The Drum Awards Festival - Official Deadline

-d -h -min -sec

Deep Learning Online Advertising Invalid Traffic

What is invalid traffic and how can advertisers prevent it?

RTB House

|

Open Mic article

This content is produced by a publishing partner of Open Mic.

Open Mic is the self-publishing platform for the marketing industry, allowing members to publish news, opinion and insights on thedrum.com.

Find out more

September 23, 2022 | 5 min read

With advertising fraud set to cost advertisers $100bn by next year, it’s vital that digital marketing platforms adopt a zero-tolerance stance against invalid traffic

With advertising fraud set to cost advertisers $100bn by next year, it’s vital that digital marketing platforms adopt a zero-tolerance stance against invalid traffic.

Advertisers expect the marketing partners they engage with to manage traffic quality so that their brand messages reach fraud-free and real audiences. This makes it imperative for advertising platforms to adopt a multi-faceted approach for their customers that embraces AI and uses cutting-edge tools.

In today’s competitive digital marketing landscape, providing a protected eco-system that supports clients and delivers the best quality traffic and inventory is the minimum standard that needs to be achieved.

What is invalid traffic?

Invalid traffic includes any activity (bids, impressions, clicks, or site/app events) that doesn’t come from a real user with genuine intent, and therefore may artificially inflate an advertiser’s cost.

This can be intentional or accidental, and range from automated tools or traffic sources to accidental clicks encouraged by flawed ad implementation.

There are two types of invalid traffic:

  • General invalid traffic (GIVT)

This type of activity comes from non-human, potentially malicious traffic that could easily be detected and filtered using industry blacklist and standard parameters.

  • Sophisticated invalid traffic (SIVT)

Activity of this kind is dangerous and hard to combat because it is intentionally fraudulent traffic whose purpose is to resemble real users.

Experts suggest that GIVT is like the ‘white noise’ of fraud, meaning it is always on in the background, but its impression volume is much lower than SIVT.

SIVT is more challenging to tackle because of the strategy often used by fraudsters to obscure their fraudulent activity by blending it in with legitimate inventory, making it harder to detect.

What lengths might fraudsters go to?

Just like fraud in any industry, fraud within the advertising sector comes in various shapes and sizes.

Reputable digital marketing platforms will view any level of fraud as unacceptable, and will have systems in place to spot suspicious activity rapidly before client money is wasted on false impressions.

One of the most sophisticated schemes of recent years involved a firm purchasing legitimate Android apps, monitoring the users’ behavior, and then creating bots to mimic this behavior.

The scheme ended up fraudulently earning its owners hundreds of millions of dollars from brands whose ads were shown to bots instead of humans, according to a report by Buzzfeed.

Data from Pixalate suggested that just one of these apps could generate $75m a year in stolen ad revenue, showing how quickly advertising firms can spend money on invalid traffic.

With greater levels of online activity and different types of online audiences, invalid traffic is not only growing in volume, but is spreading across various different sectors, with the likes of online in-game advertising being targeted too.

How can advertisers protect themselves?

The optimum way for advertisers to protect themselves is to partner with a digital marketing platforms that has a track record of uncovering invalid traffic and that has systems in place to identify fraud.

Advertisers should check that their partners have a set of rules in place to protect their brands from delivering ads to illegitimate traffic, regardless of whether it is general (GIVT) or sophisticated (SIVT).

Digital marketing platforms should have end-to-end processes that include gathering data from the advertiser’s website or app, proactive pre-bid filtering of non-human activity, and identification of fraudulent impressions and clicks.

Most industry leaders would state that the best products should include comprehensive strategies to protect client reputations and revenues, including data center analysis and deep learning in-house algorithms.

Conclusion

By continuously pursuing the highest standards of IVT detection, digital marketing firms can ensure that every dollar of every advertiser’s media budget is used to reach real audiences, and to produce only legitimate impressions from transparent purchase paths.

Advertisers cannot afford to ignore the risk of IVT, because doing so not only wastes their time and effort, but it wastes their hard-earned money too.

In a world where fraudsters are only becoming more sophisticated, it’s vital that advertisers partner with agencies that bring equally sophisticated tools and methods to the fight against illegitimate activity.

Why tackling IVT is vital for clients

Every digital marketing platform’s clients have worked tirelessly to develop their brands and the advertising campaigns that support them, meaning it’s essential that they reach a legitimate audience.

RTB House receives close to one trillion bid requests daily that need to be validated and filtered in real-time.

Platforms that experience this amount of bid requests will need in-house expertise and proprietary deep learning engines, supported by industry registers of invalid sources of traffic.

It is also important for advertisers to work with partners that are accredited by institutions such as the Trustworthy Accountability Group, or TAG, an industry threat-sharing network that works to uncover the most malicious ad fraud botnets.

Deep Learning Online Advertising Invalid Traffic

Trending

Industry insights

View all
Add your own content +