Data-Driven Thinking

We now have plenty of evidence that AI can be an incredibly useful tool. But proof is also piling up that it erodes our creativity.

Machines can now flag anomalies, surface trends and spin insights in seconds. Yet, even with its speed, AI still can’t decode the “why” behind the “what.”

If you want to protect your margins, your data and your long-term strategy, here are five questions worth asking before you sign anything.

The play “Data” sparks conversations about AI, surveillance, data tracking and predictive modeling. The ad tech world should be paying attention.

As we head further into 2026, the biggest risk for marketing orgs isn’t choosing the wrong AI – it’s letting “AI perfectionism” delay decisions so long that no decision gets made at all.

Should businesses start advertising on AI media platforms? Yes. But it depends on how consumers in your category are already using AI and how quickly your organization can respond.

Customers don’t need perfection. But they do need to know that a brand will make it right when something goes wrong.

If a buyer wants real control, “domain rationalization” alone will never be enough. Buyers need “bid rationalization,” shaping the supply at the bid-request level so they see fewer, better, more outcome-relevant opportunities.

Many organizations are encountering a paradox: Performance appears strong, while durable growth remains elusive. How should performance be evaluated when the strongest signals no longer align cleanly with growth?

The advertising industry is full of noise about AI making buy-side and sell-side processes more efficient. That framing is convenient, but it misses a broader point, writes Snowflake’s Dennis Buchheim.

The US government wants to use digital ad infrastructure for the exact type of surveillance the industry’s critics have long warned about. We should have seen this coming.

Yes/no binary logic was the right tool for programmatic. But autonomous AI agents can evaluate 10 million impressions per second, combining data from hundreds of sources, and human-defined targeting logic can no longer keep pace.

The future of retail media depends on a shift away from siloed in-store and online metrics to a full-funnel approach that reflects how customers actually shop.

ICE wants to know how ad tech and location data could aid investigations. The question is: Will ad tech finally draw a line – or cross one?

Brands allocating a sizable chunk of their marketing budgets outside of Google and Meta achieve the best customer acquisition costs, proving that diversification is a performance strategy.

For brands, this opens a new lever beyond impressions. For publishers, this is a business problem, as their first-party data and on-site inventory could become less valuable for targeting.

Ripping out battle-tested infrastructure and starting from scratch – as some emerging agentic protocols propose – is the slowest and most painful path to advertising’s agentic future.

The brands seeing the greatest lift from Andromeda have adjusted not by outsourcing judgment to the algorithm but by refining their creative strategy. Here are three best practices to keep in mind.

Advertising’s “agentic” future is a compelling vision, but one that relies on capabilities that do not exist yet. AdCP may one day make buying easier, but it will not make buying better by itself.

For most of the last decade, privacy compliance lived in a gray zone. Companies could point to a cookie banner, update a policy and reasonably believe they were doing enough. Not anymore.

argue the Outcomes Era is already fading, writes Joe Zappa. But the evidence suggests the opposite. The Outcomes Era isn’t receding; it’s accelerating.

Bold, accountable leaders who publicly embrace conflict and challenge industry fraud are essential to spark innovation in digital advertising and ad tech, writes Alessandro De Zanche.

Don’t ask whether any other company has created a similar product to yours; the answer is probably yes. But there are other ways to stand out.

AI shopping agents are breaking the traditional attribution loop. Marketers must ensure that AI is aware of their brand and track how that awareness influences ecommerce outcomes.

Almost everything in AI feels big. But are you looking at the next Gangnam Style or the birth of a new industry? Here’s how to assess whether an AI vendor is likely to matter in two years.

For years, MFA was a mostly web-based problem. Now, generative AI has supercharged the made-for-advertising model, and it’s infecting social media feeds and vertical video platforms.

Broadsign may actually be building a platform that will make an attractive acquisition target down the road. And one of the major cross-platform Big Tech players feels like the most likely buyer.

CTV spending is flattening, performance is plateauing and buyers are hesitant to push budgets further. The reason is not complicated. When buyers cannot see what they are buying, they cannot commit their spend with conviction.

Generative AI is the new ad land obsession: a shiny promise of a fully autonomous world of self-driving advertising.  But behind the hype lies the costly delusion that automation can replace judgment and more content somehow means better marketing. We’ve entered the age of machine-made abundance, where content can be generated faster than it can […]

With privacy under unprecedented attack by data brokers and social media, it is the wrong time to weaken CIPA’s private right of action protections, as has been proposed in California Senate Bill 690.