“How to Keep Iterating in a State of Product-Market-Fit”
Interview with Sensei Labs CEO & Co-Founder Jay Goldman.
There are many challenges startups face when they try to spin out of another company, especially when they have a product that wasn’t initially built for commercial use. However, it’s possible to find success in these kinds of situations. Toronto-based startup Sensei Labs is a living example of that. They offer an enterprise orchestration platform that spun out of a parent organization called Klick Health and are finding success all on their own.
Sensei Labs co-founder and CEO Jay Goldman joined the Midstage Startup Momentum Podcast with Roland Siebelink this week to talk about the challenges and triumphs his company has experienced in this situation. Jay offered many key learnings that can be applied to startups in all industries and all situations.
- How Sensei Labs was born out of a book - The Decoded Company
- The challenges that await when turning a proprietary software into a commercial product.
- The headwinds and tailwinds that become possible in a spin-out startup.
- Why Sensei Labs started its go-to-market with a partner channel.
- What startups should keep in mind when customers ask for exclusivity.
- What a carve-out startup should focus on during its first year after spinning out from its parent organization.
Roland Siebelink: Hello and welcome to the Midstage Momentum Podcast. This is Roland Siebelink and I am a scale-up ally and advisor to many of the fastest growing startups in the world, one of which is in our studio today. It is Sensei Labs, and we are so honored to have co-founder and CEO, Jay Goldman with us today. Hello, Jay, how are you?
Jay Goldman: Hello, Roland. Pleasure to be here.
Roland Siebelink: Absolutely. Thank you for finding the time to talk to us.
Let’s talk about Sensei Labs. What do you guys offer? Who do you target? And what difference do you make for them in the world?
Jay Goldman: Sensei Labs makes an enterprise orchestration platform called Conductor. It is sold into mid and large size enterprises globally to help them orchestrate their most critical initiatives. That tends to be things like transformations, acquisitions, post-merger integrations, supply chain and procurement optimizations, ESG programs, technology delivery. The large programs that can often run multiple years and have sometimes even hundreds of people working on them and are responsible for delivering often hundreds of millions of dollars in benefits to those organizations.
Conductor is what we call an orchestration platform. It is the combination of project portfolio management and collaboration, KPI data, and benefits tracking, and collaboration all in a single platform together. And we really help those organizations to de-risk delivery of those critical initiatives and to reduce the cost of delivering them as well so that they can accelerate benefits realization.
Roland Siebelink: As a layman, my impression is it’s project management software on steroids. Is that my layman understanding in the right direction?
Jay Goldman: Yeah. That’s definitely directionally correct. We think of it as that, but really a lot of what we think of traditionally as project management software - the kinds of tools that come to mind, Microsoft project, for example, might be one of those examples - those were built for a different kind of environment and a different scale of problem and challenge. When we think of transformation, particularly - and really everything we do fits broadly under the banner of transformation - that has traditionally for organizations been a transformation project. We’re going to define a scope and a scale, we’re going to deliver a two or three-year long program, it’s going to generate $200 million in benefits, and then we’ll be done. And that is the same kind of thinking that tends toward traditional waterfall-driven project management solutions.
But we think of that as the business-as-usual state, and the way that the world exists today is more business as unusual more than it is business as usual. As we are in more and more of these unprecedented conditions, pandemics, supply-chain crisis, conflicts, and inflation, and all of those things, they are leading to an environment in which the tools and the ways of working and the approaches that enterprises have traditionally taken are no longer sufficient for them to keep up with that rate of change. We have to shift from this mentality of transformation projects to this idea of a constant state of evolution as a strategic advantage. That’s what we really refer to as enterprise orchestration. And the tools and technologies that built the environment that we’re in today can’t therefore by definition be the ones that help to resolve the challenge in the future. And Conductor is one of those next generation - if not, maybe the only next generation - tools that are built for this new world we live in.
Roland Siebelink: Okay, perfect. Is this an observation you’ve had and a particular insight into how enterprises are changing and that they need to start thinking of a different way of transforming their business that you then built a tool for a trend that was already happening? Or does it feel more like you have a way enterprises should be more effective in their transformations and you’re trying to sell both the methodology and the tool at the same time?
Jay Goldman: Yeah, a bit more of the latter. We have had the opportunity to have a seat at the table to help orchestrate some really significant transformations globally for Fortune 50 companies, Fortune 1,000 companies. As we’ve run thousands of projects that fit under those transformation banners, we’ve had a chance to see what works and what doesn’t work. We’ve had a chance to observe where the early warning detection points are in those transformations, and that is now built into our platform as an intelligence layer that helps people to look for those early warning signs that things might not be going well and to correct for them so that they can get back on track. This is the combination of technology and tools that help you to run those transformations better and then also the methodology that fits around. And all of that together is what we call enterprise orchestration.
Roland Siebelink: I’m curious how you got into this field. Did you have functional expertise and experience in this field already? Or was it just like a light bulb that went off one day? Can you talk a little bit about the history of Sensei Labs?
Jay Goldman: Yeah, absolutely. And it’s a bit of an interesting origin story. It’s a bit different than I would say most startups. For your listeners who come from a few team members who got together and started a company and have built and scaled it, we are not that story. But there’s definitely some parallels that I think they might find interesting in terms of how you pivot toward market fit and how do you think of that as a constant state. Rather than a check mark that you have achieved product-market-fit, but rather the understanding that over time you need to continue to iterate in a state of product-market-fit.
If I go back to 2013, the founding members of Sensei Labs were part of the senior leadership structure in a larger organization called Klick. Klick Health today is the largest commercialization partner for life science companies in the world. Think of it as a marketing agency, technology partner, commercialization expert that works with biotech and pharma companies to bring life-saving therapies to market. And we had built a platform for ourselves to use that was part of what was powering Klick’s ability to scale 30 to 40% year over year, which it has consistently done since it was founded in 1997. We got asked about how we were able to continue that rate of growth.
We decided that we would write a book that explained how we went about thinking through building this talent-centric, data-driven organization using technology and data as a significant underpinning, this platform that we had built internally for ourselves called Genome that runs everything that happens at Klick, and that we would tell some stories of other companies that were taking a similar approach. We wrote a book called The Decoded Company. It came out in February, 2014.
The unexpected thing that happened is we happened to write a book at exactly the right moment that was at the intersection of a trend around big data being a topic that people were starting to talk about. And at a time where people were really starting to dig in and talk about people-centric or talent-centric cultures, and what it meant to build that hyper-growth environment. And so the book did incredibly well, far better than we had expected. It became a New York Times best seller and gave us a chance to speak around the world.
And everywhere we went, very consistently, when we came off stage, there was a lineup of people who wanted to know where they could buy this software that we were talking about at a hyper-growth organization. And they wanted to know how they could build that same culture in that same environment. We are entrepreneurs and we said very quickly, maybe we should really think about how we can bring this to market as its own tool. It had never been built with that intent, so there was a lot of thinking that we had to do around that. And we got dragged into that market because at around the same time while we were out on the speaking tour, we got contacted by a CEO of - today, they are actually the second largest mortgage lender in the US, it’s a company called UWM - he had seen a demo of Genome on a tour of our office and he really wanted it for his team. He basically said to us,”Look, you’ve given me a choice here. Either I can build this thing for myself or I can buy it from you. But if I have to build it for myself, then you can never call yourselves entrepreneurs again.”
And I’m sure for all the entrepreneurs who are listening to this, that is the ultimate threat to your self identity. We weren’t gonna take that as a challenge. We said, “Okay, we’ll sell the platform to customer number one. But we’ve never done this before. We have no idea if it’s going to even work, so you have to go into it with that same spirit of experimentation and we will try this together.” And we signed them up as a customer. They launched in June of 2015, and we have just renewed into the eighth year of our subscription together.
Roland Siebelink: What were the particular challenges of turning a product that has never been designed for external use, as you said, into something that could be commercialized?
Jay Goldman: There are a few things that come up in there. Some of them are just strictly from a technology perspective. When you build a platform for internal use, you don’t take into account a whole bunch of things that you would if you were building this externally. There were some really funny ones that came up. If Klick’s chairman wasn’t the top of your org chart, the whole platform didn’t work. And so, we can only sell it to companies that were willing to take him on as the head of their org chart or we have to change the way that that whole system worked.
But there is a bigger issue that comes up here. And I think it’s probably different if you were already a software company and then you have a new product that you were maybe using internally and you decide to turn it into software.
Most of the time what happens when you are building a software platform inside of something that’s not a software company, there is a big cultural difference between running a professional service organization and running a software organization. And we were very fortunate that Klick’s co-founders, who are also co-authors of The Decoded Company, Leerom Segal and Aaron Goldstein, were both very much on board with building the software company and understanding that we were going to have to look at it differently. But I think that’s a big challenge that will come up for listeners who might be today part of a professional service organization of building a platform.
Roland Siebelink: It’s really about turning from an intra-preneur to an entrepreneur.
Jay Goldman: Right. And getting even to the point where you’re enabled to do that is a bit of a challenge. When you look at it, for example - and I’ll use professional services and software because it illustrates the point - professional services companies tend to be bootstrapped. You are from day one selling your services and your services pay for themselves. And if you’re able to realize a decent margin on that, you have investment capital to spend on growing the business. But you’re never in a position where you are investing more than the business is able to earn - generally speaking, that’s not always true. But you don’t tend to go out and raise external investment into services businesses. It’s very difficult to do that anyway because the multiples just are not there on scale.
However in a software world, that’s not necessarily true. If you have a culture that’s been built up in services around that idea of bootstrapping and funding yourselves, and then you decide to create software, you may be in a position where you have to over-invest relative to culturally what the organization is used to investing into future growth. That’s one thing to keep in mind there.
The second is that just the structure of the organization may not lend itself well to that. And by that I mean, literally at the level of something like the structure of the general ledger from an accounting perspective tends to be built around how do we track a professional services business and not built around how do we track a SaaS platform. And if you’re in the mode of, we sell contracts which are hours against delivery, rather than we sell annual subscriptions to a software platform, very different business model.
There are a whole bunch of roadblocks and obstacles. It can be an incredible headwind or an incredible tailwind as a spin-out out of an organization. And that’s really just how you manage that process and how you learn from it as to which one of those things it will be.
Roland Siebelink: Let’s go back a little bit to the go-to-market. You said that the origin story of Sensei Labs was that so many people are lining up to try and buy your software, so I guess initially your go-to-market was not that much of a challenge. But I am also guessing that that line dried up at some point in time and you had to go and find proactively customers for yourself. That’s a big challenge for many of the listeners of this podcast as well. How did you identify your ideal customer profile? How did you start reaching out to them? Basically, what have you learned about your go-to market motion?
Jay Goldman: We go to market through two different channels. One is direct to large enterprises and the other is through partners. For us, partners look like large service delivery firms, often in management, consulting, systems, integrators, that kind of thing. We actually started with the partner channel, which is a rare thing. Most of the time you get to the partner channel later. But we were very fortunate in that one of our early direct customers, some partners from Kearney, the management consulting firm, were in there delivering a pharma-related solution and they had a chance to see something that we had put in place on our platform to manage a global product launch for a pharma company. And they got very excited by what they saw, both as this should be the gold standard for pharma launch but also as this is something that we could use internally for Kearney. They have been looking for a platform for about two years at that point. That brought together into a single platform both project portfolio management and KPI benefits track in a single place.
They got quite excited and asked us if we were interested in together customizing a solution for them on our platform that they could use as their delivery platform or their orchestration platform on a global basis, which of course we were very excited to do. We’ve now had a partnership in place for several years.
Roland Siebelink: It’s like a multi-level marketing game you’re playing here. Your customers are using it and then also are bringing in new customers for themselves.
Jay Goldman: Exactly. That’s why we really liked the partner side of our business. We’ve got such a great case study there, and Kearney’s been quite happy for us to speak about that publicly, and that’s been helpful in us scaling that partner side. Sometimes, you luck out and you get an ideal customer profile in a customer that you have. Then your job is to look for the lookalikes to that ICP and find who are the - in this case, who are the other management consulting firms that might look somewhat like Kearney, maybe on a different scale. Kearney is in the top 10 from a global management consulting firm. There are obviously only nine others in the top 10. Maybe you want to look into what does the top 50 look like? And maybe they’re a little bit smaller, but they’re aspirationally trying to look like an existing customer that you have. And that makes it a great case study that you can share.
Roland Siebelink: Okay. How do you feel about partnerships when partners ask for certain exclusivity or being the only partner in their field? That’s often a barrier that I find many founders have to overcome. How do you feel about that?
Jay Goldman: Yeah, I don’t think there’s a black and white question. You can just say never do exclusivity or always do exclusivity. It’s not that simple. There are definitely shades of gray between the two. And you need to think about this as something that you are giving up in exchange for value. If you have a potential partner and they insist on exclusivity and you’re not getting anything for it, then just say, no. There’s no reason for you to give up that thing. This is an exchange of value that you need to get out of having given up exclusivity because it is something that you’re giving up.
But you need to look at that exclusivity and make a decision around what you are actually giving up. And that’s a bit of an internal calculation. If you have an extreme example and someone says, “You can’t work with anybody else in our industry,” then you are giving up a very significant volume of business in exchange for that. And often, you are the one taking on the risk. What they’re saying is “Look, we’re not working together now or if we are, it’s just a small pilot amount of revenue, but we want to expand this significantly.” And they will say, “We’re going to take a risk on picking this as the platform that we go ahead with. And you’re going to take a risk by not working with anyone else.” Those two things are not equal. Don’t be deceived by that trade off in value. You are taking on all of the opportunity costs of this being the only way that they work.
The other way to think about exclusivity or a term like that is think about it on a time-limited basis. If you have a partner that comes to you and wants exclusivity, one of the ways to manage it is to look at it and say, “Okay, we’re comfortable with this on a 12-month basis. We agree for the next year that we won’t work with this exclusive list of people that you don’t want us to work with. And we will renew that exclusivity if the following conditions are met in that 12-month period.” How do we time box this in a way that’s minimizing the risk that we’re going to take on and guarantee that for them to preserve that exclusivity, they have to do some work to get there.
Sometimes for a partner - it depends on what you’re selling and what they’re selling - just having that exclusivity is the thing that they care about. If they’re going to use you in their sales pitches - and we can use an example of you’re going to talk to a service company who wants to bring Conductor into their sales pitches, it’s going to help them sell more consulting services. They’re not really in the business of selling software. For them, they don’t necessarily care whether the software part of the deal closes. What they care about is that they look great because they have a technology platform that they brought into the pitch. They don’t want to be pitching against someone else who’s using the same technology because then that advantage goes away. But really all they care about is that they’re the only ones coming into that pitch with that technology. The exclusivity is worth it to them because they’re going to sell more consulting services, but you’re never going to get subscription revenue out of that deal. You want to time box that and say, “Let’s try this together for 12 months. And if at the end of this, we haven’t closed at least this volume of new revenue.”
Roland Siebelink: Right, exactly. Make it conditional.
I do want to ask one specific question that I think is only applicable to you folks, as people coming from a professional services company then turning into a software company, how do you avoid spending too much time on selling services as part of your software?
Jay Goldman: Yeah. It’s a great question because you will get dragged into services. And you will also get dragged into customization, and customization is a dangerous word. Configuration is a great word. When we look across our customer base, it is exactly the same code, same platform, everyone gets the same release on the same release schedule. That is the only way you should be thinking about building this if you want to look for scalability and you won’t raise money if you are really building a whole bunch of custom solutions that happen to be sharing the same product name but aren’t sharing the same code base.
What you don’t want to do is get into the business of building custom features that require custom code, then require customer leases and all of that stuff. One of the dangers here is you end up in the field of professional services, which is really just writing custom code. People are paying you an hourly rate to build features on your platform. And it’s a nightmare from a maintainability perspective and a scale perspective. But early stage, it’s attractive because you’re getting revenue in the door. That’s the thing that you want to try to stay away from.
The right way to think about it is if it will get us to the next stage of growth, we should do it today. We will either figure out how to make it scale or we’ll stop doing it when we get to the next stage of growth. If you’re an early stage enterprise software company out there right now, and you’re trying to keep the lights on, and the only way for you to do that is to take some money for building some custom features, build the custom features. That’s the right thing to do for where you are today. Just don’t look at it as a scalable practice. Cause it’s going to break when you start to get to a higher level of scale.
Roland Siebelink: Jay, last question. How big has Sensei already become? What kind of traction are you seeing? And how big could it become 10 years down the road?
Jay Goldman: We grew about 40% year over year from last year, which is a pretty solid enterprise. We would love the number to be higher and are aiming for a significantly higher number this year. But last year was really our first year of post carve-out. One of the things that is just a general expectation on a carve-out - this should be something if you are a carve out and you want to manage with your investors - you should think about that first year as a foundational year. You can get lots of stuff done, but you’re also going to have to put a whole bunch of things in place that you didn’t have as part of the parent. That is rebuilding some of your fundamental systems.
Often as a carve out - and this was the case for us. We were using our parent company’s financial system up until that point. We had to set up our own financial system. We didn’t have a CRM in place. We had to put that in place. We had relied on a whole bunch of roles and responsibilities in our parent company. We had to hire for those same roles and build those teams. You lose about a year there to a whole bunch of that foundational work. Really important, and in a way nice because you’re actually starting at a clean slate. We were able to put in place the right systems and the right people, rather than having organically built this thing over a period of years.
We’re about four and a half million in ARR when we closed the round. And we’re about 6 million in ARR today. It’s been a good, solid growth period for us, even though we were doing all of this carve-out operation. But that was really a big part of that year. We have now got all of that in place. And so, for the last few months have really been in growth mode and focused on that scale.
In terms of total adjustable market for us, it was never really a conversation that came up that much with investors because it has the potential to be almost a virtually infinite market in the sense that we help organizations to orchestrate human potential. That’s the best way to think about what we do. And so, today we’re focused on the large enterprise end of that market. But there’s lots of potential there. Really, we work with companies who do things - is our market. And then the market of companies who do things is pretty big. We’re not concerned that we’re going to run out of market space for us. In all likelihood, the most likely path we end up on is that we are disruptive enough in the enterprise software business that one of the much larger players wants to acquire us. And at some point, someone is likely to put an offer in front of us that we can’t say no to.
But we build our company from a culture perspective, from a systems perspective, and especially from a people perspective with that long term in growth. We are always building toward the state of we will continue to build and scale this company forever. We will have an IPO at some point. We will continue to grow past there. And that puts us in the right long-term mentality rather than a short-term mentality, where we might make some bad trade-off decisions that emphasize our quarterly results over what will be a long-term growth for us.
Roland Siebelink: A hundred percent agree. This is very inspiring. Really appreciate it,Jay Goldman, co-founder CEO of Sensei Labs. Where can people go if they want to hear or find out more about Sensei Labs and what should they download?
Jay Goldman: Yes, absolutely. You can buy a copy of The Decoded Company anywhere where you can buy books. If you prefer audio books, there is an audio recording on Audible and other places where you can buy audio books. I will warn you in advance that if you’re tired of the sound of my voice after this podcast, don’t buy the audiobook cause I did the audio book as well. If you’d like Decoded, it’s available in any of those places.
If you want to learn more about Sensei Labs and our conductor platform, you can visit us at senseilabs.com. You can also find us on various social platforms. We do post a lot of content on LinkedIn, for example. Find us on LinkedIn and follow us there. And if you’d like to connect with me, I am Jay Goldman on pretty much every platform out there. You can find me on whatever social network you enjoy and reach out to me there.
Roland Siebelink: Thank you so much, Jay Goldman. Again, co-founder, CEO for Sensei Labs. Thank you so much for your time and inspiring teachings, I really appreciate it.
And for the listeners, thank you again for listening, tuning in. And we’ll have a new episode for you next week.
Roland Siebelink talks all things tech startup and bring you interviews with tech cofounders across the world.