Free Livestream event with Roku and AdAge discussing OTT on May 20th.  Register Here Free Livestream event with Roku and AdAge discussing OTT. Starts at 9am PST.  Register Here

Painless Ways To Prevent Ad Fraud

By May 14, 2020 May 26th, 2020 Blog

Marketers must not play into the false economy of digital advertising.

Many mobile marketers are operating in a false economy because of the pervasiveness of fraud. Fraud can  infiltrate campaigns in many ways, making it difficult to extricate it and justify the resources to do so especially when performance numbers look favorable. If you take your campaign data at face value, you are relying on inaccurate data that appears to show a broad reach to customers. And we all know that appearance isn’t everything.

The misbelief that half of marketing is fraudulent is unacceptable. Marketers shouldn’t have to gamble in launching campaigns. Although fraud has become an inherent part of campaigns, it does not mean marketers should pay for it as part of the cost of doing business. 

We all know that fraud takes many forms, and while there are numerous algorithms to identify them, loopholes exist where a fraudulent entity can hide and appear legitimate. In a perfect world, fraud would be completely eradicated but that is unrealistic.  However, there are several relatively simple actions you can take to greatly reduce fraud in your campaigns. 

It’s all about the ad signal. Signal clarity reflects how reliable the click and impression data sent are. Each must be sent in a separate data stream and is how media partners transmit valuable data such as device IDs, creative IDs, transaction IDs, etc., necessary for attribution, fraud prevention, and campaign optimization, among other marketing criteria. 

These data streams are what fuel the attribution process where credit is adorned to the channel in which the conversion was obtained. Fraudsters aim to infiltrate this system and win attribution through any means necessary (eg, click flooding, click injection through malware, SDK spoofing, junk publishers, fake installs, etc.). Work with your media partners or find one through the Kochava Traffic Index as a way to ensure you are sending a clear ad signal from your campaigns.

Avoid buying cheap CPM. When you buy cheaper inventory, you increase the risk of purchasing fraudulent traffic. Understand where your media is being purchased and aim for quality over quantity, this might mean retargeting a high-quality user audience segment or creating a lookalike audience. When you (your agency) buys inventory with a low CPM, you run the risk of dealing with fraudulent publishers masking their identity and appearing to be a legitimate site.

Check names: Request that ad partners provide publisher site IDs. Fraudsters in this arena are guilty of falsifying transparency by masking their traffic sources by cycling through their sub-publisher site IDs. You can pull a summary report to include site IDs and installs to easily tell which sites are acting suspiciously. If you see a high number of sites with few installs, be wary. This is an unrealistic and ineffective way of finding traffic. For example, having hundreds and maybe even thousands of sites yielding upwards of 20K clicks for a few dozen installs is a red flag that there is fraudulent activity involved. It’s better to have fewer sites with more clicks than numerous sites with just a handful of installs.

It’s also important to have unobscured site IDs to understand where your ads are being served. Are they coming from Candy Crush? Or are they coming from some unknown Peruvian windsurfing blog?

Set the rules: Use traffic verification. With traffic verification, marketers set thresholds to avoid exorbitant impression and click volumes and/or block certain types of traffic that you specify. If traffic doesn’t meet your parameters, it will omit the traffic from attribution. This greatly reduces the presence of fraud and is your first tier in your fraud strategy once your campaign is underway. 

In an informal three-month analysis between advertisers using traffic verification and those not, Kochava saw only 3% fraudulent installs and 41% fraudulent traffic caught by our anti-fraud algorithms, after traffic verification dropped invalid traffic. This was in contrast to 16% fraudulent installs and 88% fraudulent traffic without traffic verification. Note: Without traffic verification in place, those fraudulent installs were attributed.

Fraud is not a fact of life.

Fraud is a fact that marketers don’t have to accept and absorb in their ad spend. But doing nothing about it is an impractical way of doing business. Install numbers may look as expected or extraordinary, but peel that initial layer away and you’re left with few actual conversions. Having protective measures in place will show your true conversions, and if done incrementally, marketers can ease into their anti-fraud strategy without drastic changes in their expected campaign outcomes. 

Preventing and mitigating fraud doesn’t have to be time-consuming and exhaustive of your resources. By selecting any of these measures, you have taken the first step in your anti-fraud strategy. At minimum, have the Global Fraud Blacklist enabled for basic protection against known fraudulent entities (devices, IP addresses, site IDs). If you’re unsure of how to establish your custom settings in Traffic Verifier, contact us to help you establish settings right for your brand and app traffic.