Latest / The DTC Marketing Podcast with Fexingo: Direct-to-Consumer Brands, Shopify, and E-Commerce Growth / How DTC Brands Use Data Clean Rooms for Ad Targeting
Transcript
- Lucas: So a DTC mattress brand — let's call them SleepWell — was spending about $150,000 a month on Meta and Google ads, and their cost per acquisition had climbed thirty percent year over year. The usual fixes — better creative, tighter audiences — weren't cutting it. Luna: That's the signal-loss story we keep hearing. iOS changes, cookie deprecation, all of it. Lucas: Exactly. But SleepWell did something that's still fairly rare among mid-market DTC brands. They set up a data clean room with their ad platforms. Luna: I've seen that term thrown around. Can we back up — what actually is a data clean room, in plain English? Lucas: Sure. Think of it as a neutral, encrypted environment where a brand and a platform — say, Meta or Google — can match their customer data without either side seeing the other's raw data. The brand uploads its hashed customer emails and purchase history. The platform matches it against its own user graph. What comes out are aggregate insights and audience segments — not individual records. Luna: So it's privacy-compliant because neither party actually shares raw data. They just query against each other in a locked room. Lucas: Right. The analogy is two people standing on opposite sides of a frosted glass wall. They can see each other's silhouette, but not the actual face. That's the clean room. Luna: And SleepWell used this to do what exactly? Lucas: They uploaded their list of 80,000 past purchasers — hashed email addresses. The clean room matched about sixty-five percent of those to existing users on Meta. Then Meta — without revealing which users — told SleepWell: 'Here's an audience of lookalikes based on the common traits of your matched customers.' That audience performed at an eighteen percent lower CPA than their previous broad targeting. Luna: Eighteen percent — that's significant for a brand spending $150K a month. That's like $27,000 in savings every month. Lucas: Assuming the spend stays constant. But the key insight is that the clean room didn't just improve targeting — it also let SleepWell measure incrementality. They could run a lift test where half the audience saw the ad via clean-room targeting and half got a PSA — and see the actual sales lift from the ad itself. Luna: That's the holy grail for DTC brands right now. Real attribution that isn't just last-click or probabilistic. Lucas: Exactly. And the trend is picking up. According to a recent survey by the IAB, about thirty-eight percent of advertisers have used a data clean room as of mid-2026, up from maybe fifteen percent two years ago. DTC brands are still under-indexing — but the ones that have adopted it are seeing a clear competitive edge. Luna: What's the barrier? Is it technical complexity? Cost? Lucas: Both. The biggest clean room platforms — Amazon Marketing Cloud, Google Ads Data Hub, Meta's own solution — each have different interfaces and query languages. You need someone who can write SQL or at least work with a data engineer. For a small brand with a two-person marketing team, that's a non-trivial investment. Luna: And the cost. I've seen clean room minimums around $50,000 a year for some platforms. Lucas: It varies. Google's Data Hub is essentially free if you're already spending on Google Ads — you just pay for the compute. Amazon Marketing Cloud charges based on queries. Meta's is more bundled. But the real cost is the people and the time to set up the matching and interpret the outputs. Luna: So it's not a silver bullet for everyone. But for brands at SleepWell's scale — let's say $5 million-plus in annual ad spend — it starts to make sense. Lucas: Absolutely. And it's not just about acquisition. Clean rooms also enable closed-loop measurement for retention campaigns. You can match your email list to a platform, suppress existing customers from acquisition ads, and then measure whether a retention campaign actually drove repeat purchases. Luna: That suppression use case alone can save a lot of wasted spend. I've heard brands say they were accidentally targeting existing customers with acquisition creative — which is basically burning money. Lucas: Right. And clean rooms fix that by creating a match that the platform uses to exclude those users — but again, without the platform seeing the raw email list. It's a win for privacy and for performance. Luna: Is there a downside beyond cost and complexity? Any risks? Lucas: One risk is data leakage if the clean room isn't configured properly — but that's more about implementation than the technology itself. The bigger risk is that brands over-rely on platform-provided clean rooms, which are essentially walled gardens. You're still trusting Meta and Google with your data — just in a more controlled way. Luna: So if you want true independence, you'd need a third-party clean room provider — like Snowflake or Habu — that sits between you and all platforms. Lucas: Exactly. And those are gaining traction, especially among larger brands that have data across multiple platforms. But for a DTC brand starting out, the easiest entry point is usually the platform's own clean room — because it's already integrated with their ad manager. Luna: Let's talk about a concrete example besides SleepWell. Any other DTC brands using clean rooms in creative ways? Lucas: There's a supplement brand — I'll call them VitaBloom — that used Amazon Marketing Cloud to understand which products customers bought together. They found that people who bought their magnesium gummies were three times more likely to buy sleep aids within 30 days. So they built a cross-sell campaign on Amazon that targeted magnesium buyers with sleep-aid ads — and they saw a 25 percent increase in average order value. Luna: That kind of insight is hard to get from standard analytics because you don't see the cross-platform journey. The clean room gave them a privacy-safe view. Lucas: Right. And it's not just Amazon. Another brand I spoke with used Google's Data Hub to match their CRM data with Google's search and YouTube data. They discovered that customers who watched a certain type of video ad — a longer-form testimonial — had a 40 percent higher lifetime value than those who only saw a short bumper ad. So they shifted their YouTube budget toward longer formats. Luna: That's the kind of actionable insight that justifies the setup cost. It's not just about cheaper acquisition — it's about understanding your customer better. Lucas: And I think that's the real value of clean rooms. They're not a targeting hack. They're a data infrastructure play that lets you ask better questions about your customers — safely. Luna: I want to ask about the future — but first, a quick thing. If we're having a conversation about data infrastructure and marketing, and it's sparking ideas you're actually using — a lot of listeners have mentioned that these episodes help them think differently about their own strategy. Lucas: Yeah, we hear that a lot. And honestly, the reason we can keep doing deep dives like this — without a sponsor interrupting — is because of listeners who toss a few dollars in the tip jar. It's a couple of dollars a month that genuinely keeps these going. Luna: If these conversations have helped you, buy me a coffee dot com slash fexingo. No pressure, no perks — just knowing it's useful. Lucas: And we're back to our future question. So Luna, where do you see clean rooms going for DTC brands? Luna: I think we'll see more brands using clean rooms for predictive modeling — like training machine learning models on matched data without exposing raw data. And I also think smaller brands will get access through platforms like Shopify, which might offer a clean room as a service. Lucas: I've heard whispers about Shopify building a clean room layer. That would be a game-changer for DTC brands because it would lower the technical barrier significantly. Luna: Imagine a small brand with 10,000 customers being able to upload their list and get a lookalike audience on Meta without hiring a data engineer. That's where the industry needs to go. Lucas: And I think the privacy regulators are pushing in that direction too. The more that clean rooms become standard, the more they'll be seen as the responsible way to do ad targeting. It's better for consumers, better for brands, and eventually cheaper. Luna: So if you're a DTC brand listening and you haven't looked into clean rooms yet, the first step is simple: check if your ad platform offers one. Often it's already available in your account — you just haven't turned it on. Lucas: And start with a small test — like suppressing existing customers from a prospecting campaign. That alone can prove the concept and save you money on day one. Luna: That's a low-risk entry point. And from there, you can graduate to lookalike audiences and lift measurement. Lucas: Exactly. Clean rooms are not just for the enterprise anymore. They're becoming a practical tool for any brand that takes its data seriously. Luna: And with privacy regulations tightening everywhere, it's smart to get comfortable with the infrastructure now. Lucas: Alright, that's it for this episode of The DTC Marketing Podcast. Thanks for listening, and we'll catch you next time. Luna: Take care, everyone.