Outcomes might be the most popular word in advertising right now.

Everyone wants to know what happened after the impression: Did someone visit the site? Request a quote? Walk into a store? Buy something? And further down the line, is brand affinity growing?

For Drew Paquette, Director, Audience & Outcomes at iSpot, the pressure to answer those questions extends well beyond the marketing team. Proving the value of media investment, he says, is as much a CFO conversation as a CMO conversation. And counting conversions only gets you so far.

That’s the question incrementality measurement aims to answer. But before media teams can credibly connect an action to advertising, they need to answer a more basic one: Who actually saw the ad?

A conversion, site visit or store visit tells you that an action occurred. It does not, on its own, tell you what caused it. Connecting that action to verified ad exposure is a starting point; determining whether advertising drove additional activity requires a sound methodology, too.

For cross-platform attribution, that work starts with a clean, deduplicated view of the audience: who was exposed, how often and where. Without it, the outcome story gets shaky, quickly.

Your Audience Doesn’t Live in a Media Plan

A person can see your ad during a linear broadcast at night, encounter it on a streaming app the next morning and receive another impression later that day. A fragmented measurement system can treat those exposures as separate audiences.

That’s why cross-platform outcome attribution needs a way to reconcile exposures across screens before assigning credit for what happened next. Otherwise, downstream reporting can become precise-looking math built on a bad audience count.

Paquette’s perspective adds another requirement: the methods behind those numbers need to be consistent. Combining reports in one place is only useful if the results can be meaningfully compared. 

Here are four ways a weak audience foundation can distort those comparisons and the decisions that follow.

1. You Think You’re Finding New People, But You May Be Finding the Same People Again

Say one household sees your campaign on linear TV and again on a CTV app. If those systems can’t recognize that it’s the same household, adding their reach figures together can count it twice. Reach looks stronger than it really is, while the frequency building across platforms disappears from view.

A planner looking at that report might keep spending because the campaign appears to be expanding its audience. In reality, it may be reaching the same households again.

Paquette explains that making those exposures comparable takes substantial work behind the scenes. Different publishers have different integrations and ways of supplying data. Some require clean rooms. Those inputs still have to come together in a unified measurement solution.

For a buyer, the useful question is how a provider reconciles those differences and deduplicates exposures across partners. A combined total alone does not tell you how much new audience each publisher contributed.

2. Everyone Gets to Claim the Conversion

Now take that same household and assume someone makes a purchase after exposure across several channels. If each platform measures itself in isolation, multiple platforms may claim credit for the same transaction.

You’re left with one purchase and multiple claims of credit. Add those reports together without reconciling them, and campaign performance can look stronger than the business result that actually occurred. The next planning cycle may then reward channels for results they did not independently generate.

This is where Paquette puts particular weight on neutrality. Marketers need to examine both how credit is assigned and whether the company assigning it has a financial interest in the media being measured.

An independent view gives marketers a basis for evaluating publishers together. Deduplication helps establish which exposures occurred; a consistent attribution methodology determines how to assign credit. And incrementality measurement asks whether those exposures drove activity beyond what would have happened anyway. 

Each addresses a different part of the outcome question.

3. You Connect Business Results to Media Proxies Instead of Actual Exposures

A household can watch a program but miss the commercial break. A program rating alone cannot establish that it received the brand’s ad, making it a weaker starting point for connecting exposure to a business action.

To explain the distinction, Paquette borrows a framework he credits to Auren Hoffman: “truth versus religion” in data. At one end are observed, immutable events, such as a transaction or ad exposure. At the other are models that interpret signals to describe audiences or explain performance across disparate channels. The more interpretation involved, the more the result depends on the assumptions behind it. And the more the buyer needs to have faith in that solution.

Modeling has a role. Marketers need to understand what was observed, what was inferred and how those inferences affect the result.

An ad-first approach starts with evidence that a household received the brand’s ad, including when and where. Deduplicating those exposures across publishers lets marketers evaluate a subsequent action against the household’s combined exposure history. That reduces reliance on assumptions about who saw the ad based on program viewing.

The next step is applying consistent methods to that evidence.

Using the same exposure definitions, attribution windows and crediting rules makes publisher comparisons more meaningful. Otherwise, one publisher could look more effective simply because it counts conversions over a longer period.

Together, verified exposure and consistent methods reduce uncertainty about who received the ad and how results are evaluated. Proving the ad drove an action that would not otherwise have happened still requires incrementality measurement.

4. By the Time You Untangle It All, the Campaign Is Over

There’s also the operational reality. 

Before a media team can compare results, it may have to reconcile different campaign definitions, line items and naming conventions across buying systems. Paquette describes that work as a substantial undertaking, especially for agencies managing multiple platforms, and jokes:

It lands because the work comes before the analysis. Teams have to standardize the inputs just to make measurement possible. Then they need the methodology and expertise to interpret what the data means.

When that process stretches beyond the campaign, its value becomes retrospective. You find the duplicated reach after paying for it. You uncover excessive frequency after the audience has experienced it. You identify an inefficient allocation after the budget is gone.

A centralized, deduplicated audience view reduces the reconciliation burden and gives teams a better opportunity to act while media dollars are still running.

Outcomes Are Only as Good as the Exposure Data Underneath Them

More conversions, transaction data, and web activity cannot repair a broken audience foundation. They give marketers more signals to interpret, but they do not resolve who was exposed or how credit should be assigned.

Paquette’s recommendation is to use a consistent approach across the funnel: from creative and audience measurement through outcomes, with a trusted partner that has no stake in which medium ultimately gets credit.

That gives media teams a clearer basis for seeing where frequency is building, identifying which publishers add incremental reach, and adjusting allocations while a campaign is still in-market. It also supports a more credible evaluation of business results against the audience actually exposed.

Without that foundation, you may still get an outcome number. The harder question is whether you can trust what it’s telling you.

iSpot’s approach to outcomes starts with unified audience exposure across screens, giving marketers a consistent foundation for connecting media to business results and making better investment decisions.