The hidden power of latent data with Justin Evans

BY LIZ MOOREHEAD, BEELER.TECH

Justin Evans opened his Navigator NYC session by talking directly to the operations people in the room. If you work in ops, or manage the people who do, you’re close to the work in a way most other parts of the company aren’t.

You know how campaigns actually deliver. You know what clients ask for. You know where the reporting breaks down, where the performance questions show up, and which data points everyone keeps asking about without knowing what they really mean.

That puts operations in a powerful position. Not just to report on what happened, but to help publishers find new value in the data they already have.

Justin’s talk, “The Hidden Power of Latent Data,” made the case that data and AI aren’t really about technology. They’re about ideas. The opportunity isn’t to chase every new tool or drown the business in another dashboard. The opportunity is to ask better questions, find the data assets already sitting inside the company, and turn them into something useful.

The first problem is intimidation

Data has a way of making smart people feel like they’re standing outside a locked room. The language can be arcane. The AI conversation makes that worse. Everyone feels behind, and the pace of change can turn curiosity into fear.

Justin’s view is that intimidation is the enemy of innovation. When people treat data knowledge as power over others, the work gets smaller. The jargon gets louder. The room fills up with people proving they know the terms instead of asking what the business actually needs.

That’s why he encouraged the audience to cut out the noise. You don’t need to become the most technical person in the company to make better use of data. You need to understand the business problem well enough to ask the question that matters.

Better data starts with a better question

One of the strongest examples Justin shared came from his time at Samsung Ads during the pandemic. Streaming was exploding, advertisers were nervous, and agencies wanted to understand what was happening to their audiences. At first, the question seemed obvious: how much more streaming TV are people watching?

So his team answered that question. Again and again. They created reports for advertiser after advertiser showing how audience behavior had shifted from linear TV to streaming. The work was technically useful, but it wasn’t changing buying behavior.

Eventually, Justin realized they were answering the wrong question. Advertisers didn’t only need to know that streaming was growing. They needed to feel comfortable spending money in an environment they didn’t fully understand.

That changed the work. Samsung Ads created reporting that showed advertisers what their target audiences were actually watching in streaming environments, giving them the context they needed to trust the opportunity. The result wasn’t just more data. It was more confidence, and that confidence helped unlock revenue.

For publishers, that distinction matters. A report that answers the question someone asked may still miss the question underneath it. The real job is to understand what a client, executive, or team needs to decide, believe, buy, change, or trust.

Latent data is already inside the business

Justin used the term “latent data” to describe data that already exists inside an organization but hasn’t been put to work. Every publisher has some version of this. It may be clean and structured. It may be buried in systems. It may be sitting in plain sight, treated as operational exhaust instead of a strategic asset.

He organized latent data into six categories: traits, traffic, transactions, tokens, tastes, and treasuries.

Traits are the characteristics you know about audiences or customers. Traffic is the movement and volume you can observe across your properties. Transactions show what people buy, what they pay, and how pricing behaves. Tokens are the IDs that connect accounts, users, articles, stores, or other valuable objects. Tastes show up in what people search, read, watch, save, or revisit. Treasuries are the large stores of valuable information publishers often forget they have, like years of articles, images, research, reviews, or archives.

The point of the framework is simple: most companies have more usable data than they think. The challenge is that teams get so used to seeing the business one way that they stop noticing what else is there.

The creative step doesn’t have to be dramatic

Justin isn’t asking publishers to blow up their business model or build something wildly new from scratch. He was talking about a lateral step: take something the organization already knows or already has, then use it in a new way.

Pinterest Predicts was one of his examples. At one level, it’s search data. People search Pinterest for ideas, styles, products, colors, themes, and odd little signals of future interest. Pinterest turns that behavior into trend insight for marketers, then uses those insights to create audiences and commercial opportunities.

That’s the move publishers should pay attention to. Search behavior can become trend intelligence. Content archives can become audience insight. Campaign performance can become planning guidance. Repeated client questions can become product strategy. Delivery patterns can show where demand, pricing, or inventory strategy should change.

The data may already be there. The product may come from packaging it differently.

The data doesn’t need to be perfect

It only needs to be better than the next best alternative. That’s an important standard for publishers because incomplete data can become an excuse to do nothing. The dataset has gaps. The taxonomy is messy. The sample size isn’t ideal. The methodology needs caveats.

All of that may be true. It may still be valuable.

Business decisions already happen with imperfect information. If a publisher can give a client or internal leader a better view than they had yesterday, and can be honest about what the data does and doesn’t show, that can create real value.

Operations can help lead this work

The people closest to campaigns, clients, delivery, and performance often understand the business reality better than anyone else. They know where the friction is. They know which questions keep coming back. They know which data gets ignored.

That gives operations a larger role to play than simply producing reports or maintaining dashboards. Ops can help the business see what it already has, ask sharper questions, and turn latent data into something that solves a real problem.

For publishers, that may be the bigger takeaway. The next growth opportunity may not require a new system, a new data source, or another AI panic cycle. It may start with someone in the business looking at the data already there and asking: what could this help us solve?

👉Want to stay in the loop with what’s happening in the Beeler.Tech community? Subscribe to our newsletter. If you’re interested in attending or sponsoring a future event, you can explore our upcoming events here.