The ad tech industry loves to talk about the agentic future. I’d rather build it.
I’ve sat in a lot of these agentic AI conversations lately. And my honest reaction most of the time is: we’re describing a DSP and an SSP. We’ve just changed the name.
The vision gets laid out — buying agents talk to selling agents, the whole auction infrastructure gets reimagined — and I’m sitting there thinking: okay, so who builds the buying agents? Companies that specialize in it. Who builds the selling agents? Same answer. That sure sounds like what we already have. Different code, same concept. And more importantly: better for whom, exactly?
So I keep coming back to a more useful question: what problem are we actually trying to solve, and for whom?
Here’s my current answer for the buy side, as well as some observations. Most mid-market agencies are running campaigns across three, four, or five buying platforms simultaneously. Setting up a campaign correctly takes two hours per platform. Multiply that out, and a single campaign brief can eat an entire trading day before anyone has made a single strategic decision. Senior traders are doing data entry. Junior traders are making guesses on campaign setup. Mistakes are made. Nobody is doing the work they were actually hired to do, or all the work required to effectively set up and manage a campaign.
The last paragraph articulates a specific, solvable problem, one that QuantumPath was built to fix.
The major buying platforms aren’t going away. And when each of them builds its own agentic layer (and they will!), the ecosystem won’t be simpler. It’ll just be fragmented in a new form. Five walled-garden AI agents instead of five browser tabs, each with its own blind spots and its own agenda.
The answer to platform complexity isn’t fewer platforms. It’s smarter orchestration across the ones that already exist.
QuantumPath is that layer. It sits above existing buying platform seats — TTD, DV360, Amazon — normalizing setup, enforcing accuracy around the clock, and surfacing cross-platform intelligence that no individual buying platform can provide on its own, using agents to simplify the work. And critically, zero displacement. No new contracts. No workflow upheaval. It sits across the top of what agencies already have and makes it dramatically more intelligent.
In a blinded case study from January 2026, an agency using the QuantumPath platform hit a 4.6x increase in campaigns managed per trader and a 58% reduction in execution errors — without adding a single headcount. Traders who used to lie awake worrying about a misconfigured campaign walk in Monday morning with a single cross-platform dashboard that scores every active campaign and tells them exactly where to focus.
That’s not theoretical. That’s traders finally having the right tool.
The platform is led by GM of Advertising Solutions, Jeff Hirsch. Jeff was the former CCO of PubMatic and CEO of Audience Science. QP was designed by Zach Lenchner, who built and ran the only compliant programmatic digital offering for Canada’s national election campaign. Ken Rona, our CAIO and co-founder of both Chalice and SWYM, rounds out the QP leadership team, responsible for integrating QP with our existing demand solutions and the ongoing build-out. These are people who have lived the problem they’re solving. That matters.
2026 is the year agentic AI moves from promise to production on the demand side. We’re not waiting around for some theoretical future to announce itself. We’re building on top of the infrastructure agencies already rely on and making it smarter. That’s the right problem to solve.
And we’re excited to get to work.
John Nardone is CEO of JWX. Learn more at www.jwx.com.