Walled Gardens: Not As Bad As You Think

krishnaboppanaData-Driven Thinking" is written by members of the media community and contains fresh ideas on the digital revolution in media.

Today’s column is written by Krishna Boppana, chief data scientist at Digilant.

With Facebook’s recent Atlas launch, Apple’s iAd refocus and Google’s restrictions on DMPs’ pixel firing, we’re seeing the rapid emergence of the “walled-garden” model of digital advertising.

As a result, advertisers may feel that they’ll be locked in when choosing a platform to work with. But in reality, they’ll have access to consolidated high-quality inventory, while audience reach, targeting and attribution will be in silos.

The closed environments will give those networks the ability to improve inventory quality, provide in-network attribution and develop different creative ad units and solutions. Industry consolidation will undoubtedly cause some players to exit or pivot in the near future, but ultimately, the walled-garden model will be positive for all elements of the digital ecosystem.

Accurate Audiences

“Walled garden” generally refers negatively to the boundary between two resources that should otherwise be integrated. It’s a positive term in this context. Programmatic vendors have always touted their algorithms as key differentiators, but even the most accurately predictive algorithm won’t produce desired results without separating each advertiser’s data into individual silos.

For example, if an algorithm ingests a massively disproportionate amount of data from one advertiser, results for all customers will be skewed towards the largest advertiser and category. While this may not concern the biggest brand players, smaller ones may be incentivized to turn to Facebook or Google for what they perceive as more accurate audiences.

Impact On Fraud

As 2014 comes to a close, many advertisers still lack true insight into their campaign performance, thanks to opaque buy-side platforms and the persistence of fraud. While one could argue that fraud will always be an issue to some extent, like a typical software virus, advertisers can expect to get a better handle with controlled environments. Facebook is promising the holy grail: fraud-free advertising with walled user data, cross-device tracking and ad serving. With user registration data (Facebook login), controlled inventory, ad units and built-in attribution, it promises to be a fraud-free solution.

Walled gardens help to limit publishers’ data leakage and hold publishers to higher levels of transparency to control fraud at the source of the problem. The presumed default onus of providing higher-quality advertising is shifting from buy-side platforms to publishers and sell-side platforms. This is an important step forward for the programmatic ecosystem and for advertisers.

Reach Across Walled Gardens

The main drawback of a solution like Facebook’s Atlas is that it still only reaches Facebook’s user base and relies on its (albeit enormous) user reach. While Facebook will be a component of most campaigns, it’s not the only channel that can reach an advertiser’s target audience at any given time.

With the success of cross-device identification solutions, it’s more important than ever to reach users at the right time on the right device with the right creative. This simply means advertisers do not have to get “stuck” in a walled garden, but rather leverage them by using their buy-side platform to reach across walled gardens.

The walled-garden phenomena won’t doom the industry. Increased competition and higher-quality inventory will produce a race to the top, where the most successful technology providers are also the most transparent and effective. This will go a long way toward helping digital marketers deliver on the promise of true one-to-one marketing at scale.

Follow Digilant (@Digilant_US) and AdExchanger (@adexchanger) on Twitter.

 

Add a comment

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>