ThoughtSpot CEO Sudheesh Nair On How AI Is Transforming Data Analytics And The End Of ‘Froth’ In Silicon Valley

On the eve of this week’s ThoughtSpot Beyond ’23 conference, Nair spoke with CRN about the company’s new Monitor for Mobile software, how the new generation of GPT AI technology will transform how people interact with business intelligence tools, and the changing competitive landscape in the data analytics technology sector.

ThoughtSpot Beyond ’23 … And Beyond

Data analytics software developer ThoughtSpot is holding its ThoughtSpot Beyond ’23 virtual conference this week where the company debuted its new Monitor for Mobile software for delivering analytical results to mobile devices and previewed its ThoughtSpot Sage AI-based search technology, among other announcements.

ThoughtSpot Beyond ’23 comes amid the uncertain economy that’s impacting many segments of the IT industry, causing customers to take a closer look at their IT spending, and drying up the venture capital that fueled many Silicon Valley startups in recent years.

The conference is also taking place as the buzz over AI has grown to a roar in recent months as a wave of IT companies rush to add OpenAI’s ChatGPT and the underlying large language model technology to their products.

ThoughtSpot CEO Sudheesh Nair spoke with CRN on the eve of the Beyond ’23 event, offering his views and opinions on what AI means to ThoughtSpot’s products, the current state of the competitive data analytics market, the channel impact of the cloud hyperscalers’ rapidly growing online marketplaces, and a range of other topics. The interview has been edited for readability and clarity.

What do you think are the highlights of ThoughtSpot Beyond?

I think, like it or not, everyone’s thinking about natural language, large language models and GPT. I think that will be the primary point of view that people want to hear [about]. Partly because people have questions and partly because people are curious. We are one of those companies who always had a point of view that natural language is critical for democratizing access to insight. So I think that will be the focus.

What has been your core message to customers and partners at ThoughtSpot Beyond?

I think that we want to show that we are living in a time that is unlike anything that we have experienced because the hype behind natural language and AI is no longer hype. It is probably the biggest opportunity and the biggest threat for pretty much every business that’s built around this [data analytics] space.

And this is not the time to sit on the sidelines. Particularly speaking about channel partners, this is the time to go help your clients navigate [AI] in a responsible way. In the Silicon Valley bubble ThoughtSpot and OpenAI and others are focused on fast-paced innovation. But a large pharmaceutical company or a bank, they have constraints and questions around, ‘Can I responsibly deploy these innovations while having security, governance and privacy?’

And that gap is the opportunity for forward-looking partners. If they do it right, this could be one of the biggest things [in] value creation for their own business, but also value creation for their customers.

What kind of feedback are you seeking from partners and customers at ThoughtSpot Beyond? What do you expect to hear?

First, because [AI] is not a fad, I would like customers to understand that ThoughtSpot has designed probably the most sophisticated, most well-designed, secure platform to leverage natural language for data analytics. What I would like to hear from customers and partners are their questions and concerns about productizing [and] bringing it into production.

For example, if you are an insurance company and you have 20 years of claims data, you don’t want to give it to Microsoft or Google or OpenAI because that’s your competitive advantage. How would you be able to bring that into the world of AI without losing control of the data? That’s the kind of questions they will be thinking about.

And we would like to hear more about their plans around [AI], their questions and concerns around it, because I think the days of selling and buying, you know, vendors and customers, are necessarily over. Things are changing so fast. You [partners] have to truly build a co-innovation partnership with vendors with partners being the connecting tissue between customers and vendors.

At ThoughtSpot Beyond the company introduced Monitor for Mobile. Why is this important and what does it mean for partners and customers?

It’s very simple. There is so much noise in the world that we’re living in that often the answer that you’re seeking is there in the data, but you don’t find it because you did not ask the right question. That’s the status of current BI [business intelligence]. The data, the signal, is hiding somewhere. Analytics, as defined today, will not give you the answer unless you ask the right question.

The reason I’m excited about Monitor for Mobile is because we are moving analytics from insight to foresight. Two things have to come together. No. 1, you have to have contextual conversation, not search, search, search. No. 2, you should not have any tax on your curiosity. In the data world the tax comes from SQL and other [query technologies]. With a natural language [interface] Monitor for Mobile solves for both. Insight will come to you when you need to know, where you are. You don’t need to come to ThoughtSpot.

So going from data to insight to action will be fundamentally different if you open it up and make analytics follow you where you are and speak the language that you speak. That’s Monitor for Mobile.

ThoughtSpot has made a big move in the AI-enabled data analytics space with ThoughtSpot Sage. What does this mean for ThoughtSpot, its customers and its channel partners?

Anyone can take GPT and translate [queries] into SQL. That’s not a complex thing anymore. What is complex is how do I know that the answer I just got is the right answer? Remember, the biggest difference between words and numbers is that words are very fungible. If I go into a meeting and I say something and someone disagrees with me, there is a lot of wiggle room for me. Numbers don’t give you any wiggle room. Answers have to be precise. They have to be believable, they have to be defendable.

What Sage does is connect this massive gap between expressability and accessibility. Sage gives you the option to communicate with all the beautiful vagueness of natural language but then it understands the intent and then translates that into precise queries, executes them on extremely complex and massive-scale data warehouses like Snowflake, comes back with the right answer, visualizes [it in] the best way, and then lets you interact with it, drill [down] anywhere, ask anything further.

And the beauty here like [former U.S. President] Ronald Regan [said] is, ‘Trust but verify.’ You verify and then you trust and then we introduce humans into the loop, which means that a human can give feedback so the system will learn from its mistakes and then improve. This comprehensive governable package is what Sage is. This is why we believe that we are making [AI] accessible. All this crazy innovation that’s happening in Silicon Valley is now accessible to the world’s largest enterprise organizations that are extremely secure-conscious and governance-conscious companies.

ThoughtSpot recently announced an expanded strategic alliance with Google Cloud. How does that impact ThoughtSpot’s competitive position and what does it mean for partners?

We’ve always been a believer that systems should be open and we should give choices. And our primary reason for working with Google is to give customers choice—which cloud platform they should build on, which data platform to build on. That was the primary goal.

The secondary goal is [Google business intelligence platform] Looker. Looker has become a good modeling tool and a good semantic layer. ThoughtSpot delivers good answers, accurate answers, but the data needs to be right and it should be precisely modeled. So we have [integrated] the Looker modeler and ThoughtSpot Data Modeling Studio. These things will make a customer’s job simpler because we don’t have any illusion that ThoughtSpot will be the only tool customers need. They will need multiple different tools. And we want to make sure that we are not introducing a proprietary layer that forces customers to be stuck in the world of ThoughtSpot.

So partnering with Looker, partnering with Amazon [AWS], partnering with Google: We are simplifying the modeling …using large language models and AI, but at the same time making it accessible through Looker or [data transformation technology developer] dbt, which means customers can have a central [data] repository and ThoughtSpot will seamlessly join them. So it significantly reduces the cost of operating ThoughtSpot.

The third reason is that we are part of Google Marketplace. A lot of customers have made commitments to Google and they’re trying to figure out how to get more out of it. Customers can now buy ThoughtSpot through Google Marketplace. And this is really important for partners because a lot of partners have their services business and sales business tied with Google. And giving them the option to sell and service through Google Marketplace credits will be a one-plus-one-equals-three model for partners.

In December ThoughtSpot announced that ThoughtSpot Team and Pro Edition were available through the AWS Marketplace. Do you see such online marketplaces as the wave of the future? And what does this man for channel partners?

OK, I’ve got to be honest here. And the honest answer is, it’s not good for partners because large hyperscalers are getting more greedy and they’re trying to have end-to-end control of the entire procurement process … In that process, they’re going to steamroll everyone, including partners. That’s what’s going to happen. That’s the reality of it.

So in that context, partners have to be really careful that A) they have a hybrid option, and B) they have to build capabilities, and C) they have to make sure they’re going with the best-of-breed stack. They can’t keep saying, ‘I’m going to be all in with Salesforce, all in with Microsoft, all in with Google.’ Because the reality is, these companies are looking out for themselves. And I’m not saying ThoughtSpot is magnanimous enough to look out for everyone else. We are all trying to figure out our P&L.

So, if you’re a partner, A) you have to be ready with optionality, B) you have to build core capabilities, C) you have to figure out how to deliver best-of-breed. That is, instead of saying, ‘I’m just going to sell [the] Microsoft stack for everything,’ you’ve got to figure out what is the best use case where ThoughtSpot will shine. And that becomes a difference maker for a partner. That’s the reason why a customer would want to work with a partner. I go to them because they will understand my requirements. They will figure out how to put together the best stack, make it operationally simple enough, and make sure it is aligned with the procurement goals and the marketplace credits. So that is the opportunity and threat for … partners.

What are the key technology trends and what are the demand drivers in data analytics today? What are partners and customers asking for?

I think BI companies and analytics companies have become fat, lazy and happy because once BI goes in, it doesn’t come out. They get complacent that customers will never throw us out. Because remember, an analytics dashboard is like a security blanket. Once you get it, you don’t want to give it up. So companies like Tableau [Software] have exploited the customers because of the complexity that comes with being an incumbent.

That is what is changing because of AI, because for the first time there is a compelling reason to do something different. … A dashboard tells you what happened in the past. When you saw [ThoughtSpot’s new product announcements], you saw we are taking it from ‘what happened’ to ‘why it happened’ and ‘what can I do about it?’ I think this continuum requires constant insight, constant analytics on real-time data. For the first time analytics companies are realizing that you can’t be counting on … the power of incumbency to build on the next 10 years. So I’m very excited because of the chaos [this has] unleashed in the world of analytics.

So, for example, Tableau will come out and say we have GPT support too. [This week Salesforce announced Tableau GPT for AI-powered analytics.] The simple question that customers need to ask them is, ‘What are you doing on top of dashboards?’ Dashboards are static and non-interactive. The only thing you can do on top of a dashboard with large language model is to explain the dashboard. My question would be, if you need to put [an] explanation on top of the thing that you built, doesn’t that mean that your end-product is not easy to understand?

My point is what we have done is fundamentally different. We are not simply explaining things, we are helping with input and output. … Input is search. That’s what natural language is built for. Because that is how we as human beings learn, we ask questions. … Natural language is built to interact, conversational analytics is the future. ThoughtSpot is the only company with that point of view. …

So yes, they have the scale, they have the distribution, whether it’s [Microsoft] Power BI and [Salesforce] Tableau, they all have a point of view and they will say something [about AI] so they can check the box. AI is not a check-the-box thing. AI is upset-the-applecart. Can you rebuild everything with AI as a baseline? So sometimes when people ask me, ‘How will you compete against Salesforce? Or Microsoft?’ I’ll tell them, ‘Look, it’s not a question of how ThoughtSpot will compete. How will they compete against ThoughtSpot?’ Because we do only one thing, we are not distracted. [We’re] a smaller, younger company, which means we are going to be faster. Today more than anything else, you need to be faster rather than bigger.

I am so confident that we will focus and solve this one problem with AI in a delightful way and iterate so fast that the large companies will have a hard time keeping up with us. So, unlike the last 10, 15, 20 years, this is a unique time where BI will have to fundamentally be flipped on its head.

­What is your assessment of the competitive landscape in data analytics? And specifically, how often are you seeing Tableau Software in competitive situations?

Tableau [Software] was a company that built an amazing partner ecosystem. They built [on] partners like InterWorks, Slalom and others, from small to powerful medium-sized partners. And Tableau got so much in return in the sense that [partners] did passionate advocacy, they actually built services, they built solutions and all of that. Then they got acquired by Salesforce [in 2019].

First two years, things were stable. Guess what, this year, Salesforce went to [partners] with a Faustian bargain, basically saying either you sell our entire [technology] stack or we will prioritize the [partners] who are selling the entire stack. Which means that unless you are able to sell the entire Salesforce [product line] … pure Tableau partners have no place left. That is not good or bad. That is the business of doing business. That makes sense. Because [Salesforce is] building a $30 billion, $40 billion business and in that context, a $100,000 deal or a $300,000 deal means nothing. So as the laws of large numbers kick in, partners will get squeezed out. …

And to be very honest, ThoughtSpot loves that because for us to go win Tableau customers, we need to have more partners. So we are running a race, [an] extremely lucrative partnership program for former Tableau partners to come and do what they did with Tableau one more time with ThoughtSpot and AI. We are not shameful at all. We have been very shameless in our way of reaching out to Tableau partners and saying, ‘Would you like to do what you did with Tableau last decade one more time [with] ThoughtSpot as a partner?’ That is because Salesforce had to do what they had to do, and it is our opportunity today. So these are the sorts of dynamics that’s going on in the partner world.

ThoughtSpot, being privately held, obviously doesn’t disclose revenue or earnings details. What can you say about business and growth over the last year given the uncertain economy, inflation, rising interest rates and other macro-economic trends?

It’s been a mixed bag. I will tell you what worked well and what did not work for us. I’ll start with what we are struggling with, which most companies are, [purchase] decision-making is taking more scrutiny, the sales cycles are increasing.

Having said that, I would rather be—and this might sound self-serving, but this is the honest answer—a year ago, maybe two years ago, anyone with a PowerPoint slide was able to get funding and make noise. Because there was so much froth in the market. Crypto, all of that was going on. You know, like Warren Buffett said, when the tide goes out, that’s when you realize who’s swimming without trunks. And the tide was so high we had no clue.

I like that there is a little bit of scrutiny. I like that it is back to substance over style. Because we are a company that has substance and we’ve been yelling but there’s so much noise people couldn’t hear us. So I like that.

The second thing I like about this is, we are a growth-stage company. Silicon Valley companies sometimes forget that our job is to make a profit. We are a profit-making entity, we have to make a profit at some point in time. We have to [provide a] return to investors. That kind of message is hard to keep up when companies are [making] work-from-anywhere possible, [giving] free massages, food every time—the benefits were getting out of control. Some of those things had to go. I like the fact that this tightening has brought it back to some sort of reality.

What I also like is that ThoughtSpot went through a transformation. We had to pick three different debts. Metaphorical debt, not actual debt—the company has zero debt.

The first was going from on-premises to cloud. Second was our cost model was built with most of the engineering being here [in the U.S.]. We had to move that organizational debt by finding people in other economies, other countries. So we found multiple locations, including three different locations in India.

And the third is ‘tech debt,’ the [technology] stack. We rebuilt the stack so that we will be ready for OpenAI, Google Bard and large language models. We used the [COVID-19 pandemic] crisis to do all of that.

Today we are seeing customers spending more. Hopefully in the next few months we will have some numbers to announce. But our business took that front-hit with respect to business getting delayed three quarters ago. The last two quarters it has picked up steam again and we are seeing tremendous growth—in fact the quarter we just closed is one of the biggest quarters in the history of the company. It just closed in April.

So I’m cautiously optimistic. I love the fact that the world is back focused on substance over style and companies that are built to last, companies that will make tough decisions during critical times, with good leadership and culture, will thrive.