If the advertising industry were a family dinner, CTV would be the teenager screaming for attention while the adults pretend they “didn’t hear that.” Meanwhile, billions of dollars pour into streaming environments being held together by DSPs built when Angry Birds was the height of mobile innovation.

Pontiac’s Keith Gooberman sat down on The ADOTAT Show and, without flinching, said the thing legacy DSP architects would rather eat fluorescent office carpet than admit: “We don’t have cookies. That’s a very large difference in terms of our infrastructure costs and the way our DSP works compared to others.”

Translation: Pontiac is not dragging a rotting, decade-old cookie infrastructure carcass behind it like a dead whale on a beach. Everyone else is.

And in CTV, that difference is the entire story.

Let’s break the industry taboo here. The DSPs running your CTV dollars were engineered for banner ads, Y2K-era tracking scripts, and the fever dream of retargeting every warm body with a browser. They were never designed to handle streaming television with fixed floors, curated inventory, and publishers who guard their supply like it is the Dead Sea Scrolls.

Yet everyone pretends this is fine.
It is not fine.
It is a fire hazard disguised as an ad stack.

Keith laid out what those legacy DSPs don’t want you to look at too closely:
“Cookies make the core bidder very expensive. You need enormous computing power to maintain cookie maps and segment lookups. Every bid costs a DSP a sizable amount of computing power.”

Think about that.
Every time a big DSP fires a bid, a cloud server sighs, clutches its chest, and wonders why it wasn’t born an S3 bucket instead.

This is why your CTV results often feel like someone threw your budget into a leaf blower and prayed.

CTV was never meant to be propped up by a technology stack that thinks “impression number one” is a chance to do identity syncing homework and “impression number two” is the first opportunity to actually bid.

Imagine explaining this workflow to Hulu.
They would stare at you like you just tried to pay them in Monopoly money.

CTV is a different species. It is a lion, not a golden retriever. Yet legacy DSPs keep showing up with tennis balls and good intentions.

Keith said the quiet part with the tone of someone who has explained this to too many people who nodded politely while understanding nothing:
“Other DSPs are built for display. They want to get the CPMs down. They’re designed to lower CPM until they don’t win enough inventory. That doesn’t work on CTV.”

Of course it doesn’t.

Because in CTV:

A general-purpose DSP trying to run CTV is like watching someone try to cut down a tree using a Roomba. Enthusiasm is not enough. The tool is wrong.

Pontiac’s entire argument is simple and almost insulting in its obviousness:
If CTV is different, build a DSP that is different.

Keith explained the design choice with the conviction of someone who knows he is about to annoy half the industry:
“We don’t have cookies. That’s a very large difference.”

The absence of cookies is not a privacy pitch.
It is an architectural jailbreak.

What Pontiac skipped:

What Pontiac built instead:

Pontiac is not a prettier banner-DSP.
It is a different species entirely.

And if you’ve ever run CTV through a legacy DSP and wondered why the results felt like a breakup text, this is why.

Because 2025 is the year CTV stops pretending it’s “emerging” and starts fighting for territory like a street brawler with nothing to lose.

Streaming viewership is exploding.
Retail media is linking arms with CTV like a power couple looking for tax advantages.
Supply is consolidating.
Curation is turning into a gold rush.
Publishers are fortifying their walls.
Marketers want transparency, performance, and actual control.

And in back rooms, everyone whispers the same sentence like a confession:
“Why does CTV still feel like 2015 programmatic with three times the cost?”

The answer is almost depressing:
Because the tools running CTV are still built like 2015 programmatic.

Keith called legacy DSPs “battleships.”
Perfect analogy.
Big.
Impressive.
Powerful.
And absolutely terrible at tight turns in shallow water.

Pontiac believes the future belongs to DSPs that weren’t born in the cookie era.
DSPs that weren’t designed for retargeting.
DSPs that do not need identity-plumbing just to look at an impression.

And honestly, Pontiac might be right.
Someone had to build the first real CTV-native DSP.
It appears one already has.

This free section is your appetizer.
ADOTAT+ gets the wiring, the blueprints, and the parts of the story the incumbents would desperately prefer you not read.

Stay Bold. Stay Curious.
Know More Than You Did Yesterday

The free version gave you the polite handshake.
ADOTAT+ is where I actually rip the wiring out of the DSPs and show you how the machine works.

Here’s what’s behind the paywall:

• The full CTV war map
Who’s rising, who’s collapsing, which DSPs won’t survive 2026, and the publisher politics nobody speaks about in public.

• The real bidder diagrams
Stateless vs stateful, allocation loops, pacing windows, pod rules. The engineering truth the vendors avoid like a dentist appointment.

• The ugly stuff
The floor-sniping games, the PMP tricks, the failures during live sports, the fee structures everyone denies exist.

• The buyer survival playbook
How to stop wasting CTV budgets, how to build a real PMP strategy, and how to test DSPs without getting snowed.

If the free section felt sharp, the ADOTAT+ version is the part that gets forwarded in group chats with “don’t share this.”

Join today unless you like being the last person in the room to understand where CTV is actually going.

Stay Bold, Stay Curious, and Know More than You Did Yesterday.