By Charlie Swift, General Manager, Adstra Services

Most lead generation strategies are built on a flawed premise — that the path to better performance is a smaller, cleaner, more refined audience. Marketers layer on criteria, tighten parameters, and chase the “perfect” prospect. Every time they do, the pool shrinks. Cost per lead goes up, volume drops, and the program quietly starts losing its long term maximum ROI potential.

What’s missing with this type of approach is the opportunity to expand an audience. While it might make sense to refine a large pool to find the right leads, too much refinement typically strips away the ability to learn and identify new audiences. 

What can a new approach look like? It starts with expanding to new like-minded audiences, cultivating them over time, and letting the data tell you what’s working. 

Start With the Decision

Before touching audience strategy, some questions need to be answered. First: what specific decision is this program trying to make and what data will actually move it forward? Second: what is that decision worth in dollars? Not in leads, not in impressions, but in revenue. Those two answers are the foundation. 

Most lead gen programs jump straight to the channel, the list, the creative — and optimize for cost per lead without ever asking whether the leads being generated are worth the cost. It feels like progress. The dashboards look active. But optimizing toward the wrong outcome faster is still the wrong outcome.

The result is a program that gets more efficient at generating leads that never close, or customers who churn before they’ve returned their acquisition cost. How many of us have seen the results of chasing “cost per click” metrics with no customers to show at the end? The decision framework (defining what you’re trying to decide and what it’s worth) is where most programs fall apart. It’s also where the real performance gains live. Get this right first, and everything downstream gets sharper: the audience strategy, the testing criteria, the definition of success.

The Performance Marketing Curve

Once you know what a decision is worth, you ask another useful question: where is your audience strategy actually operating and where might money be left behind? 

Think about performance marketing as a curve. On one end sit the proven audiences, high conversion rates, predictable ROI, and a ceiling that most programs hit faster than they expect. On the other end sits the unknown: audiences that haven’t raised their hand yet, that require cultivation before they convert, and that most programs never pursue because the short-term numbers look uncertain.

But that’s exactly where the growth is. The brands that expand their performance curve are the ones willing to invest in the unknown with an effective testing framework, clear decision criteria, and the data infrastructure to measure what’s actually happening. Over time, yesterday’s unknown becomes tomorrow’s proven audience. The curve extends, the economics improve, and the program stops recycling the same shrinking pool

Now let’s get into it. How can marketers expand their Audience Framework to capture more leads and customers? 

• Anchor to your proven audience: Know who your best converters are. Build the model, and understand the attributes that predict performance.

• Identify the like-minded adjacent: Use identity resolution and behavioral data to find audiences that share key attributes but haven’t yet engaged. Consider attributes of your proven audience before engagement, not just what they look like now.

• Test with a clear decision value:  Before investing, define what a successful result looks like in dollars. What conversion rate justifies the spend? What LTV makes the acquisition cost worth it? Think positive return, not bettering proven performance.

• Let the data decide: Follow the performance, not the legacy assumption or the comfortable vendor relationship.
• Cultivate over time: New audiences don’t convert at the same rate on day one, and you may need new messaging.  Measure long-term value, not just first-touch response.

Knowing where to look is half the battle. The other half is having the infrastructure to actually get there.

AI: Accelerant, Not Strategy

We can’t discuss anything these days without exploring how AI fits in and audience strategy is no exception. The good news is that AI enables faster, cheaper, and better testing than was ever possible before. What used to require weeks of manual analysis can now happen in near real time. That speed changes the economics of audience expansion entirely.

But AI doesn’t define what a conversion is worth. It doesn’t distinguish between a subscriber who churns in 90 days and one who stays for three years. That context — the business logic, the lifetime value model, the definition of a quality outcome — has to come from the program. The strategy still has to be owned.

Where AI earns its place is in the middle layer: pattern recognition at scale, lookalike modeling across large identity graphs, predictive scoring, and dynamic optimization based on real-time performance signals. Programs building that loop now will have a structural advantage over those that wait and the gap is already widening.

What is one constraint AI doesn’t remove? It doesn’t make bad data good. If the identity foundation is fragmented or stale, AI will optimize confidently toward the wrong outcome. The quality of the data layer underneath determines whether the acceleration is pointed in the right direction.

Identity Resolution: The Foundation Underneath

The second enabler is identity resolution. Expanding to new audiences requires the ability to reliably identify who those people are and connect them back to the performance signals that matter. Without a stable identity spine, audience expansion becomes guesswork. You’re reaching people you can’t measure, testing segments you can’t track, and optimizing against signals that don’t connect back to real outcomes.

Identity resolution is what makes the loop close. It’s what allows a program to say: this person saw this message, took this action, and became this kind of customer. Without that connective tissue, AI has nothing real to learn from, and the performance curve never actually extends; it just looks like it does until the numbers catch up.

Audience expansion is the strategy. Revenue is the result. Those have to move together and identity resolution is the infrastructure that keeps them connected.

The Question Worth Asking

The decision framework, the curve, the AI loop, the identity foundation–– all point to the same reorientation. Start asking “how do I make money from a bigger audience?” The conversations worth having are about defining decision value before spending a dollar, building the data infrastructure to test new audiences without blowing up cost structure, having the resources to understand performance, and cultivating those audiences so they become the next wave of proven performers.

Programs that respond by filtering harder are accelerating their own decline. Audience expansion programs that respond by expanding smarter, testing with discipline, and letting results drive decisions are the ones that will still be growing two years from now.