Relevance Is Not Personalization: Rokt’s Gupta On The Transaction Moment
Relevance Is Not Personalization: Rokt’s Gupta On The Transaction Moment
In the data-driven ad world, we often hear executives talk about the importance of ad relevance. What what is relevance, really? For many, that word means simply personalized ads. But Srishti Gupta, Chief Product Officer at Rokt, goes deeper than that. Relevance from Rokt “Relevance is not personalization. Completely different,” Gupta explains, in this video interview [...]
In the data-driven ad world, we often hear executives talk about the importance of ad relevance.
What what is relevance, really? For many, that word means simply personalized ads.
But Srishti Gupta, Chief Product Officer at Rokt, goes deeper than that.
Relevance from Rokt
“Relevance is not personalization. Completely different,” Gupta explains, in this video interview with Beet.TV editorial director Lisa Granatstein.
“Personalization would be like, ‘Hey, Srishti, we have this offer for you because we understand you.’
“But what we are really talking about is, well, if I am in a very different mindset, depending on what transaction I’m even in or what the context of my persona at that point is.”
Rokt‘s software platform leverages algorithms to help online businesses optimize customer engagement through targeted advertising and personalized experiences during the transaction process.
Their services include ad serving, offer recommendations, and data analytics to increase conversion rates and customer acquisition efficiency.
To achieve this level of relevance, Rokt leverages a combination of first-party data from e-commerce companies, such as cart contents, shopping frequency, and customer loyalty status, along with derived data on customer interactions and preferences.
The company’s machine learning algorithms then make real-time decisions based on this data to determine the most relevant content to display.
“To actually harness the power of this data, we need to make decisions in real-time,” Gupta says.
Maintaining a High Bar for Customer Relevance
Rokt’s machine learning models have a high threshold for customer relevance, and if the likelihood of a customer finding an offer relevant falls below that threshold, nothing is shown at all.
“We only show something if it’s net additive to the experience,” Gupta emphasizes.
“We need to maintain that super high bar, even if it means that there’s a certain percentage of times we will just not show anything.”
Experimentation and Testing for Long-Term Trust
In addition to its AI-powered decisioning, Rokt relies on experimentation and testing to validate the impact of its offerings on customer experience.
By monitoring metrics such as cart conversion rates, the company ensures that it is consistently adding value and not detracting from the user experience.
“We always need to be back testing what it is that we are adding,” Gupta explains. “We look at things like cart conversion rate, which is, if we showed something versus not, what’s the actual impact on that? And making sure that we are only being net additive and not taking anything away from the experience.”
By maintaining this long-term view of customer relevance and trust, Rokt aims to create a win-win situation for both its partners and end customers.
You’re watching “Why Relevance Rules in Ecommerce, a Beet.TV Leadership Series presented by Rokt”. For more videos from this series, please visit this page.