We need proof: Why every click needs an impression
Posted Feb 12, 2019
Mobile ad fraud has been one of the most discussed topics in our industry for the last few years, and during this time Adjust has led the charge by establishing industry-standard definitions and measures that actively fight and reject fraud before it reaches campaign budgets - rather than just detecting it after the funds have gone.
Today we’re taking the next logical step in cleaning up our industry and protecting our clients. We’re doing this by asserting what should be a very obvious standard:
That for every click, there must be an impression.
The emperor has no clothes
Common sense dictates that a user can only click on an ad if they have actually been served it and have seen it.
However, in mobile marketing, ad channels are free to claim a click on an ad without informing attribution partners of a prior impression. This creates a real issue where many cases of fraudulent clicks are delivered which could easily be validated with impression data attached.
“Of course the clicks are real!” is an adage heard far too often in mobile marketing. Which is surprising, since proving otherwise is extremely simple.
This is not a new problem either, but has been a “necessary evil” since the inception of mobile advertising. It also hasn’t really been challenged since.
Our clients are always stunned to learn that many partners won’t provide us with proof that an ad was actually delivered before the user clicked on it. This surprise is heightened by the fact that network partners must, for their own analytics and ad delivery systems to work, track impressions and ad delivery. So the technology to share this crucial data with attribution partners already exists, but it isn’t shared with us.
Adjust’s new standard for Click Validation requires ad partners to share this data with us so we can answer a very simple question: was there a matching impression for a click?
With this, we can end two pervasive forms of fraud for good: Click Spamming and Click Injections.
The numbers don’t add up
Over the years, Adjust has made great strides in developing a new anti-fraud methodology to ensure the authenticity of app interactions measured by our SDK. In doing so, we have prevented even the most sophisticated fraudsters from gaming the system from the attribution data side. But data from the network side still offers an opportunity for ad spend theft.
Right now, measurement partners have no choice but to base their metrics off ad network claims of ad interactions, some of which may originate from fraudulent publishers. While there are giveaways that help us identify fraud methods such as Click Injection or Click Spamming, for the vast majority of clicks there is no reliable indication if the user ever actually saw an ad or not.
To give this some perspective: Adjust receives over one billion clicks a day. This is an incredible amount of activity. However, this figure does not include major self-attributing network partners, and it only represents Adjust's share of the market. Given that it's clicks we’re looking at, there are simply not enough people on earth to actually produce this many legitimate engagements.
With our new standard, we’d filter through this number immediately, and be able to deliver accurate data to our customers. Without it, everyone is still in the dark.
Raising the bar
Adjust’s latest white paper lays out in detail a new standard that requires ad channels to send impressions to us with a unique identifier that will also be attached to any resulting clicks before the click claim is sent.
This approach allows us to confirm if a click is logically possible, as we’ll be able to check if there was a matching impression within a reasonable timeframe made by the same device.
This seemingly trivial requirement dramatically increases the workload for fraudsters seeking to steal ad spend.
Given a normal click-through rate of 1%, an attacker would have to create 100 times more data to spoof plausible looking clicks, which makes Click Spamming more expensive, but also easier to spot. Click Injection would be impossible to perform, as an attacker could not generate a previous impression. In fact, many other fraud methods would struggle to prove that an ad was actually served.
While there still be some ways to steal attribution and spoof engagements, the cost of doing so will increase significantly.
An open approach
This new standard is open and Adjust invites everyone to adopt it as soon as possible. This step towards transparency was long overdue and Adjust will, once again, lead the way.
Members of the Coalition Against Ad Fraud (CAAF) have already begun the work to send this data to enable Adjust’s upcoming Click Validation filter. We also invite all networks to reach out to us for more information.
It’s just common sense: for every click, there must be an impression.