Market segmentation in a dual speed economy

For decades, market segmentation practices prioritized behaviors and beliefs over demographics, assuming consumers had the agency to act on their preferences.

Demographics, we were told (often rightly), are crude proxies that don't predict purchase behavior. A 35-year-old in Sydney, Australia may have more in common with a 55-year-old in Berlin than with another 35-year-old in the next suburb.

Behavioral segmentation, attitudinal clustering, jobs-to-be-done frameworks—all assume that the why matters more than the who. But lately, I've been wondering: what if that assumption no longer holds?

What if something has shifted permanently underneath our segmentation maps—not cyclically, but structurally?

What if the economic structure has fractured so significantly that who someone is—their structural position in the economy—once again constrains what they can do, believe, or become?

Are we segmenting the wrong terrain?

A good segmentation model is like a map: it should show us the terrain, help us navigate trade-offs, and expose viable paths to value. But most segmentation maps today assume the economy is a single, connected system.

Data and lived experience increasingly reveal that economies like the U.S., Australia, the UK, and parts of the EU function as structurally stratified systems, where economic agency varies significantly across populations.

• A "fast lane" made up of asset-rich professionals, AI-augmented knowledge workers, and high-income households with capital leverage.
• A "slow lane" comprising precarious workers, renters, underemployed service workers, and debt-burdened graduates.
• And perhaps a "median strip" of emerging adaptive earners—freelancers, AI-native operators, career-switchers—who could fall either way.

This dual-gear economic structure suggests that people's capacity to act on their preferences or beliefs is now increasingly gated by structural realities.

That's the part that keeps sticking in my head: capacity to act on beliefs.

Beliefs are not enough

Two people may both believe in financial discipline. One has a mortgage offset account and a six-figure bonus. The other is choosing between rent, groceries, and paying down a Buy Now Pay Later balance.

Two founders may both believe in investing in AI. One has $4M Series A funding and a 12-month runway. The other bootstraps from a co-working space while juggling freelance contracts to make payroll.

They behave differently not because they believe different things, but because their constraints are different. That suggests that belief-based segmentation, while still useful, is insufficient on its own.

We need to layer it atop a structural understanding of economic agency—including how group identity now mediates economic behavior, making tribal signaling as important as utility optimization.

In other words, segmenting by belief alone is like assuming everyone who wants to travel has a passport and paid leave.

The case for a multi-step segmentation framework

I'm beginning to think we need a new approach—something like a two-layer segmentation model:

1. First: Structural Segmentation

Not demographics as cultural proxies ("males 25–44"), but demographics as structural position indicators: age as automation-risk exposure, geography as policy regime effects, education as economic resilience capacity. For example:

• Asset ownership vs. debt load
• Income volatility vs. salary stability
• Renters vs. owners
• Institutional vs. gig-employed
• Access to affordable credit vs. subprime exposure

This gives us a map of the economic terrain—where your customers are standing, and what constraints govern their decisions.

Note: It might be necessary to account for temporal positioning; people moving between economic speeds due to life events, market cycles, or technological disruption—but with recognition that structural barriers now make speed transitions less fluid than previously assumed.

2. Then: Belief and Behavioral Segmentation

Once we know the terrain, we can interpret beliefs and actions in context:

• Simplicity in the fast lane might mean "automated and premium."
• Simplicity in the slow lane might mean "predictable and low-risk."

Same stated preference, different operational meaning.

By sequencing the segmentation this way—structure first, beliefs second—we can design offers, messages, and experiences that are truly aligned to desire and reality.

Is this a return to the past?

Not really. This isn't about going back to crude demographic targeting. This is about recognizing that in a fractured economy, we can no longer pretend every customer has similar capacity to act, regardless of what they believe or desire.

We're not giving up on behavioral segmentation—we're grounding it. We're making it legible again.

And maybe that's the shift we need in strategy today: not more nuance, but the right kind of foundation to layer the nuance on top of.

Do we need a radical new approach?

If segmentation models were built for an economy that no longer exists, then tweaking them may be a false economy.

It might be time to admit that segmentation, as commonly practiced, has failed to evolve with the real world. Not because the frameworks are broken—but because the assumptions underneath them no longer hold.

What if the core question isn't "What kind of customer is this?" but "How free are they to act like the customer they want to be?"

In a structurally bifurcated economy, where purchasing power is unevenly distributed and economic volatility is asymmetric, the meaningful axis of segmentation may no longer be beliefs or behaviors, but degrees of constraint.

Think of it as economic agency mapping: understanding not just customer preferences, but the structural realities that determine whether those preferences can become purchasing decisions

The radical move, then, is to treat structural positioning—income predictability, capital access, geographic arbitrage—not as metadata, but as the first layer of segmentation. Only once we understand capacity, does it make sense to map intent.

It's a shift from classifying customers by what they say or do, to segmenting based on what they can do—under current and forecast conditions. In some ways, that's not a radical redefinition of marketing. It's a return to its core: deploying scarce resources where structural readiness meets meaningful demand.

Is this your reality too?

If you're seeing your customer behavior models lose predictive power, if you're watching spend patterns bifurcate in ways your personas don't explain, you're not alone.

Maybe it's time to stop segmenting markets just by what people want and start segmenting by what their economic position lets them do, when it lets them do it, and how group membership shapes what they're willing to do.

I'm exploring this in my consulting work and would love to hear from others wrestling with the same tension. Are we seeing a reemergence of structural segmentation? Or something entirely new?