.
Harry Stebbings: I always question whether people are destined for certain roles. How important do you think it was that the Dataiku founders were so plastic in terms of their willingness to take you in a very different role to what you expected?
Kurt Muehmel: I think it’s really essential to success. And I think a lot of what Dataiku has done right, and this is really the founders who were very good at that, was looked for the qualities that they felt were universal qualities around being hardworking, having the right level of intelligence, but not always having the specific background that you’re looking for, or that would normally lead you to that role, especially early on, right? And especially early on when looking to hire. It can be difficult, right? And it can be difficult to find the person with the perfect profile.
Kurt Muehmel: And so I think bringing people in, giving them the opportunity to succeed in a given role in taking that as proof that they may be able to succeed in another role as well, I think it’s a really good strategy for founders to take to build a successful company. Then, you have this stable of, let’s say, multipurpose players who have been with you a long time, who maybe have technical exposure, who have business customer-facing exposure. And with those folks, you can really put them in the position that you need in the future, because I think that’s one of the challenges, as well, when you’re building your team, especially as a founder, is making sure that you’re not only meeting your needs today, but you’re equipping yourself with the right people who are going to be future leaders in the organization as well.
Harry Stebbings: I want to start to stay on the buzzword of AI and all enterprise boardrooms are talking about it. When you think about going from nought to one in the core disciplines of AI, what does that take and I guess what needs to be in place for it to be successful?
Kurt Muehmel: If you’re a large organization, a traditional organization … What I mean by traditional is not a digital native company, right, for whom these topics are probably their bread and butter … but for a successful 20th century company that is now looking to be a successful 21st century company, indeed getting started can be a daunting task. What do we do? Do we hire a team of data scientists? Well, that’s difficult because I think I’ve seen some statistics saying that the large Silicon Valley companies are sucking up something on the order of 85% of the top data science graduates. What do we do there? Which technology do we start with? Do we need to be in the cloud or not?
Kurt Muehmel: I think there’s actually a lot of ways that that question can be simplified. One, is it almost certainly, especially well-established companies, they have data available, and we have to start with some raw material here, and that raw material is the data. They have that data available in their data warehouses, in their CRM systems, so there’s that available.
Kurt Muehmel: Now, the second question, though, is how do we actually formulate the problem and how do we start tackling that? That one is one that the company really does need to own themselves because they know better than anyone else what they’re going to need to do for their business.
Kurt Muehmel: Now, where external help can help is in thinking about ways in which more modern analytical techniques, data science, machine learning, can be applied to those business problems because ultimately where the value is, where you get from zero to one, is not just by going and doing some sort of toy example, but really by identifying a business problem, identifying the right data, which might be useful for informing that business problem, and then matching that with one of these new techniques to approach it.
Kurt Muehmel: And there’s a number of different ways that you can do that if you don’t have that skill in house. One of course, you could work with a consultancy, and there’s a lot of great consultancies out there. We have the pleasure of partnering with many of them. Two, there is a technology element to it, as well, where you want to make sure that you are equipping yourself with the technology which is going to allow you to apply some of those more modern techniques without necessarily being a top flight programmer yourself or having one on staff. And this gets into some of the automated machine learning capabilities, which Dataiku and other solution providers that we have in our products as well.
Kurt Muehmel: And so there’s ways in which you can start getting a step into that, but also, right, making sure that you’re leveling up the people that you have in your organization, because ultimately, you could take somebody who’s been doing some pretty advanced SQL queries, things like that, to get good traditional analytics out of your data, and that person may have the desire to move up, to start learning some more advanced techniques. And that’s, again, where equipping them with the right technology, but also the encouragement and the means to learn those capabilities, that’s where I think that companies can really start to see some of that success.
Kurt Muehmel: What I don’t believe in is that there can be a pure technology solution, that you just need to pay somebody enough money and they’re going to come in and solve it for you. What we believe in at Dataiku is really that this needs to be a capability that a company internalizes, because the question, of course, is zero to one, but then it’s really from one to 100 and from 100 to 10,000, and that’s where building out that capability internally can be a really strong advantage and strategic differentiator for that company.
Kurt Muehmel: But indeed, the first question is zero to one, and that’s just what do we need to accomplish? Where are we going to show value? What data do we have available and how do we start attacking that? And that’s not an easy problem to solve, but it’s one that most companies are now tackling and seeing some success with.
Harry Stebbings: It’s a fundamentally new learning curve for these enterprises really to get their hands over and really understand. How do you advise them in terms of the learning curve and maybe potential challenges or missteps that they make in the early days, and is that just part of the process?
Kurt Muehmel: Yeah, absolutely. There always is. That in some cases just is going to take a few iterations to get right. The reason that this is called data science is because it’s an exploratory process. You start with a hypothesis and you look to see if it’s actually going to be solvable. And in many cases it may not be, and that’s not because you have the wrong people. It’s not because you’re using the wrong technology. In some cases, it may just be you don’t have the data yet to answer the question that you’re asking yourself.
Kurt Muehmel: Another dimension, though, of the expectation setting is that, in fact, a lot of the value that we see customers realizing may not actually be using the most sophisticated AI or ML techniques. In fact, in some cases, they’re using just more sophisticated traditional analytics, just doing aggregations and sums and joining different data sources and developing new analytics, which really are answering those business questions.
Kurt Muehmel: We are also careful to caution against the belief that so long as we can do some sort of machine learning, so long as there’s an algorithm somewhere in it, that it’s magically going to solve the problem. In many cases, it’s actually not necessary. And the problem can be scoped down, can be formalized more clearly and through a new combination of different data sets, which have already existed, but have never been combined in this way or never been aggregated or sliced up in a certain way, you can actually come up with really interesting and valuable business results.
Harry Stebbings: Speaking of the results there, in terms of the use cases, I mean, it applies to all areas of the org, but one in particular where it can be really striking is the element of customer success and customer service. In terms of applications, are there best practices you’ve seen for these new technologies in their deployment towards the customer itself?
Kurt Muehmel: No, absolutely. And honestly, for a lot of our customers, I mean, at Dataiku, we work with mostly large enterprises, Fortune 500, Fortune 1000 companies, but across all sectors, right, from financial services to retail, CPG, as well as healthcare, pharmaceuticals, manufacturing, et cetera, so really quite a bit across the board.
Kurt Muehmel: What’s interesting is that a lot of them do tackle a lot of customer-facing questions first and foremost, and this can be about customer activation, which promotions do we want to send to whom? It may be about adapting their practices to what they believe to be customer preferences, and it may be customizing that customer experience for an individual, which is a really interesting one.
Kurt Muehmel: We had an online vacation retailer, let’s say, selling package vacations. And they had this fascinating use case where basically they wanted to take essentially the same offer, the same vacation package, but present it in a different way depending on who was there, which makes perfect sense, right? Two different people going on that same vacation to the same resort, for example, they might be having very different experiences. One might be spending their time enjoying the pool, enjoying the cocktails, having a very leisurely and very luxurious experience. The other one might be signing up for all of the different adventure experiences which the resort offers, and it’s a very much active, very sporting experience for them.
Kurt Muehmel: And so what this retailer did, or this online travel agency did, was they, one, understood their customers, right? They built some predictive capabilities to classify their customers and put them in some different buckets. This person would like a more luxurious, relaxing experience. That other person, they would like the more, let’s say, adrenaline-fueled experience.
Kurt Muehmel: But then very interestingly, they took all of the different hero images, the main images that they would use to represent that, and they built a deep learning model to understand what that is, to classify it, if it’s more of a relaxing image or if it’s a more of an exciting adrenaline-rich image and then put those two together so that in real time, a user coming to that website, depending on what their preferences were expected to be, would see an entirely different package when in fact, of course, it was the same actual offer behind the scenes, which is really interesting.
Kurt Muehmel: And I see a lot of potential there, not just for online retailers, but for, really, anybody who needs to interact with their customers on a regular basis, to think about the way in which that experience can be customized in real time. It’s really fascinating to see that capability today.
Harry Stebbings: I love that application, but I also don’t underestimate the challenge in terms of implementing AI and implementing data science so rigorously across an organization. And it’s real change management. How do you think about change management effectively and how the best have done it and where often people maybe make mistakes?
Kurt Muehmel: No, absolutely. You mentioned change management and it really is that, because we’re talking about doing something new for these organizations. We’re talking about relying on capabilities that they did not have previously and in many cases, making core business decisions based on that. In the past, those business decisions may have been based just on the expert and you knew who you could trust in your organization, but also maybe you were reassured that if something went wrong, you would know who to blame as well.
Kurt Muehmel: When we start introducing some of these capabilities where it’s more automated, where there is an algorithm which is influencing that decision, that can be a change. And in some cases, change can be scary for the organizations. There’s a couple of different dimensions that we like to talk about when we’re engaging in that level. On the one hand is explainability. And so of course, there’s a lot of different machine learning techniques out there, and they have different trade-offs in terms of their performance, the speed at which they can make predictions, the accuracy at which they can make predictions, but also, let’s say, their opaqueness or the way in which we could actually get inside of them and understand what’s going on. There are certain that are really complicated and really to understand how the decision was actually made. You’re telling me mathematically the answer is going to be X, but I can’t understand why.
Kurt Muehmel: There’s others where that decision, that prediction has been made in a way which is much easier to understand. Given that this person has this value in this dimension and that value in that dimension, we think that. And so one of the things that we want to equip organizations with is the ability to understand how these decisions, these predictions are being made, but also to understand how then they’re integrating that into their broader processes to ensure that everybody is aware of what’s going on so that we’re aware of what data is being used to develop these models … because, of course, the model doesn’t come out of nowhere, it’s based on historical data … so that they as an organization are reassured by that process that’s been engineered.
Kurt Muehmel: For us, this is part of what we call responsible AI, as well, because it allows you to build with confidence so that one, as a decision maker, you’re reassured that this does make sense for the business, but also, two, that it’s not having some negative impact that you wouldn’t want to have, right? There’s various stories of AI gone wrong, where it’s done something which is harmful to a certain population or to a certain individual. And we also want to be equipping our customer organizations with the ability to avoid that, to reduce the risk which may come from that.
Harry Stebbings: Speaking of the risks, say, you have to educate them, but then you also don’t want to put them off or be too scared. How do you think about the right way to educate them, and what does that look like?
Kurt Muehmel: Yeah, it’s a combination of factors. One, it’s by educating them about the need, right? And so that’s part of just the engagements that we have with our customers, talking about the reason that they should be concerned about this, that they should be anticipating the need to have the ability to answer these questions.
Kurt Muehmel: There’s also then a technology component and we build into the product the functionality to make that easier for our customers so that they can understand where their data is coming from and how the prediction is being made. We think that’s really core to how we do things. And indeed, it’s really aligned with our core product vision that you should be able to bring in both the very advanced data scientist, but equally the decision maker, as well as the business analyst into a common platform.
Kurt Muehmel: But then yes, there’s definitely an education element as well. And we’ve developed a training curriculum for responsible AI which covers these, which covers how to identify potential biases in your model to potential biases in the data, which is informing your model, and how to make sure that that doesn’t come up so that you’re not having a negative impact, because of course, we believe our customers are trying to do good or trying to make good business impact. They don’t want to cause any harm with this, and they’re looking for support in doing that. So indeed, part of it is just, let’s say, executive level education, that this is a topic that they should have on their radar, technology, and then training, training and education. And we do that both through in person trainings, consulting services, but we’re also building that out in our online training platform, our online academy, Dataiku Academy, so that we can equip anyone who’s interested in that topic with the information that they need.
Harry Stebbings: I’m really intrigued there. In terms of the education and the more hands on personal elements, I’m a big believer in services revenue, but many are very scared of it. Many VCs are very scared of it, and most shy away from it. How do you think about the services element and the real spending time on the ground with the customers?
Kurt Muehmel: No, absolutely. And we see that as a part of our broader enablement services, right? And that’s where we are working with our customers, and this goes back to a broader customer success question as well. We’re, of course, a company that sells our software licenses. That’s how we make our money. But being a SaaS company, we need to make sure that our customers are happy and that they’re getting value out of that.
Kurt Muehmel: So we work very closely with the people who are making the decision at our customers to purchase the license, to bring the technology in house, to make sure that there is a really clear rollout plan. And that includes both just core education on how to use our technology, but also then when it’s something which is relevant to that customer, to have them then being, let’s say, the project managers on the customer side, build in that process and that training into their curriculum so that that then becomes explained not from the Dataiku side that this is something you should do, but from a customer perspective, this is something that we as an organization believe in, we’ve brought in this technology, but we also need to make sure that people are getting trained up on the other elements as well.
Kurt Muehmel: Ultimately, we’re not in a position to convince an organization that they should care about this. We will have that discussion with them, but ultimately, it’s their decision where this falls in their priority. But when they do decide that this is something that they do care about, then we work with them to make sure that that’s communicated to their teams.
Harry Stebbings: In terms of selling into these organizations, I’m always questioning whether to go top down, bottoms up. Many obviously [inaudible 00:19:01] bottoms up because of the adoption cycles. I’m really intrigued, how do you think about this? And is there ever a case for too small a contract to start?
Kurt Muehmel: We’ve never shied away from landing small within a large organization, even if it could be an organization which could reasonably have a very large footprint for us. We’ve never hesitated to land small. What we have been cautious about, though, is doing pure pilot work or pure POC work and mistaking that for success, right? We always want to be driving towards what for us is our key metric, which is annual recurring revenue, which means that somebody is signing an agreement saying, “Yes, we’re looking to work with you for at least one year.” For us, that’s always been the key. When we have done anything shorter than that, we are clear that we’re structuring this … clear internally and clear with our prospective customer … that we’re doing this as a way to make a decision about a 12 month engagement or not.
Kurt Muehmel: And so, that said, though, frequently though, we’ve landed very small at some organizations and then grown very large, in some cases, by multiple orders of magnitude, four or five orders of magnitude in some cases, to grow big. Especially when you’re new and unproven, right, it’s difficult to sell at your highest potential. You need to have that proof. And I think it’s fair that customers would want to see some proof as well.
Kurt Muehmel: Now, I think where founders need to be clear with themselves is which metric is most important for them. For us, we made a decision that it was that ARR number, and we were very careful to not confuse anything that could feel like progress, could feel like activity, but which was not ARR for those actual results.
Harry Stebbings: Now, help me out here, Kurt. In terms of scaling up those slightly smaller contracts into the much larger contracts or converting the POCs into contracts themselves, customer success is at the center, and the question always is should they be involved in the sales process? Help me out. Where do you stand on this very challenging topic?
Kurt Muehmel: Fantastic question. We are constantly discussing that ourselves. And so allow me to reassure your listeners that this is not a solved problem in a universal sense. Customer success is a relatively new practice in the scheme of things. And I think that every company is going to need to figure this out for themselves.
Kurt Muehmel: Now, where are we landing on that topic? We see a lot of value in having these two different roles and having both roles present and involved, both in retention activities and expansion activities for our customers. And in a lot of case, of course, the same thing is going to contribute to both, right? Ultimately, we see customer experience and positive customer experience as being the key there. And if somebody is having a good experience, that’s going to protect you, you’re going to have retention, and it’s also going to create the conditions in which expansion is going to be easier.
Kurt Muehmel: But the way that I like to think about the role of customer success is in some ways like an ambassador, an ambassador from one country to another. This is somebody where you never doubt the affiliation. The American ambassador to France is an American citizen and is clearly representing the interests of his or her home country, but they take pains to know the language, the local language of their host country, to be ingratiated with their host, to understand what’s going on in that organization. And that’s a different role than, typically, what you’re able to do when you are going to be the person who’s responsible for managing the contractual aspects, managing the financial engagement of that.
Kurt Muehmel: I think insulating the customer success from those questions, for us, it’s worked very well and allows us to develop a different relationship with our customer, a positive relationship. The customer, of course, they understand that the customer success manager is working in the interest of Dataiku, but there’s a different rapport which can be created. And so for us, we’ve still landed on having both a customer success manager and an account executive present in every single account, clearly differentiated roles, but working very closely when they come back to the office together, to make sure that we have a clear and coherent strategy.
Harry Stebbings: Speaking of CS’s role in the expansion, I’m always questioning pricing mechanisms, because you can have the variable pricing mechanisms, which can disincentivize, really, the upsell, but you want to be so aligned to your customers. So how do you think about the variable pricing mechanisms and the right one that aligns you to the customer?
Kurt Muehmel: We try to be very pragmatic about that and we don’t ever want the licensing model … Allow me to distinguish between the licensing model versus the price point … to be a barrier to expansion. And so where we have been very flexible is … very flexible, moderately flexible, let’s say. Not everything can be customized. But we have, where needed, shown a lot of flexibility and, I think, response to customer needs in terms of adapting the model. Are we charging per seat? How are we allowing for expansion? Are we giving you some ability to have some test licenses so that you can expand and then true up to that?
Kurt Muehmel: At the same time, though, ultimately we always are going to be having a pretty clear sense of what the total dollar value should be for that contract, looking at the number of users, looking at basically our standard pricing model. And for us, it’s very important, right, to defend that price point because it’s the clearest signal of the value which is expected from that. Now, we do have some standard discounting mechanisms in place for volumes, so it’s not strict binary, “No, we never do that ever.” We have some mechanisms in place for that that are standardized and that are consistent across our customer base, as well, that we feel strongly about that things should be fair for all of our customers. But then we do like to talk about value. And I know that myself, when I’m purchasing software in the negotiating phase, I say, “Oh yes, yes, sure, value, value, right, but let’s talk about the cost.” But for us it really is core to what we’re doing.
Kurt Muehmel: And we’ve set up a value consulting capability as well, internally to help our customers understand the value that it represents to them, not just our software, Dataiku DSS, but also maybe the broader analytics initiatives so that they can then communicate that internally as well. And so once again, we want to help them with that communication. We want to make sure that the value is there, but also to make sure that that value is getting communicated internally, as well.
Harry Stebbings: Now, obviously the CS is crucial there, but there’s also unit economics to think about. If the client’s contract isn’t large enough, it maybe doesn’t warrant the CS exposure that they maybe get. How do you think about the unit economics and the CS attributed to it?
Kurt Muehmel: Yeah, absolutely. For us, I think the answer to that question can change as your company grows and matures. In the early days, it’s difficult to imagine over-investing in it. And for me, frankly, at Dataiku, we weren’t asking ourselves too many hard questions about that. Is this profitable? Is this scalable where we’re at right now? Because those first customers, making sure that they have just a fantastic success and just a wonderful experience working with you, that is going to pay dividends over the coming years because they are going to be your references. You’re going to be putting them in touch with the next big contract coming in. With analysts, depending on the market that you’re in, you might be talking to Gardner, to Forrester, or others. And so ensuring that they’re having an excellent experience, I think that it’s hard to overinvest early on.
Kurt Muehmel: Of course, later on, though, the question of margins does become more relevant. And I think key to that is being clear about the segmentation of your own customers. At Dataiku, we serve a very wide range of customer installs and average contract value. And so for us, it’s important that we’re bringing the right level of service to meet the needs of that customer and doing that in a way which ultimately in aggregate makes sense for our business as well. But there’s not a hard cutoff. What we’re looking to do is adapt to the needs of that individual customer, where they’re at today and of course, where we’re hoping that they’ll go to in the future.
Harry Stebbings: Now, I could ask endless questions, but I do want to dive into a quick fire round. So Kurt, I say a short statement and you hit me with your immediate thoughts. Starting with this, what’s your biggest challenge with Dataiku today in your role?
Kurt Muehmel: The biggest challenge today is, I think, knowing which are the priorities and where I should be focusing my efforts on any given day, any given week, any given month, because there’s just so much. There’s so much potential for a company like Dataiku. There’s so much that we could be doing, as well. Things are going well, but we’re all compelled to want to improve. And so I think knowing who you can refer to in the organization internally, but also with your customers to give you that feedback on, “This needs to be the priority, this thing we need to fix, that’s the next thing that we need to work on,” or, and I would say our CEO Florian, he does a great job of helping with this, is we need to get out ahead of this because this is going to be an issue in two years from now so we do need to start working on that today. That’s extremely helpful. And so knowing who you can rely on to help you with that prioritization, I think, is really essential on a role like mine..
Harry Stebbings: What do you know now that you wish you’d known when you started your role with Dataiku over five years ago?
Kurt Muehmel: Yeah. I wish I had known that it was actually going to work as well as it had. It would have made the initial decision to join that much easier because it was a little bit difficult. We had just had a child and I was moving out of a very stable consulting job into something that was yet unproven. But aside from that, I would say that I wish I had known just how valuable this could be for some of our customers. I think maybe we even underestimated to a certain degree the value that we provide to customers in those early days. And I think in some cases, we may have undersold it, to be quite frank. And so I think having that information back then would have been helpful, but at the same time, that’s impossible, right? There’s certainly no regrets, either, for what we’ve done or how we’ve done it.
Harry Stebbings: What would you like to change about the world of SaaS and startups and why?
Kurt Muehmel: I would change the focus on growth. And maybe the current crisis is just, by definition, changing that. But I think in some cases you can end up with a mentality of growth at all costs, that all we need to be doing is showing more growth. We’re not yet worried about having a stable business or a sustainable business. We just need to grow at all costs. And I think that one of the things that Dataiku has done extremely well is to grow at the best possible pace, not the fastest possible pace, but the best possible pace that allows us to maintain product quality, that allows us to maintain team cohesion. We’ve done fabulously well in having very high retention numbers for our employees, which now five years in is just extremely useful that we have so many people who have been here for so long while we’ve of course been hiring very aggressively at the same time as well. And so I would say let’s not focus so much on growing as quickly as possible. Let’s have each company focus on growing as wisely as they could.
Harry Stebbings: When you think about who you admire in the customer and customer experience game, who would that be and why?
Kurt Muehmel: Yeah. I would say that ultimately, people who have achieved a positioning for their product or the service that they provide that exceeds just the lower level good experience, delivers me value from a customer perspective, but those who have attained that level of, in some cases almost transcendence, right, where they’re serving some higher order need or desire in their customer are really those that I look up to most. And I want to get Dataiku to, and in some cases where we have, but it needs to be generalized.
Kurt Muehmel: And so examples there, right, classic ones, Steve Jobs and Apple, right, where just being a customer was meaningful to the individuals. I think, just to bring it back a little bit closer to home in the enterprise software world, I think a company like Tableau has done a fantastic job in building out that enthusiasm among their user base, and ultimately, the great success that they’ve had over the years that they’ve been working on it.
Harry Stebbings: Kurt, I can’t thank you enough for joining me today. This has been such a fantastic discussion, and it’s such an exciting time ahead with Dataiku.
Kurt Muehmel: Thank you, Harry. It’s really been fantastic for me, as well. And like I said, I’ve been a fan, so it’s a great opportunity for me as well, so thank you so much.
Harry Stebbings: Such a fantastic guest to have on the show. And if you’d like to see more from Kurt, you can find him on Twitter @kmuehmel. Likewise, it’d be great to welcome you behind the scenes here on Instagram @hstebbings1996 with two Bs.
Harry Stebbings: As always, I so appreciate all your support and I can’t wait to bring you an incredible episode with the founder of Contentstack, Neha Sampat, next week.