Double-Sided Markets Break Most Audience Research. Here's the Fix.

Sometime last week, someone sat in a SparkToro Office Hours session and asked a question that almost nobody in a boardroom asks out loud: how do you do audience research when you have two completely different groups you're trying to reach at the same time? The question lived rent-free in Rand Fishkin's head long enough that he wrote about it. It should live rent-free in yours too.

Double-sided marketplaces are everywhere now. Gig platforms. B2B SaaS with a buyer side and a seller side. Retail aggregators. Specialty communities that monetize from both the contributor and the consumer. The structure is common. The research discipline to serve it is not.

The Single-Persona Trap

Most marketing teams, if we're being honest, collapse both sides of their market into one representative persona. They find the overlap, declare it a target audience, and build a campaign. This feels efficient. It is not. What they've actually done is averaged out two distinct tribes into a fictional third tribe that exists nowhere in the wild. The messaging lands cold on both sides. The acquisition funnel looks normal on the surface and then stalls at a number nobody can fully explain.

Each side of a marketplace has its own identity contract with your brand. The person supplying inventory — whether that's a freelancer, a seller, a content creator, or a service provider — comes to you with a set of status needs and permission thresholds that are simply different from the person consuming or buying. They read different publications. They trust different signals. They respond to different pretenses of value. A single research brief can't hold all of that without compressing something important.

Where the Arbitrage Window Opens

Here's what the research discipline actually demands. Two separate audience briefs. Not one brief with two sub-sections that share an introduction. Two fully independent tracks of inquiry, each asking: what does this cohort read, watch, and follow before they've ever heard of your category? What's the adjacent habit or ritual that already primes them for the problem you solve? Where do they go to feel status within their own peer group?

The supply side of a marketplace almost always has a stronger professional identity signal than the demand side. The demand side almost always has a stronger appetite signal tied to an outcome they want. If you map those two appetite structures separately, you end up with two precise media and channel maps. You stop guessing which ad placement is wasted. You start placing each piece of content exactly where one tribe is already paying attention.

This is the window. Your competitors running blended personas are spending budget on impressions that confuse one side or the other. You can step into that gap by simply doing the unglamorous work of maintaining two research tracks instead of one.

The Operator Move

Start with the harder side. In most double-sided markets, one side is more expensive to acquire and more critical to the product working. Identify which that is for your specific business. Build the audience research brief for that side first. Go deep on what media they trust, what language they use to describe their problem before they've found a solution, and what peer communities shape their professional habits. That brief becomes the foundation of a separate creative and channel strategy.

Then build the second brief for the other side with the same rigor. Resist the urge to reference the first brief while you're writing the second one. The contamination risk is real. You'll unconsciously look for connections between the two audiences and end up rebuilding the blended persona you were trying to escape.

When both briefs are complete, look for intersections deliberately rather than accidentally. Some channels will serve both cohorts. Some creators or publications will have real reach into both tribes. Those overlaps are genuinely valuable. But you can only see them clearly after you've mapped each side without the other one in the room.

Three Questions to Pressure-Test Your Marketplace Research

Could you describe the media diet of your supply side without referencing your demand side at all? If the answer is no, your research is blended and your targeting is probably blended too. Ask which side of the market, if it disappeared tomorrow, would make the product immediately worthless — and check whether your research budget is proportional to that answer. If you audited the last six months of content your team produced, would a stranger be able to identify which pieces were written for which tribe, or do most pieces seem addressed to a general audience that doesn't quite exist?

The brands that build real liquidity in a two-sided market are the ones that resist the elegant simplicity of a single audience story. Messy. More work. Considerably harder to present in a quarterly business review. That difficulty is, in fact, the whole point.