Enhancing Search Experiences: AI and Information Management with Gunjan Sood – Atlassian Team ’25 NA

This is Textron tv. Welcome to Anaheim Man. Atlassian, team 25.

We are just down the street from Disneyland and from where the Mighty Ducks play. And, um, I'm here with our first guest. Yes.

Uh, gsu, who's the head of product at Rovos Search. Welcome to the show. Happy to have you here.

Same here. Thank you for inviting me and having this conversation. Yes, exactly.

You know, search, when I think of search, I think of Google. Tell me a little bit about what Ro o does and how maybe you differentiate what you do from what Google does. Makes sense.

So I think you just said it pretty well by saying what Google does. What has happened is in the enterprise for decades and decades, you are getting a better search experience every day When you search for your famous basketball team or you're trying a vacation in somewhere, you get great answers, you get great experience, but when you go and search for your enterprise content, you get the same blue links every time. You have to scan through pages and pages of it to find what you're looking for.

But with Rover Search, we are changing that. We are making it so that anybody in your organization can find whatever they're looking for, and they don't have to spend so much time and drudgery in finding the content they're looking for. We make it intelligent and personalized to you.

And we just announced today, it's not about just searching Jira or Confluence. You can search across 50 plus AppSec in your organization. So be it Salesforce, be it Asana, be it Google Workspaces, be it Microsoft SharePoint.

So you have a single place where you can come ask a question, get the answer, and don't have to spend a lot of time in clicking through it. So That brings me to my next question. So how long typically would it take to search across an entire enterprise infrastructure and operation to find a specific piece of data?

It's, It's actually very concerning today. Like we did a survey of thousands, thousands of people, and the survey came back and said, 25% of workforce time is wasted in finding information like that is like, That's just, that's just not search. That's finding anything.

Finding in, yeah, finding information. And then when you're trying to find information, you usually feel it's your fault that you don't remember where this information was. Oh, I think it was in a Slack thread.

Mm-hmm. Oh, no, I think it was a document which was sent to me. Mm-hmm.

Oh, no, it was in a meeting recording, which somebody sent on a chat. So you humans are blaming themself all the time, but the reality is every organization has around 250 plus application they use every day to get their work done. So humans can't keep up with that.

So with Rover search, like what we're trying to do is like, you don't have to go through these hundreds of AppSec to find the information. You come in one place, you type the keyword, you ask the question, and we try to find the most relevant information for you without you having to click many places. So a typical search, maybe perhaps you could gimme an example of one of your customers and one of the typical searches they do and how long it takes for them to do it with vo search versus the old way.

Yeah. So we have seen, uh, user spending four to six minutes, which seems like a small thing, but when you're doing that at tens of times in a day, that it adds up up very quickly. Yeah.

So with our answers, we see those like six minutes shrinking down to few seconds. You ask a question, Hey, how much can I spend on my trip to Disney when I'm here on a vacation, on a, on a business trip? The answer is no, I can't spend any.

But you will spend going through five or seven different documents to find that answer. But we find that answer, we show it in a nice card. So you have the information you're looking for, we cite the sources where we find the information so that you can also validate the answer we gave you is the correct answer.

Um, can you tell us a little bit about some of the customers you work with and like what type of industries and how, in a sense, as you mentioned earlier, Google's may think we think of Google as a consumer search experience. That is correct. Although they are starting to turbocharge their search experience through ai.

That is correct. But I I I I'm not sure if, if you consider that a challenge or a, a threat on the enterprise side from Google. And if so, if that's true, number one and number two, how do you kind of appeal to what your solution is?

Yeah, so I'll, I'll probably portray it in two different ways. One, uh, Atlassian has been serving customers from like healthcare to finance, uh, from defense to rocket building and from government to any other sector you can imagine. And people have trusted us with decades and decades of their data.

When we build search, we build a search in the same platform, right? We extended our platform to give you that first class enterprise search experience. That's number one.

Number two, we have 300,000 customers, which every day are using our products. So our search is available in all of those products. In single click you can find the information, you don't have to teach your employees to go somewhere to find it, and every other product has to spend time teaching their employees to go to a new place to go find this information.

But with Atlassian, use Jira, use Confluence, JSM Bitbucket, the information is just in the search box that you search every day as an example. Um, well, you mentioned healthcare. That's interesting to me because a huge upside, I, I would assume in AI would be in healthcare area, which includes inordinate, almost indecipherable amount of data that's right to go through.

Is that, is that kind of one of the areas a sweet spot for you is healthcare or is that something that's kind of longer term? It's, Uh, so, uh, we need to be very careful, like when we talk about healthcare, government and fedra stuff. So we are constantly building the right controls, uh, the right securities infrastructure we need.

We are gonna have a HIPAA compliance example in our roadmap for us to support. Uh, right now we are starting and learning because our, we are still new in this business to win the customers who have little less risks associated with it. We support financial systems today.

We support a lot of manufacturing consumer products as an example. So we are looking forward to supporting health industry as well. We are building the use cases, which are so useful in everyday things you do that I can't like foresee health sector staying without them.

Is your, do you consider your product an agent of the sorts? And if so, it's interesting to me how you might play into this whole ecosystem because there are so many AI agentic announcements that are going on. Yes.

I would think it would be overwhelming for the IT folks. Yes. They're thinking, oh my gosh, you know, what fits where, how do we keep track of everything?

How do we coordinate collate? Yes. And as in a, in a sense, could you see an opportunity where you help organize what the agents are doing?

Because I, I'm assuming that a lot of these companies are gonna use multiple agents from multiple vendors. That Is correct. That's a great question actually.

So, uh, it's funny that we announced agent 18 months ago in Vegas last year. Uh, since then, like a lot has changed in the agent industry. Agent has become a very popular word.

Uh, so there's two foundations to agent agents work well when they have the knowledge. And Jira and Confluence are the brain of your company's function, what you're building, what your challenges are, what your finance team's up to, what your sales teams, our targets are. All that foundational knowledge is in Atlassian team of graft today.

And our agents work alongside, you don't have to go find them, you don't have to go like learn, uh, spend a lot of time learning, building it. So while you're in Confluence as an example, and you wanna create a project poster, once you start creating a project poster, our agents comes up and say, Hey, you know, I can help you create this project poster because I have this decades of learning on how project posters I was created in the, in the field. So the discovery is minuscule because it's right there when you need it, when you're trying to look through hundreds of tickets, which you got overnight.

Yeah. Uh, we have a theme analyzer which looks through those tickets so that somebody doesn't have to do the mundane task of looking at every ticket to find the answer. It just shows up, say, I'm here to help.

Do you want help? And so discovery is very seamless in our Atlassian ecosystem with agents. So that's something that you, that you're kind of stressing operational efficiency, uh, uh, eliminating if not significantly reducing the amount of time spent on things.

That is correct. There was a, uh, survey that came out a couple of days ago, uh, global CEOs worldwide, and they, I think 91% of them said they were going to be using AI in some case study. Yes.

They were gonna adopt and use it. I'm wondering, are you seeing an influx of interest from companies you hadn't heard from before who are interested in becoming customers because they are, they're afraid they're not going to be on the AI agent bus, so to speak? Yes, it is.

It is paramount. How we work Changed multiple times in last four or five decades. Like computers coming in was a big deal.

Web coming in was a big deal, mobile was a big deal. And if somebody didn't follow those trends, they were an obsolete business. And we know the world's top 10 companies who are no more in existence because they didn't keep up with the change.

And AI is the next wave of that. Everybody needs to come on board. And we are seeing a huge amount of interest, especially our announcements, uh, from today on VO being included in your, uh, paid cloud subscriptions.

Like our booth has been full of people asking us tons and tons of questions. Even the regulated industries where they haven't made a leap into the ai, they are actually curious. They're asking questions.

They wanna have demos, they wanna have a test tenants, and of course, like bunch of manufacturing companies and uh, finance sectors, like we already have customers who are actively trialing, using and saving tons of time. So there's only one way, like if you want to be ahead in this era of having your employees have the modern tools and accomplish more mm-hmm. AI is your way to embrace it.

So, and this is one way to get there, there Is one way to get There. I appreciate your time. It was great interview.

Um, it went by seemed by, it went by fast. But search is absolutely, I think is like one of the most crucial areas, especially now because I, I think we're seriously in an era where, um, companies from top to bottom are overwhelmed by what they're gonna do. Yes.

And they need some sort of guidepost or some sort of traffic cop to, to get them to where they need to be. It's your first stop, like when you're in enterprise, uh, search, Search Is what you do every day and it is the foundation of how our chat and agents can work on top of it to make it even better. Alright, thank you.

It was a pleasure talking to you, John. That's Great talking to you. Yes.

Well, that's, uh, our first interview. We're gonna have many more here from, uh, Atlassian team 25 in Anaheim. So, uh, stay tuned to Techstrong TV and uh, we'll see you soon.