David Wadhwani | Making Change at Scale

Thank you for doing this with me. Thanks Sarah, and really good to be here. And thank you all for joining us. A little bit about my background, probably there are three relevant pieces of it. The first is I was at Adobe for the better part of 14 years, and for about a decade of that, I ran the digital media business. That's products like Photoshop and Illustrator, After Effects, InDesign, the workhorse of the creative industry. Of course, what many people now know is that that business had started to level off about 9, 10 years ago, and one of the things we had to do was really reinvent ourselves. We had to embrace cloud, we had to embrace mobile, we had to transition the business to subscriptions. And if you look at the results of that, we went from probably close to 10 billion in market cap then to Adobe has now touched or come close to a 250 billion dollar market cap. So that was a major transition that I think I learned a lot going through. After that, I joined AppDynamics, which was a Greylock portfolio company, as the CEO when it was about 75 million in ARR. And there were really two big transitions, macro transitions, that I drove there. One was moving the company from a single product company to a multi-product company, and in the context of that, continuing to elongate the growth cycle there. So we were able to grow revenue about 9x in 4 years because of that transition. At the same time, we really had to bring a lot of discipline to the company as well. We had been, as many great growth companies are in the beginning, a company that was about grow at all costs, but we needed to transition, especially as we prepared ourselves for the public markets, to be a much more responsible grower. So we wanted to be much more disciplined. We moved the revenue from 50% subscription, 50% perpetual to 100% subscription, and we took operating margin from negative 80% to cash flow positive and eventually just crossing over the profitability mark in about four to five years. So that was another big shift. And as we'll talk about, the transitions that we went through, the phase shift that we went through at Adobe and AppD, had the same learning, even though the methodologies that we used were very different. And of course, as Sarah mentioned, now I'm at Greylock, have been here for just coming up on a year, investing in companies and helping other founders go through this kind of transition.

Let me start with what I don't believe a phase shift is, because I hear this all the time. I hear people talk about companies going from your first phase is 0 to 10 million, then your second phase is 11 to 50 million, and then 50 to 200 million. I just think that's the wrong way to think about it. A phase shift, in my mind, is when one or more major functions of an operating business just need to operate differently. That idea of what got you here isn't going to get you where you want to go next, that's a phase shift. And it happens at a functional level throughout the organization on an ongoing basis. And in fact, in a growth company, I would argue that one or more of these departments are always in transition. And we need to start thinking about how we run our businesses and growth much the way we think about how we build our product. This movement towards CI/CD and agile application development really needs to be there for agile business development as well. And these five must-dos can work in the context of a big corporate transition, but I'd encourage all of you to think about it more as this continuous transition at the departmental or functional level, and those same five must-dos work at that level as well.

Phase shifts come in all shapes and sizes. Things that I've noticed through my years are deal sizes. When deal sizes start to grow, every startup as they go through starts with lower pricing. As they optimize pricing, deal sizes get bigger. As deal sizes get bigger, customer expectations grow, so customer success organizations need to go through a phase shift at some point where they're treating all customers the same way to tiering how they handle customer relationships. Same thing, as deal sizes get bigger, marketing teams need to evolve the way they think from predominantly digital to maybe a mix of digital and events. Another area is when organizations or companies introduce a new product. The sales team then has to completely change the way it looks at these products that it's selling from an enablement perspective, from a structure of the sales organization perspective, from incentive structure. Those kinds of things can have major negatives and major positives if you get it right. Another big one is organizations moving from mid-market to enterprise. In that context, you don't think that much about the product team, but sometimes the product team then has to think completely differently about field escalations and how requirements from enterprise customers that are urgent need to be managed and how they organize themselves around that. And even some of the internal functions, as orgs get bigger, the data and ops teams have to completely rethink the way they do internal prioritizations and SLAs. Comms teams have to think differently about how they work with and communicate with the company as a whole. So all of these things are phase shifts, and if you try to do them all at one time, it's just an overhead that a lot of people can't handle. So that's why I really encourage companies to take an iterative, incremental view on all of this.

Yeah, so let me talk sort of at a macro level about simplicity. My view on organizations over my 25 years is that one truism of every organization is that its natural motion is towards complexity. Because anytime there's a new opportunity or new challenge, generally speaking, organizations continue to do the old stuff and then they add the new stuff onto it. And actually, that's not necessarily a bad mode of functioning. There are some optimizations you should be constantly doing, but I look at it and I say, look, organizations kind of fit into two macro states. State one is the steady state, which is your goal as an organization is to scale what you've done and to do it efficiently. And to do that, you push decision-making lower in the organization, you optimize more locally as opposed to at a macro level. And that local efficiency actually extends the growth of the business, but it also, by localizing efficiently, you also increase the complexity because everyone's doing things slightly differently. And then when a business hits a transition or a phase shift, that complexity ultimately turns into the enemy of that transition. And so successful transition requires centralized decision-making and simplification so that you can actually get your arms around what you want to move from and to. And the only way to do that is to really push hard on simplifying the core of what you've spent complicating over the last many years before you can actually efficiently transition it.

Yeah, so getting very specific and tactical as an example of simplicity, the business I inherited was a business that was in steady state for I'd say maybe a decade, 10 years. And to that point on optimization, it had been hyper-optimized. We had dozens of products that we sold into hundreds of countries, shipping in tens of languages, and we had optimized every single angle of pricing. So we'd optimize by segment, so what could photographers afford to pay versus videographers versus illustrators versus animation folks. And we had optimized by pricing, if you were one version behind, there'd be a particular price to upgrade, if you were two versions or three versions behind or a new buyer, there were all these different points of complexity that had come in. And if I'm remembering correctly, we had well over 2 million SKUs. Think about that, we had 2 million SKUs for what we were selling, and we were selling it through thousands of global partners. But as I came in, one of the things I realized is that we were actually pretty late stage into the steady state. And one of the big markers I recognized to your question is that units were actually dropping. The growth that we were getting was marginal growth, and it was coming from a price point that we were increasing release after release. And that didn't compute for me, especially the point that units were dropping, because what I was hearing from the user base was that cross-discipline was a new trend. Everyone wanted to, if I was a photographer, I wanted to learn how to be a video editor. If I was an illustrator, I wanted to understand photography. I also saw a lot of non-professionals as the next generation was coming up, not a lot of non-professionals being interested in the tool. So the solution we realized needed to be the ability to innovate faster across desktop, of course, but also starting to pull in mobile and web into the mix. We knew we had to actually innovate more quickly than we had been doing. So we needed to release weekly, and we needed to engage our customers directly on adobe.com as opposed to the partner ecosystem. So all of this needed massive simplification to actually be doable. I mean, you can't take 2.5 million SKUs, thousands of partners, and all of this hyper-optimization and move it into a new direction. So things we did as an example was we took product releases where we used to try to coordinate dozens of products to release on the same day every 18 months, and we freed the teams up. We said, look, just release whenever you're ready. If it's a small feature, just push it out, don't wait for this massive train to align. We took our messages from hyper-optimized in video and photography and illustration to just one uber message, which was creativity for all. If you want to do anything creative, just come and use Creative Cloud, which was the new product direction we were taking. From a route-to-market perspective, we started to really deprioritize all that complex multi-tier channel stuff, and we just said, look, all that matters, the number one priority by far is adobe.com. And from a pricing perspective, we took that 2 million SKUs and hundreds of price points for different upgrades and segments into a couple dozen offerings with literally two price points: a $10 a month offering and a $50 a month offering. And that really set us up to be able to say, look, this is something that we can get our arms around and that we can start to take control over again at a global corporate level.

When we transitioned the business as we were going through this, there was obviously the obvious one, which is revenue impact. Moving to a subscription service means a short-term, I think it was like $700 million of revenue impact in the following year. On the other side, however, when you look at it, we had employees and customers and investors to respond, so it was a pretty significant effort.

First of all, I put conviction on this list because if you think about when you're going through a phase shift, you have to recognize that the steady-state decision-making systems, all the systems you've been relying on to make good decisions, the reports, the research that you do, institutional instincts, that's a big one too, is defined by the framing of the past. And the data you get as you look to justify the change is only as good as the questions you ask and the user base that you talk to. And so in the Adobe case as an example, just remember I just said this, that units were declining, the growth was really coming from price. So the entire system was set up around that. At that moment, we commissioned a pricing study to figure out what should we charge as we move to subscription on a monthly basis. And the study came back with these staggeringly high monthly prices, and it just didn't feel right to me. My personal belief was it had to be consumer-based pricing. So that's where $9.99 and $49.99 was stuck in my head. And as I dug in, the problem was that the study was skewed very much toward the professionals, because it optimized for highly qualified users and highly qualified buyers. And my goal was exactly the opposite, which is we knew we had highly qualified buyers and users, I needed to optimize to the unqualified buyers and users. And so the systems really got in the way of that, and getting confidence in that was very, very difficult from a data perspective. And that's where conviction came in. So as we made that transition, it turned out quite well because we did launch, we forced ourselves to launch with $9.99 and $49.99. Non-pros came in as we expected, cross-discipline came in as expected, but then we got nice sort of additional tailwinds from emerging markets and probably the biggest single one were pirates, just people that wouldn't pay the $2,500 to buy something like a Creative Suite were now willing to pay $49.99 to buy Creative Cloud and be legit. So that's where you can't just rely on the data, you have to also rely on your own beliefs.

I've heard that from so many founders, but I'm glad to know you're now legit.

Throughout my entire career, I think any organization you talk to says, I wish we could make decisions faster. During a phase shift, it's never more important than that moment to be able to make decisions quickly. Because if you think about decision-making in the steady state, it's really, I look at this as sort of like a federal versus state kind of rights. In a steady state, you're really focused on state's rights, which is you're pushing the decisions down in the organization throughout. In a transition or phase shift, you really need to pull decision-making back to the federal authority or a small group of people who are just going to make that decision. I think of it actually going very much back to this founder-led, command-and-control model. And so we realized very early on that one of our core competencies had to be, if we're going to make this transition, cycle time to decision, because we had hundreds of decisions to make to really transition the business, and that was going to be a problem. We had an organizational problem. And so what we did was we put together a small cross-functional team of people that I led, that didn't all even report to me at the time. And at the end of the day, every day, we did a standup meeting. The standup meeting was for an hour, and I would say, in that group, we were able to churn through so many decisions that needed to be made. We'd probably make about 5 to 10 big business decisions every day, often with a good amount of data, but not always. Sometimes we just had to make it with our instincts, because you just have to go. The enemy to the transition is the delay on the decision, and it's just faster to make those decisions even if it's wrong and then come back later and fix it and optimize it. And the only other hack that we kind of put into this was that this was such a big shift at the company level, what I also did was Shantanu Narayen, who was the CEO I reported to at the time, he and I had a standing 5:00 p.m. meeting every day at the end of this other standup I was doing. And that meeting, whether he was traveling or I was traveling, rain or shine, we always met for 30 minutes. And what I was able to do was any decision I thought was super controversial that was going to impact how we communicated to the street or huge impact on potential revenue risk, I would be able to talk to Shantanu that day, and the two of us would be able to make a decision, and then we could go back. So no decision, no matter how big, was held up because of process for more than 24 hours. This is a critical part of it, and you just got to be willing to make decisions as a command-and-control organization, make a few mistakes, and correct it as you go.

Yeah, and this takes us down the path of communication. But first of all, tactically to what you were saying, I think without a doubt, some of the smartest people in the room were people who had actually created the last big innovation, which was creating Creative Suite, and drove that business as effectively and efficiently as they did. Many of them really needed to rethink the new framing. And so that's where we kind of get into this fourth one, which is around communication. If you think about it, simplicity is what makes the transition possible. Conviction is what gives you, frankly, yourself the empowerment you need to drive things even when the data isn't there. Decision-making is all about executing efficiently. What communication is about is bringing people along, including and especially the people you just described, which are people who had been in the business for a long time, understood the business extraordinarily well. But as I think about communication, one of the principles we put forth is that communication is about the same communication. I don't want to manage different communication to different groups. So I want to be able to tell the employees, I want to be able to tell customers, I want to be able to tell investors the same story. And we really took a perspective that in order to bring people along, we needed to have communication be a two-way street from the beginning. And the whole idea that we pushed was, look, if you believe in something, promote it, but then listen, then improve it, and then go back and promote it again. So I really took that to heart and implemented that. So what I did in terms of communication and bringing people along was to decide I was going to talk to everyone in my org in a way that could allow for a two-way communication, and that was, I don't know, 5, 6,000 people, something like that. So we started with the top 100 influencers. These were all the VPs that reported to me plus handpicked people who just had huge networks of influence within the company. I had an hour of content where I explained what we were going to be doing and three hours of discussion where they gave me all their concerns and we kind of worked through it. And then we incorporated that feedback. And then I did 25 more of those sessions with the rest of the org, with people of 200 to 250 people in a room. I also made sure that that group of 250 people was always cross-functional. I didn't want to be overweighted to product or sales or marketing or customer success or any field, because I wanted them to all work through it together. And we'd do those sessions for 30 minutes of me talking, two hours of discussion. I incorporated all that feedback. We then went to the all-hands and shared this with the entire company based on the feedback that I garnered over the previous couple months that it took me to do this. We then went to investors and had a broad investor day where the same message was there. And then eventually we had our user conference, which had I think it was about 6,000, 7,000 people in the audience plus 250,000 streaming live, and that was the first day of that conference that we announced all of these changes. And then we had three, four days with that audience to explain in more detail. And then we just took it on the road. We did about 100 regional sessions around the world. So the point on communication is it's got to be open, transparent, singular, so everyone understands the motivation and trusts the motivation.

I think you bring up a good point in sort of re-emphasizing this. This is the part I think a lot of people run past. Communication is so essential, and you frankly can't over-communicate out there. And it took us years to do it, so absolutely agree.