Similarweb, Chief Product Officer Benjamin Seror’s Exclusive Interview with MarTech Pulse on Digital Intelligence

One reason I joined Similarweb is that my academic background is in applied mathematics, and I had experience applying it to digital marketing and advertising.

Years before Similarweb, at Amdocs, I led an internal startup focused on mobile digital advertising. This was before the iPhone, so we were really early. We were using mathematical models to understand user behavior and how to optimize advertising. Today at Similarweb, I oversee the development of mathematical models and digital tools for optimizing marketing, sales, and digital businesses of all sorts, in addition to advertising.

At Red Loop Media, a mobile ad network in China, I continued to focus on the transition the internet was making from the desktop web to mobile web and apps. That really was huge. Now, we have another huge wave coming on fast with AI changing the nature of the web, of the internet, and of business in general.

When I joined Similarweb CEO Or Offer as a cofounder of Similarweb in 2012, it was the next level challenge I was looking for. Similarweb gave me the opportunity to merge my passions for data analysis and mathematical modeling of human behavior in internet connected systems. The challenge, and the opportunity, is even more intense today. The fundamental math of human behavior hasn’t changed, but the scale and speed of the acceleration we’re seeing with AI are making the challenge and the opportunity greater.

What key innovations have you driven at Similarweb, such as launching the SaaS platform, API, and recent AI Studio for market intelligence?

First of all, I’m happy to take a share of credit for everything we’ve done, but we wouldn’t have made the same progress if not for the brilliance of a lot of other people driving the product forward. Launching as a SaaS platform was very natural for a company like us focused on the web, as was adding an API – part of what we call our Data as a Service offering – to make our data available for clients to integrate into their own systems and analyze in their own unique ways.

At an even more basic level, we always have to be focused on the breadth and quality of the data itself – gathering it, cleaning it, modeling it, and making it more useful. As I mentioned, I was part of early mobile internet businesses, so I’ve put a lot of emphasis on improving our data about iOS and Android apps usage and putting that on parity with our web data, which is what we were originally known for.

One of the most significant aspects of Similarweb AI Studio is that it brings all that data together in a simple user interface. You can ask a question where the answer might require web data, search data, and app data. The AI agent will act like a lightning fast research assistant with a deep understanding of our data, meaning it knows which data points to gather and how to assemble them into a meaningful narrative with supporting tables and charts. With AI Studio, every employee can make much more sophisticated use of our data, and they can do it with close to zero effort.

You still have to bring your critical thinking skills to the task of evaluating the output, but AI Studio saves a lot of time that you would otherwise spend pulling reports from individual modules of our SaaS platform – in my example, the traffic and keyword research screens of Web Intelligence, combined with App Intelligence – and correlating the results.

There are good reasons we divide up our SaaS platform into multiple screens for multiple purposes for human users, but AI Studio can bypass all that because it lives and breathes data. That also makes AI Studio the most enterprise-level tool we offer for enterprise customers who buy the full Similarweb suite. It’s the best way of synthesizing an analysis that takes advantage of all our data.

Read More – Personalized Shopping on the Move: GenAI Mobility Experiences for Connected Consumers

How does the new Similarweb-Bloomberg Terminal integration take advantage of digital metrics like unique visitors and bounce rates? How is that useful for investors?

The news we announced recently was an expansion of an integration with Bloomberg, which was already taking data feeds from us for their {ALTD <GO>} product. We’re now making more extensive data available to our joint customers. This Bloomberg Product is an example of what the investment world calls “alternative data” as in data that goes beyond the financial metrics public companies are required to release and provides insight into company performance between quarterly reports.

When digital metrics like bounce rate and unique visitors available for Bloomberg users, it validates the Martech stack as a core driver of company valuation, not just a marketing expense.

We discovered several years ago that the data we gather and analyze for digital marketing is also very interesting to investors trying to assess the current and future performance of tech companies, ecommerce companies, CPG companies – really almost any company that has a digital footprint. A significant surge in web visitors, combined with a low bounce rate, can be an advance signal of strong sales, for example.

We have our own products, services, and data feeds for investors, but having our data bundled with the Bloomberg Terminal is a massive recognition of the value of that data. It means investment professionals can see traffic and engagement metrics in a tool they use every day, side-by-side with all the other data they rely on for decision making.

How is analyzing AI traffic and visibility in generative search engines different from the work you’ve done in the past, like expanding your coverage from websites to apps?

This is another case of a new channel emerging – a new communication channel, marketing channel, and now it looks like it’s going to be an advertising channel as well. At first, we were tempted to treat it as a new form of search, and we watched referrals from the AI platforms rising rapidly over the last year or two. But it was never going to be the same game of keywords, keyword volume, and clicks. People go to ChatGPT looking for answers to their questions, not links to click on. Even if they are offered links as part of the answer, they often don’t click.

For businesses that are addicted to clicks, that’s a crisis. You can blame the technology, but really this is a factor of human nature, human impatience. People want what they want, when they want it, and if AI can give it to them, they’re going to embrace that.

Meanwhile, it turns out that even if you’re not getting referral clicks from these platforms, you can’t ignore them. Just because people aren’t clicking through to your product detail page doesn’t mean they aren’t forming opinions about your product based on the answers they get to questions about your industry or product category. In fact, getting found in AI answers may prove to be even more critical than getting found in search.

Understanding AI brand visibility and how to optimize for it is the new challenge we’re addressing with our GenAI Intelligence Toolkit. We do track GenAI Traffic, as in referrals, because those referrals are valuable if you can get them – buyers who click have typically gone through most of the evaluation and consideration phase in their conversation with the AI, meaning they’re primed to buy.

But we’ve also built out GenAI Brand Visibility tools to help you understand whether you’re showing up in answers, what the sentiment is, and how to improve your position. Regardless of whether people immediately click on a link the way we were used to them doing with search, if they decide that your brand and your product are the best, that’s a win. It’s a win whether they buy right now or a month from now.

But to win in this new arena, brands can’t rely on brand visibility as the primary indicator. It’s an important part of the story, but what ultimately moves the revenue needle is the combination of visibility, traffic, and conversion of that traffic to sales.

What 2026 predictions do you have for digital intelligence trends, like GenAI productivity tools and their impact on MarTech decision-making?

We’re entering a year where we will see more Agentic AI. I’ve already talked a little about Similarweb AI Studio, and we have a whole series of more specialized AI agents for tasks like sales meeting prep and marketing content planning. Some of our enterprise customers are also taking advantage of our data feeds, our APIs, and our Model Context Protocol (MCP) server to create agentic systems that include their internal data and data from other vendors.

Obviously, lots of other software vendors are going to be selling their own pre-packaged AI agents. As with every other generation of technology, there’s going to be a “buyer beware” process of separating the great products from the mediocre ones. Just branding something AI doesn’t make it a smart use of AI.

Beyond the buyer challenge, there is also the user challenge. AI Studio is close to a zero-training product, in that anyone who has ever used a chatbot can take advantage of it. But you still have to ask the right questions to get the right answers. And it still helps to check that the answers make sense. We minimize the likelihood of outright “hallucination” behavior by feeding AI Studio high quality data and training it to ask clarifying questions rather than “guessing” what you meant. But even the smartest AI agent will do its best work when paired with a smart user.

For MarTech leaders seeking revenue growth, what strategies do you recommend using digital data for competitive benchmarking and opportunity identification?

Such a big question! You’re really getting at the core of what people use Similarweb for. The key principle is you don’t really understand your own performance if you don’t know how you compare with others in your industry. It’s not just about whether they have more web traffic than you do, it’s understanding how they got it, whether from organic search, paid search, display advertising, or overperforming with GenAI brand visibility. Or maybe you’re strong on the web but you want to grow app users.

Understanding how your competitors are achieving success doesn’t necessarily mean you should imitate them, but their success may give you ideas about how you can exploit or create your own unique advantages.

How can decision-makers in MarTech leverage tools like Similarweb to tie online momentum directly to revenue outcomes in 2026?

Online momentum almost always proceeds revenue. It’s the necessary prerequisite, and building that momentum is both an art and a science. We can help you make it more of a science, more of a predictable, repeatable process. Quality data about what works and what doesn’t for you and your competitors makes a difference.

In many cases, we can go beyond web traffic to model how that traffic is converting to sales. We can show you cost per click for digital advertising and let you optimize your spending and conversions.

We’re here to help you improve not only digital performance but financial performance however we can.

Write to us [⁠wasim.a@demandmediaagency.com] to learn more about our exclusive editorial packages and programmes.