TV Buyers Need Better Signals Not Just More Floss Data: Alliant's Suvadip Choudhury

Why It Matters

Because precise, high‑quality audience signals directly boost ad efficiency, the blend of transaction and predictive data will determine which brands capture fragmented TV viewers and drive measurable growth.

Key Takeaways

• TV targeting now shifts from demographics to behavior-driven audiences
• Quality of behavioral signals outweighs sheer volume of data
• Transactional data excels for repeat purchases, but predictive insight fills gaps
• Predictive models outperform when past behavior data is sparse or unstable
• Future ad buying blends content adjacency with intent‑based audience targeting

Summary

In a recent interview, Alliant’s Suvadip Choudhury argued that TV buyers need better quality signals—not just more raw data—to sharpen audience targeting. He traced the industry’s evolution from simple demographic slices to behavior‑driven segments that can identify, for example, “moms likely to spend on childcare” rather than just “moms ages 25‑34.”

Choudhury emphasized that the value of a signal lies in its reliability and provenance. He warned that advertisers must ask where seed data originates, how it’s collected, and what it truly represents before building sophisticated audience strategies. Transactional data remains the gold standard for high‑frequency, repeat‑purchase categories, but it falls short for infrequent, high‑value actions such as home buying.

Using a floss‑purchase analogy, he illustrated how knowing a consumer’s repeat purchase tells only part of the story; predictive insights fill the gaps by interpreting why behavior changes. He also highlighted the tension between content adjacency—buying ads next to relevant programming—and intent‑based audience buying, concluding that both will coexist but require intuitive selection by marketers.

The takeaway for marketers is clear: combine robust transactional signals with predictive modeling to achieve both scale and precision. As viewing habits fragment across devices, advertisers who integrate high‑quality behavioral data will unlock more accurate ROI and stay ahead of the shifting TV landscape.

Original Description

SAN JUAN, PUERTO RICO – At the Beet Retreat San Juan, Suvadip Choudhury, head of television partnerships at Alliant, delivered a reality check to the TV ad business. Yes, we have more data than ever. No, that does not mean we should blindly chase every shiny behavioral segment that appears in a dashboard.

“The TV industry has done a great job at evolving the way they think about custom audiences,” Choudhury said in this interview with Beet.TV contributor David Kaplan.

He said the shift from basic demographic targeting such as “moms ages 25 to 34” to far more behaviorally informed audiences.

Instead of broad strokes, marketers can now zero in on “moms who are likely to spend on childcare” or “college educated households that are looking to buy a new house.”

With richer signals, he said, “we now have the ability to hone in on the exact things that an advertiser, an agency, or brand might want for targeting their consumers.”

But more signal also means more responsibility. Choudhury urged buyers and sellers to question the raw material behind those segments.

“Where does the seed data come from? How did you collect it? What does it truly signify?” he said, adding that these are the questions required to execute “highly sophisticated audience strategies in the future.”

Translation: not all data is created equal.

You are more than your floss purchases

Kaplan suggested that Alliant sits at the crossroads of transactions and prediction. Choudhury agreed, then cited an example that may haunt dental aisles everywhere.

“Transactional behavior has been the gold standard for E-commerce marketing,” he said.

If you buy floss every month, marketers know you floss. Congratulations.

But, he added, “there are multitudes to everyone.”

Just because someone buys a lot of floss does not mean that floss defines their entire identity.

“The frequency at which I buy tooth floss isn’t the only thing about me,” he said with a smile.

For marketers, the lesson is clear. Transaction data shows what happened. Predictive insight helps explain why and what might happen next.

“The context of why they’re doing the things that they do matter,” he said, because that broader understanding leads to better predictions and smarter audience expansion.

Balancing scale and precision in CTV

When it comes to connected television, the eternal tension between scale and precision still looms. Choudhury framed it as a business decision.

“Balancing scale and precision really comes down to how an advertiser would like their ad dollars to be used,” he said.

If a brand wants a very specific consumer based on real behavior, precision wins. If there is enough transaction data, that precision can sometimes scale on its own. But often it cannot.

That is where predictive modeling steps in. It helps “fill the gap and size up,” he said, allowing marketers to take a precise audience and reach it at meaningful volume.

After all, “there are more to users purchase beyond just the fact that they’re purchasing,” he said, emphasizing that behavior is layered and