In this episode, we interviewed Tejas Manohar - co-founder and co-CEO of Hightouch, the leading agentic marketing and customer data platform trusted by brands like Domino's, Autotrader, Cars.com, Aritzia, and PetSmart
Tejas was an early engineer at Segment, the CDP leader acquired by Twilio for $3.2B - he pioneered the Composable CDP category with his co-founders, and is now focused on bridging the gap from CDP to the future of agentic marketing
His core argument: the model is the easy part now - the brand and customer context you feed it is what separates on-brand work you'd put in market from generic AI slop, and the action layer is where AI actually changes marketing
Key takeaways:
The warehouse became the CDP - instead of copying data back out into another silo, Hightouch solved the CDP on top of Snowflake and Databricks and pioneered the Composable CDP
Context is the thing leaders underrate - the model is the easy part, and brand and customer knowledge is what turns generic output into something worth putting your logo on
AI changes the action layer, not the single view - instead of a person maintaining 47 segments, the system decides the right message for each individual and learns from every outcome
Marketers will be judged by how well they manage agents - within a year the role shifts from making the assets to directing the agents that make them
🔗 Connect with Tejas
Who are you and what do you do?
I'm the co-founder and co-CEO of Hightouch, which is the leading agentic marketing and customer data platform trusted by leading brands like Domino's, Autotrader, Cars.com, Aritzia, and PetSmart. We're reinventing how marketing teams work with AI.
You went from engineering at Segment to co-founding and co-leading Hightouch - what problem did you see that everyone else was missing?
At Segment, I saw that traditional CDPs were too slow, fragmented, and disconnected from the data companies rely on to grow. Every company was pouring money into their data warehouse, like Snowflake or Databricks, and then paying a CDP to copy all that data back out into yet another silo. I realized that many of the challenges of building a best-in-class CDP would be better solved on top of the data warehouse.
With my co-founders, Kashish Gupta and Josh Curl, we founded Hightouch and pioneered the Composable CDP category. Today, we're focused on bridging the gap from CDP to the future of agentic marketing.
Every company was pouring money into their data warehouse, and then paying a CDP to copy all that data back out into yet another silo.
Walk us through one concrete way Hightouch changes the day-to-day for a marketing team - a real workflow, not the pitch
I'll give you an example of our customer Otrium, a fashion marketplace running end-of-season sales for 300+ brands across a dozen European markets.
For most performance marketing teams, like Otrium's, the process to create ads for paid platforms like Meta is extremely time-intensive and usually takes weeks. That means it's, first of all, difficult to beat creative fatigue with the amount of creative these platforms require, and second of all, difficult to react and respond to new ideas quickly, and get those campaigns into market when the ideas are still relevant.
They picked up Ad Studio, which plugs into their ad accounts, brand assets, product catalog, and performance history. Now they generate 500 on-brand concepts in a single session, tweak copy or swap a product, and launch in about a week, instead of four.
What's the most common thing senior leaders get wrong about AI?
Foundation models are incredible and getting better every month, but a general AI model doesn't know the intricacies of your brand, your customers, or what worked for you last quarter. Point those models at your marketing, and it'll confidently produce a lot of on-average, off-brand mediocrity (some call it AI slop).
The thing leaders underrate is context. The model is the easy part now. The brand and customer knowledge you feed it is what turns a generic output into something you'd actually put in market with your logo on it.
The role stops being 'the marketer who makes the assets' and becomes 'the marketer who directs the agents that make them.
What's in your AI stack? The one tool you rely on every week?
Claude Code. I came up as an engineer, and the workflow I lean on every week is delegating real work to a coding agent. It changed how I work faster than anything else, and it's exactly the shift I think every marketer is about to go through: you stop authoring every artifact by hand and start managing agents that produce the work. Living that shift myself is a big part of why I'm convinced it's coming for marketing.
Marketers have been promised a single view of the customer for a decade - what does AI decisioning actually change, and what's still just hype?
For a decade, the pitch was that if you unify all your customer data into one view, personalization follows. Everyone spent years and investments on unification, and then still hand-built segments and if-this-then-that rules on top of it. So the single view mostly just sat there. I don't think it fully held up to the promise of making marketing more personal.
What AI Decisioning changes is the action layer. Instead of a person maintaining 47 segments, the system decides the right message for each individual and learns from every outcome. The single view of the customer was never the point - acting on it, one person at a time, at scale, was.
The hype is thinking a clever model alone gets you there. Without your data and brand context grounding it, you just automate generic outcomes faster.
Where is your field in 12 months - one specific prediction?
Within a year, the best marketing teams will be judged by how well they manage agents, not how much they personally produce.
I'd bet that at the leading consumer brands, most campaign variants shipped twelve months from now are generated and decided by agents, with people setting strategy, guardrails, and brand rather than hand-building each one. The role stops being "the marketer who makes the assets" and becomes "the marketer who directs the agents that make them."
Where should readers find you, and what should they read first?
I post on LinkedIn the most.
I post fairly regularly about where marketing and AI are heading, usually grounded in what we're seeing with customers. Start with our piece on the Agentic CDP. It's the clearest version of where I think customer data and marketing actually go from here
AI Central Voices is where the AI Central team sits down with the founders, executives, and builders shaping AI - going behind the scenes of how they operate, what they're betting on, and where the industry goes next.
Want to be featured, or have an event you'd like us to cover? Reach out at [email protected]
