A walled garden in advertising is a closed digital platform that controls how ads are bought, delivered, tracked, and reported within its ecosystem. Google, Meta, Amazon, TikTok, and similar large tech companies all operate walled gardens. The defining feature is that user data generated inside the platform stays inside the platform. Advertisers can spend money there and see performance reports, but they can’t export granular user-level data or bring their own independent measurement tools the way they can on the open web.
How Walled Gardens Work
Every walled garden collects first-party data from its users: search queries, browsing behavior, purchase history, video views, app usage, demographic details. The platform then uses that data to let advertisers target specific audiences and measure results. The catch is that the platform owns the data and decides how much of it advertisers get to see. In most cases, you receive only aggregate-level insights, not the raw, user-level information you’d need to build your own targeting models or stitch together a full picture of how a customer moved between platforms before converting.
Walled gardens also require you to use their proprietary ad-buying tools. You can’t plug in a third-party ad server or demand-side platform the way you might when buying display ads on the open internet. The platform handles targeting, delivery, optimization, and reporting through its own vertically integrated technology stack. Ads typically appear in native formats that blend with the surrounding content, whether that’s a sponsored search result, an in-feed social post, or a promoted product listing on a retail site.
Behind the scenes, algorithms determine which users see your ads, when, and how often. Those algorithms are closely guarded trade secrets that change constantly. You have no visibility into how the platform’s machine learning models prioritize one ad over another, and platform operators sometimes use the same algorithms to favor their own products and services.
Which Platforms Are Walled Gardens
The biggest walled gardens by ad revenue are Alphabet (Google Search, YouTube, Google Maps), Meta (Facebook, Instagram), and Amazon’s advertising business. But the category extends well beyond those three. TikTok, Microsoft (LinkedIn, Bing), Pinterest, Snapchat, Spotify, and X all operate as walled gardens. Retail media networks run by large retailers function the same way: they sit on rich purchase data, sell ad placements within their own shopping environments, and restrict how that data can be used elsewhere.
Even ride-sharing apps, travel booking sites, and music streaming services increasingly sell advertising through walled-garden models. Any platform with a large logged-in user base and proprietary ad tech qualifies. The common thread is that advertisers must play by the platform’s rules to reach its audience.
Why Advertisers Use Them Anyway
Despite the restrictions, walled gardens attract enormous ad budgets for a straightforward reason: they have the audiences. Billions of people use Google, scroll Instagram, watch YouTube, and shop on Amazon every day. The platforms’ first-party data also allows targeting precision that’s difficult to replicate on the open web, especially as third-party cookies have become less reliable.
Walled gardens also offer built-in privacy protections. Because user-level data doesn’t leave the platform, there’s less risk of data leakage, unauthorized misuse, or regulatory violations. For brands navigating tighter privacy laws, spending inside a controlled environment can simplify compliance. And the platforms’ native reporting dashboards, while limited in scope, provide automated metrics on campaign delivery, demographics, ad performance, and conversion tracking without requiring additional analytics infrastructure.
The Measurement Problem
The biggest frustration advertisers face with walled gardens is measurement. Each platform grades its own homework. You see the numbers and analytics the platform chooses to share, and you have limited visibility into how ads were served, which specific audiences saw them, or how the algorithm made delivery decisions.
This creates real problems when you’re trying to understand the full customer journey. A shopper might see your ad on Instagram, search for your brand on Google, and then buy on Amazon. Each platform will take credit for the conversion in its own reporting, but none of them will show you the complete cross-platform path. Social media marketing is consistently ranked among the hardest channels to attribute revenue to. In one industry survey, 44% of marketing professionals said social was among the most difficult channels for sales attribution.
Independent measurement compounds the issue. Many advertisers want their media partners to use preapproved, third-party measurement tools. But the largest walled gardens are the ones most likely to decline. A survey of advertisers found that among those who didn’t require preapproved tools from every partner, it was typically the biggest walled gardens that got the exemption. The lack of transparency around media quality metrics has led a majority of digital media professionals to say it could affect how much they’re willing to spend on certain platforms.
How Data Clean Rooms Fit In
Data clean rooms have emerged as a partial solution to the measurement gap. A data clean room is a secure technical environment where an advertiser and a platform can combine their data sets for analysis without either party seeing the other’s raw, user-level information. You might upload your customer list, and the platform matches it against its user base to show you overlap, conversion rates, or audience segments, all without anyone’s personal data leaving the secure environment.
This matters because it lets advertisers do things like measure whether people who saw an ad on one platform actually purchased through a retailer, or build more precise targeting by layering retailer purchase data onto a social platform’s audience graph. Pinterest, for example, hosts data clean rooms so advertisers can leverage retailer data for targeting and attribution. The result is more accurate closed-loop measurement (connecting ad exposure to actual sales) without violating the walled garden’s data policies.
Clean rooms don’t eliminate the transparency issue entirely. You’re still working within the platform’s infrastructure and subject to its rules about what queries you can run and what results you can export. But they represent a meaningful step toward letting advertisers verify performance with their own data rather than relying solely on the platform’s self-reported metrics.
Walled Gardens vs. the Open Internet
The open internet is essentially everything outside these closed ecosystems: news sites, blogs, independent publishers, streaming services that sell ads through open programmatic exchanges. On the open web, advertisers can use their own ad servers, apply third-party measurement, and move data between platforms more freely. The tradeoff is that targeting on the open internet has historically relied on third-party cookies and device identifiers, both of which are becoming harder to use as privacy regulations tighten and browsers phase out tracking mechanisms.
Most large advertisers split their budgets between walled gardens and the open internet. Walled gardens deliver scale and targeting precision. The open web offers transparency, independent verification, and the ability to build a unified view of campaign performance across publishers. The tension between these two models is one of the defining dynamics in digital advertising today, and it shapes everything from how brands allocate budgets to how the industry develops new privacy-safe measurement standards.