Audience Is
the First Decision

The First Decision

When marketers talk about improving direct mail performance, the conversation usually starts with creative.

Should we test a larger envelope?
Should we personalize the package?
Should we redesign the offer?

Those are worthwhile questions.

They just aren't the first questions.

In my experience, the first decision is almost always the most important one:

Who are we mailing?

Everything else comes after that.

I've seen beautifully designed mail packages fail because they were sent to the wrong people.

I've also seen surprisingly simple creative outperform premium packages because the audience was exactly right.

That's one of the reasons I believe audience selection is the single biggest driver of direct mail performance.

Creative matters.
Offer matters.
Timing matters.

But every one of those decisions is amplified—or limited—by the audience you choose.

The strongest direct mail programs don't begin with design.

They begin with understanding who is most likely to become a customer.

Look Beyond Demographics

Years ago, marketers relied heavily on demographic variables like age, income, homeownership, or geography.

Those characteristics still have value.

But modern audience strategy goes much deeper.

Today we have the ability to identify prospects based on how they behave—not simply who they are.

Lookalike Modeling

One of the most valuable tools is lookalike modeling.

Instead of finding people who share basic demographic characteristics, lookalike models identify consumers whose overall profile closely resembles your best existing customers. The goal isn't simply to find more homeowners or more families. The goal is to find more people who look and behave like customers you've already proven you can acquire.

Transactional Modeling

Another powerful approach is transactional modeling.

Purchase behavior often tells us more than demographics ever will.

By analyzing what your existing customers spend money on, how frequently they purchase, and the categories they buy within, we can identify households with similar purchasing behaviors.

We're no longer asking:

"Who looks similar?"

We're asking:

"Who behaves like our best customers?"

Those are very different questions—and they often produce very different audiences.

There Isn’t One Perfect Data Source

One of the questions we hear most often is:

"What's the best data source?"

It's a reasonable question.

But I think it starts from the wrong assumption.

In most cases, there isn't one perfect data source.

There are simply different ways of identifying potential customers.

Every provider has strengths.

Some excel at transactional modeling.

Some have exceptional lookalike capabilities.

Others continuously refresh consumer purchase behavior or specialize in specific industries.

Each sees the market through a different lens.

If your acquisition strategy depends on a single data provider, you're limiting your program to one company's perspective.

That might simplify list management.

It might even reduce upfront costs.

But it also limits your opportunities to learn, improve, and scale.

Build a Multi-Source Audience Strategy

At Peak Response, we believe the strongest direct mail programs are built from multiple data sources and multiple modeling approaches.

Think of it like managing an investment portfolio.

Most investors wouldn't put every dollar into one stock simply because it performed well last year.

The same principle applies to audience strategy.

Different providers uncover different opportunities.

Different models identify different customer behaviors.

Some audiences respond at a higher rate.

Others produce higher-quality customers.

Some provide better long-term scale.

By testing multiple sources within the same campaign, you gain a much clearer understanding of where your next customer is coming from—and which models deserve additional investment.

Be Loyal to Performance

We don't believe in loyalty to data providers. We believe in loyalty to performance.

If another model identifies better customers, lower acquisition costs, or greater scale, we'll test it.

Our responsibility isn't to defend a particular data source.

It's to help our clients build the strongest possible direct mail program.

Connect Audience and Postal Strategy

Finding the right audience is only part of the equation.

How those names are distributed geographically can have a meaningful impact on campaign economics.

Postage has become one of the largest expenses in direct mail.

Too often, audience selection and postal optimization happen independently.

We think they should happen together.

For example, predictive models may identify qualified prospects across thousands of ZIP Codes.

Rather than mailing every qualified name equally, we often evaluate ZIP density to determine whether shifting volume into denser postal areas can reduce postage costs while maintaining audience quality.

In some cases, we'll remove names from sparsely populated ZIP Codes and replace them with equally qualified prospects in denser areas.

The predictive quality of the audience remains intact.

The economics improve.

These aren't the kinds of decisions most people associate with audience strategy.

But they should be.

Every Audience Is a Hypothesis

One of the reasons I enjoy direct mail is that it forces marketers to think scientifically.

Every audience you mail represents a hypothesis.

You believe one group of consumers is more likely to respond than another.

The mailing simply tells you whether that assumption was correct.

That's why I don't think audience selection is something you finish before the campaign begins.

It's something you improve after every campaign.

Every mailing should make your next audience decision a little smarter.

Scale Comes From the Right People

One of the biggest misconceptions in direct mail is that scale simply means increasing volume.

I don't think that's true.

Scale comes from finding more of the right people.

That requires better modeling.

Better testing.

Better measurement.

And better decisions.

As programs mature, we don't simply ask where we can mail more.

We ask where we can find more customers who resemble the ones already succeeding.

That's the difference between growing volume and growing profitably.

Start With Your Customers

If I had one hour with a marketing team preparing to launch a direct mail program, I wouldn't spend most of that time talking about creative.

I'd spend it talking about customers.

Who they are.
How they behave.
Where they live.
How they purchase.
What models identify them most effectively.
How we can reach them more efficiently.

Because once those questions are answered well, everything else becomes easier.

Creative becomes more effective.
Testing becomes more meaningful.
Postal strategy becomes more efficient.
Analytics become more actionable.

In my experience, the strongest direct mail programs aren't built around one great mail piece.

They're built around consistently making better audience decisions.

And that first decision influences every one that follows.

Questions Before the Next Mailing

Before recommending an audience, we'd ask:

• Who are your best customers today, and what makes them your best customers?
• What purchase behaviors distinguish them from the broader market?
• Should we test lookalike, transactional, demographic, or predictive models—or all of the above?
• Would a multi-source data strategy improve scale and learning?
• Can ZIP density optimization reduce postage costs without sacrificing audience quality?
• What audience hypothesis are we trying to validate with this mailing?
• If this audience succeeds, how will we responsibly scale it?

Because in direct mail, the first decision is often the most important one.

About the Author

Patrick Carroll is Co-Founder and Strategist at Peak Response. Over the past fifteen years, he has worked with more than 100 brands and helped launch, improve, and scale some of the industry's fastest-growing direct mail programs. He believes better strategy leads to better informed decisions—and better direct mail.