Empowering Your Enterprise Through Generative AI with Asana CIO Saket Srivastava

Sure, first off, thanks for having me. Looking forward to this conversation. So when I grew up in India, in New Delhi, the capital of India, did my undergrad and postgrad in computer science, worked for a few years in India, and then moved to Netherlands for a bit and came to the US. And I've been in the US now for 20 plus years. And I've dabbled my career between consulting and IT leadership kind of roles. And through that experience, I've been fortunate enough to have had experiences across companies at different phases of their growth and across multiple industries as well. The last few companies that I've been at, I was at Guidewire Software, which is an enterprise software company for an insurance property and casualty vertical, and I was leading their IT teams there. After that, I went to a fintech company, Square, now called Block, and I was leading all of their technology data teams. I was playing the role of a CIO there. And over the last almost two years now, I am with Asana, again an enterprise software company, a leading enterprise collaborative work management platform.

Yeah, absolutely. So Asana operates in a space called collaborative work management. It's all about enabling teams to work more effortlessly and achieve their goals more efficiently. So it's a platform which is used by over 140,000 customers. And essentially, it does a few things. As a CIO, as a practitioner, I can say that it's a platform that helps creating better accountability and achieve greater velocity of work. When increasingly work is becoming more and more cross-functional, rarely do you find yourself doing work which just sits within your teams and within your functions. And anytime work sort of crosses over to other functions and teams, it becomes harder to be in control of the outcome of that work. And that's where a platform like Asana comes in, wherein it helps sort of create that accountability and better predictability for work. And that's very important and meaningful to us as CIOs who are responsible for running larger cross-functional programs. So that's one. The other thing that Asana does particularly well is anytime there's a body of work, understanding as leaders, we need to understand what's the work that's contributing towards moving my company forward. And similarly, for people who are actually doing the work, it's equally motivating for them to know how their work ladders up and is contributing to the company goals. So Asana as a platform helps making those connections top to bottom and bottom to top and across. So that's essentially what Asana does.

Oh, it's... I say this and I think a lot of my peers say this, this is perhaps the best time to be the CIO of a company. I say that primarily because in the past our function used to be a little bit more on the back end. We would be the technologists who would be told go do this and we'd go and implement those technologies and be very happy being in our world. But increasingly as CIOs, our work is more about being a business leader, being that business strategist, being that business architect who's able to connect the thoughts across the company. And it's important that we're able to sort of drive conversations and drive strategy for the company. There's no other function that works so cross-functionally across and sees everything that's happening across say marketing, sales, security, product, HR, finance. It's us. So it's incumbent on us to make those connections and drive directions for the company. So that's one part of what makes the CIO role exciting. Obviously we had this sort of unfortunate journey through COVID where we had all to kind of take our workforces, make them remote, and that brought us to the forefront as well as leaders who were able to make that pivot happen overnight. We didn't necessarily know going into this that this was something that we could do well, but across the board I can confidently say that all CIOs were able to make that flip. And no conversation over the last year has ever been complete without talking about GenAI. And I'm sure we're going to talk a lot more about GenAI as well. But as CIOs, I see the CEOs looking at the CIOs and asking them to drive direction for their company leveraging all things GenAI.

It's interesting how much GenAI gets discussed, and it sort of gets discussed as if it's a very new thing, wherein our company and several other companies have been working on all things AI for a long time. I think the pivotal moment was last year, 2022 November, when OpenAI came up with ChatGPT 3.5, when it became a lot easier for everyone to engage with AI. No longer did you necessarily need to be a data scientist to leverage the power of GenAI. I believe it's still early on. And more broadly, I do feel that it does feel like GenAI might be overhyped in the shorter term, but I also feel that in the longer term it might be a little underhyped because truly, I don't know if everyone understands or anyone understands the possibilities with GenAI. We're still sort of scratching the surface. While it's taking a whole lot of airtime and it might feel that we've been talking about this and this is prime time, I think this is still the preview that we're seeing, still early days. And what's fascinating even in these early days, just imagine if you just fast forward a few years out, the kind of great possibilities that we can unlock with GenAI.

When I look at Asana and the foundational data model that Asana is built around, we internally call that the Work Graph data model, it really sets us up for greater success to tap into this GenAI growth. What this really means is around atomicity of the unit of work. We've talked about what Asana does as a work management platform, but when you have to kind of have multiple teams work on a unit of work, invariably you'll see that you duplicate that body of work, you'd copy it and make multiple copies of that. What Asana does well is Asana helps you retain that atomicity of work, which means that the data fidelity and data quality that you get on the work that your organization is managing on a platform like Asana, you'll get better results from GenAI as well. So I think that's the foundation. Aside that, the way we think of Asana and GenAI, I categorize this in two ways. One is we often hear this phrase co-pilot. This is something that can help you in your day-to-day, increase your productivity and efficiency. And Asana does that beautifully well as well. If you've got all your work being managed on a platform like Asana, you can use Asana, ask Asana anything, it can summarize for you, it can create smart answers for you, it can help you with all of that. But that's just one part of it. Besides that, we also think of it as sort of an air traffic controller, which goes beyond your sort of personal productivity and individual productivity. Now when an organization's work is being managed on Asana, and this is more game-changing, this is where Asana is now able to leverage GenAI and be able to recommend to you what are your smart goals for the company. Because we know what's the work and what are the people working on within Asana, we can recommend what are your goals for the company that you should be thinking of, or we can be doing smart planning for you, or we can help you resource plan better. We can take a look at the skill set and the capability of people and the kind of work that they've been doing and maybe we can make recommendations around that for you. So more broader than individual productivity, we can now talk about organizational productivity as well. There's one more thing that Asana is uniquely sort of equipped to do well. We've got a think tank that we call it Work Innovation Lab. So in partnership with academia and our product, we are able to tell our customers, give them a collaborative score, we call it collaborative intelligence, and we give them opportunities and ideas and recommendations. Hey, this is how you're performing in collaboration, say on accountability or velocity, and this is how you compare against your benchmark companies and industries, and these are some recommendations on how you can do better at that. And that's compelling stuff. As a CIO, when I talk to other CIOs about how they can think and leverage a capability like this, invariably the reaction is their eyes are popped open.

So we've talked a little bit about how Asana as a product is leveraging GenAI and kind of going deep and giving value to our customers. Internally, we're sort of doing similarly, wherein I'm trying to understand, me and my teams are trying to understand what our vendors, the partners that we engage with, what are they doing around GenAI, having a perspective around that. Secondly, as CIOs, it's important for us to enable our workforce and prepare our workforce to understand the nuances around GenAI and be prepared to leverage this. As a company, we have decided that we're going to be an AI-first company. What that means is that an average worker at Asana should be able to leverage the power of GenAI from capabilities that we're rolling out within our product and everything that's around us as well. So we obviously at the start we rolled out policy and some guidance and some enablement for our people. But then beyond that, we started to think about internal enablement through two vectors as well. One was identifying some lofty, not-star opportunities and go after those. And I'll talk about a couple of those specifics that we're pursuing at this point in time. And beyond that, also looking at some tactical incremental capabilities that we might need to build or we might need to buy as well. And for that, we did some surveys and identified some opportunities that our people were telling us, hey, these are things that would help us. One was this notion again around a co-pilot for our go-to-market teams, for our sales teams and our customer experience teams who need to do a ton of research around our prospects and customers, who need to be better informed around our customers and how they leverage our products before going into a conversation with the customer. There's a ton of that research and GenAI is perfect to kind of do that on their behalf. So GenAI could create a cheat sheet for them, hey, this is what you need to know going into a conversation, or research enough about a lead and then create few options of messaging that the account executive can then go and evaluate and then make a decision around. We also as a company started with some first principles around GenAI. We believe and we say that AI is in the service of humans. So the accountability should still sit with the human. So these use cases that I'm talking about, GenAI would do the work, but at the end of the day, the human should take accountability and make the decisions around that. AI can be that sort of muscle and the human can be that brain and that heart. That's sort of the way I think about this. So that's one use case. The second, customer support is a classic use case. Even before all of this GenAI, IT support and customer support, there was a ton of AI already being leveraged around chatbots and stuff. I believe that up to 40-50% of that interaction can be deflected or auto-resolved before hitting a human. If you're able to do that, just imagine the amount of money that you're able to save and you're able to redirect that investment elsewhere. The third is around as companies grow, your knowledge and content gets so distributed and fragmented. How can we sort of layer in and make that search and answering to questions easier for people, wherein they no longer need to know, they don't necessarily need to have years of tenure within the company and only they know where the knowledge resides. They can just sort of search and their questions get answered. A lot of our salespeople reach out to our R&D organizations when they have questions around a product, and there's a ton of content already out there. And we've got a Slack channel wherein these account execs and customer-facing teams can ask questions around a product to our R&D teams. One-off questions are fine, but then what ends up happening is this breaks their deep work, the R&D team's deep work. And this is where GenAI can be hugely beneficial, wherein we have a use case where what we're doing now within Slack, instead of asking someone directly in R&D, you ask a bot, and this bot can research against all of the sources that we have and give that answer. And there's valuable time that you're now saving on the R&D team who can then go build great product for our customers. So great opportunities that exist today.

So one idea that I am spending a lot of time on is how do we increase our go-to-market teams' productivity, efficiency, effectiveness in the conversations. And we started, again not saying that GenAI is going to solve this problem, but we started with understanding their day in the life of, or that customer journey from start to end. And doing that sort of experience exercise, discovery exercise, we were able to sort of identify opportunities along that workflow, that journey where automation could be helpful, where GenAI could be helpful. So while this journey is not complete, we're already seeing benefits of that exercise now where we were able to baseline that this is our rep productivity today in terms of the time that they lose because of navigating systems and processes, or even a dollar that's assigned to a rep to close for the year, that's rep productivity in some ways. How can automation and how can GenAI along that workflow be able to improve that quota for a rep or be able to save time so that the rep is able to kind of be more in front of the customer as well. So those are the ways we're thinking about some of these use cases.

That's a hard one though, right? Thinking three, five years out, it almost feels like in the past you would have said decades out. That's the pace of change and innovation that we're seeing specifically with GenAI. I do feel though that the way I and I've seen others use GenAI is more on invoking GenAI to be a co-pilot and be able to help, as opposed to... so in a pull sort of manner, I need something and I'm going to invoke GenAI. Changing that experience to more of a proactive sort of push versus pull might be a good way to kind of think about the progress of GenAI, wherein it's context-aware and it's proactively making recommendations to you as opposed to you having to, hey, can you summarize this for me or can you do X for me or Y for me. That might be good sort of progress. That's sort of more broad. Getting those specifics, I'm sure there are examples that I can think of, but there are ones out there as well.

I think it's use case by use case, and it also depends on the kind of company and the industry you serve. If you're more regulated, I think you've got more restrictions. At Asana, I do believe that we leverage OpenAI, Anthropic, the variety out there. We're not necessarily restricting ourselves to one. We also don't believe that we should be creating our own foundational models. There was some report that I read from Gartner that by 2028, a high percentage of companies that are creating their own sort of foundational models will perish. It's going to get better with the larger ones as well. So that's around the foundational models, we don't want to create our own, we want to leverage what's out there. Now more specifically around partners or building those use cases, I think that's use case by use case. I want to stay close to what my vendor partners are doing. I want to make sure that we tap into their innovation as well. So what a Salesforce or a Workday or an Oracle and the likes are doing, I want to make sure that our people understand that and tap into that. But I also don't want to be waiting for them to just create a path for us. I want us to be exploring and experimenting. So this sales co-pilot that we're talking about, this is certainly something that we've decided to go build ourselves. So leverage one of the LLMs out there and build these use cases, these GPTs more specifically for Asana's use case.

It's very, very important that as leaders we are not fully consumed with our initiatives and priorities that we've signed up to deliver for our business teams. Very important to go deliver them, but it's important that you have the right team and the right priorities and the right scope and the right understanding of how those would get delivered and when. But it's important that you're creating capacity for yourself to understand what's happening around you, which might come in the shape of networking with your peers, spending some time with the VC community. But that's primarily the way that you stay close to the innovation that's happening. And then building a point of view, important to build a point of view around what's working, what's going to work, what's not going to work, what are the problems it's solving, and where there's just all smoke and no substance.

Absolutely. So maybe to start, how do you think companies should measure the success of a CIO? Business value, business outcome. How are the CIOs impacting the strategic outcomes of the company.

Take on harder challenges. Be bold. Be a business leader first and then a technologist.

Oh, you need a short answer, right? I was going to go on.

280 characters. Understand the business of the business and be a partner who has a point of view. Build credibility and have a seat at the table, not just a fly on the wall, but have a seat at the table wherein you're influencing outcomes.

There's one that I read recently. It's by this gentleman Patrick Lencioni. He's written a couple of books. The earlier one was Five Dysfunctions of a Team, where he talks about, and as leaders, it's important that we lean in on our teams. It's through the teams that we achieve success for our organizations. So it talks about what are the factors that can help make teams more successful or less successful. But the more recent one was around teams. The book name was The Ideal Team Player. It talked about some qualities around humility, around being smart, and it also talked about being hungry for success. And it talked about needing all three of this for an individual and a team to be successful, as opposed to just being hungry and not being smart or not being humble. Having all three ingredients, take a look. I think you'd enjoy the read.

I dream of a world where robotics and GenAI comes together. I'm sure you've sort of been close to the Boston Dynamics kind of robotics, right? Almost humanoid robots here, and then you sort of combine GenAI magic to that, and these could be your assistants. They can, something as small as, can you check if there are eggs in our refrigerator? If not, then go to the grocery store, get it, or order and get it. So much beyond a virtual co-pilot and real-life assistant or robot.

Interesting. Maybe intergalactic. We're already talking about travel to Mars, but beyond that, who knows.

Our lives continue to be very busy and a whole lot of our lives are consumed by the work that we do. Being able to truly have a better work-life balance because of what technology can do better and more of is perhaps what I see in the future. Perhaps not the answer that you were expecting, but I think that would be a good dream world for me, wherein people are living more fulfilling lives, more holistic and complete lives than just doing the work.

Thanks a lot for joining us. Thank you for having me. I've had fun through this conversation.