Excluding bad placements automatically to eliminate low-quality PPC traffic
Unfortunately for advertisers, it’s extremely easy to abuse ads on Google Display Network. With paid search, Google applies a certain level of protection to your ads, saving you from lots of headaches. But running banner or video ads on the GDN’s placements is a whole different thing.
You are usually left to the mercy of the website (or app) owners, so the only thing you can do is to hope that everything will be okay with your ads. We know, however, that letting things take care of themselves usually doesn’t yield good results.
That’s why we’ve recently updated our core functionalities. Our users are now able to protect themselves from low-quality, disruptive, abusive, and fraudulent clicks coming from display campaigns.
Understanding the Problem
In display campaigns, a placement is what we call a traffic source. It could be an app, a monetized YouTube video, or any plain old website that runs ads via Google Adsense or DoubleClick. For the purposes of understanding the scale of the threat, one need look no further than the Google Play Store.
We read multiple reports of drastic, almost criminal activities that are happening daily on Google Play Store, usually in the form of malicious ad-clicking apps. Google has labeled these apps as PHAs - potentially harmful apps.
Before 2018, click fraud apps were not considered a PHA. It was shocking to see that once they were recognized by Google, they instantly became the foremost type of potentially harmful apps on the Google Play Store. Ad fraud PHAs simply dominate the ecosystem. They account for 55% of all PHAs.
But that’s not all. Think about all those reports coming from Asia about click farms where hundreds of mobile phones are re-programmed to click on ads, produce video views, or like Instagram or Facebook posts. You’ll easily understand why display campaigns, even though undoubtedly useful, are a potential source of trouble.
Our Reply: A New "Placement as a Source" Algorithm
We listened to your suggestions and realized we needed more tools to fight click fraud on placements. So we created a placement as a source algorithm update which gives you more possibility to deal with click fraud. By adding placement as a source, we’ve enabled the creation of protection rules against placements, not only clicks, as well as post-click behavior analysis for each placement.
If you want to learn about the quality of traffic you’re getting from a placement, or whether you’re a victim of ad fraud, post-click behavior analytics data is where all the insightful information is stored. Looking through it is the only way we can be sure if a click was fraudulent or not.
There are two ways to get the data you need. You can use the tracking template that you put in the Google Ads settings, or you can use ClickGuard’s tracking pixel that lives in the source code of your landing pages. Either way, you get the information about, among other things:
- The origin of the traffic
- The frequency of delivery that certain placement brings
- The intensity of display ad clicks
- How much time do visitors actually spend on the website
- Whether the visitors’ devices have a proper digital fingerprint
All of this gives us an answer if the visitor is legit and is the placement sending quality traffic. Here’s an example:
The screengrab above shows a placement report. You can clearly see that a big number of placements brings a substantial amount of suspicious and flagged clicks. Every other click coming from the placement in the fourth row, for example, is problematic. That’s a prime candidate for exclusion.
IP exclusions have a limitation of 500 entries per campaign. When it comes to placements, the limits are much more generous. You can enter a maximum of 20,000 placement exclusions at a time, and a total of 65,000 exclusions per account.
This is fantastic news for you since it enables you to be far more restrictive when setting up protection rules. You can be comfortable with your ability to eliminate all sources of bad display traffic.
Now when a placement is defined as a source, you are free to use it in other types and categories of protection rules. A common use case might be: “If a placement brings more than 10 clicks from devices that don’t have a digital fingerprint, then exclude the placement”. Before the update, you were only able to count clicks which weren’t good enough to protect display campaigns.
Display campaigns are most sensitive to no-fingerprint and time-on-site rules, therefore placement as a source is the most powerful when used with those protection rules. As previously mentioned, please make sure to have ClickGUARD’s pixel and tracking template properly set.
This is a placement report for a single website/app. It’s the place where we can investigate the overall quality of a placement. Are visitors converting, are they staying on the website or bouncing straight away, what is the level of threat they are presenting? This level of detail was previously available only when looking at “click as a source”.
The idea of expanding your brand’s reach by placing ads on specific websites sounds inviting. From a simple remarketing to more advanced interest based targeting, display campaigns can produce great results. But they hold a ton of uncertainty, especially with increased ad fraud.
So, we invite you to test few protection rules. Once you get the hang of the basic setup, take a deep dive into the various capabilities and optimization settings. Add conversion based rules, time-on-site rules, and find the “sweet spot” of protection rules that gives the best results for your business.
As always, we’re here to help! If you have any questions about Placement and Display rules, let us know in the chat below.