How WeQ Helps You Fight Fraud
A Closer Look at Mobile Fraud
Mobile ad spending is growing by 20% this year and is valued at over 75$ billion, according to Forbes.
It’s inevitable that as mobile ad spending goes up, so does mobile fraud: RetailDive reports that fraud rates have almost doubled since 2017 and are costing the industry billions of dollars — somewhere in the region from 6.4 to 7.2 (TUNE) — based on install rejection rates by fraud prevention tools. It is estimated by AppsFlyer that financial exposure to app install fraud in Q1 of 2018 was as much as $800 million – a 30% increase over the last year. There are many forms of fraud, ranging from click spam, SDK spoofing and false impressions to faked installs. Its effects can be devastating on budgets and data sets.
Advertisers are finding increasingly limited inventory to spend budget, while demand continues to grow, resulting in a huge opportunity for fraudsters to steal from advertising budgets across the supply chain (advertisers, publishers or supply partners). So, how to tackle it? While mobile fraud is increasing, one of the key impediments to fighting fraud remains the lack of unified benchmarks and actions in the industry. When it comes to fraud, there is no black and white: for example, in some regions it is normal to receive many installs from one IP, since all phones in this region will use the same IP. In other regions, this would be considered unusual. Another highly debated topic is CTIT (Click-To-Install-Time) – with every big player having a different benchmark – which can be affected by size of the app or speed of the internet connection and other factors. As such, understanding fraud and educating ourselves is the first step towards dealing with this persistent and ongoing battle.
As mobile fraud gets more sophisticated and harder to detect, it’s vital to understand its dangers and how you can control it. In this post, we will discuss what ad fraud metrics look like and which tools we can use to help fight mobile ad fraud.
Introducing WeQ SHIELD
WeQ are here to drive change in the mobile industry and are proud to announce our new tool WeQ SHIELD: our answer to protect our mobile advertising clients.
For the past 6 months, our WeQ Compliance Taskforce worked tirelessly on creating a comprehensive in-house fraud prevention solution that addresses the current compliance needs of the mobile marketing industry. The combined expertise of our in-house team of developers, data scientists and BI specialists ensured that we stay ahead of the curve with our knowledge of different types of fraud. Our team worked with clients and attribution partners (MMPs) such as Appsflyer, to understand their perspective. The result is WeQ SHIELD.
We believe that we have to make tools available internally for our account managers to be able to proactively monitor data for clients, brands, and app developers. This allows us to identify and prevent the risk of fraud affecting the campaign performance.
Sven Lubek, Managing Director at WeQ, said:
“For all WeQ account managers striving to provide high quality traffic to their advertising partners, WeQ SHIELD is an AI product which unlike other manual or statically automated systems uses Machine Learning to constantly adapt our anti-fraud detection mechanisms. This empowers our team to take immediate action in the fight against fraud.”
How does WeQ Shield help to fight fraud?
We check our advertiser’s traffic regularly and proactively for fraudulent patterns and non-compliant behaviour. It’s essential for our account managers to sit with the client and define their performance KPIs as well as all-compliance benchmarks that apply to their type of app (defined by the client’s MMP or by their internal metrics) to ensure that everyone is aligned.
We have also identified patterns for flagging potential fraudsters, laid out strategies to block fraud, and developed methods to prevent the rising level of fraud. Since fraudsters are always finding new ways to outsmart advertisers, catching fraud is like a game of cat and mouse and the tools need to be constantly revised. In order to do so, we have gathered an expert team of data scientists and BI specialists who use machine learning to comb through hundreds of millions of events and data points daily such as installs, post install events, traffic distribution; and device and operational system variations. This builds a knowledge base which helps to predict and take action against any non-compliant behaviour. Based on our experience in running global mobile advertising campaigns, we have identified two stages of fighting fraud and developed our tools accordingly, to address each part.
First stage: Once the campaign is launched, the monitoring of data starts. All the traffic goes via our WeQ SHIELD solution: WeQ’s compliance tool and internal BI platform, where we can immediately spot non-compliant traffic. It contains interactive dashboards and visualisations that help our account managers drill down into their campaign to the sub id level to take data-driven actions based on different compliance metrics, thus reducing the risk of fraud for our advertisers.
Second stage: AMs regularly monitor their campaigns efficiently and detect issues, so that they can take action by blocking traffic that is not compliant and communicate this proactively to advertisers. We screen our traffic to ensure that all our campaigns are meeting client’s anti-fraud benchmarks.
Below are some of the metrics WeQ SHIELD monitors:
- Industry standards and CTIT distribution: short and long click-to-install time can be an indicator of different fraud types — it can be considered an issue if it presents itself in a disproportionate amount within a campaign.
- Behavioural data on post install events: retention rate, for example, which can also reveal information on the traffic quality. Our account managers aim to frequently optimize based on these events, in order to achieve the highest performance and block anomalous user behaviour.
- Session IP duplicate: if we see multiple conversions coming from the same IP address it might indicate VPN traffic or fraudulent installs, therefore we always keep our eyes on duplicates to ensure advertisers are not being charged for potentially fraudulent conversions.
- CR per source: the hourly conversion rate, is monitored in combination with other compliance metrics such as CTIT, to recognize suspicious patterns in the traffic. This ensures that the conversion rate is within our client’s benchmarks.
- Unexpected OS version distribution: the industry standards for OS version distribution and device brand distribution per GEO are monitored. WeQ Shield can detect if there are any anomalies, which our account managers can swiftly deal with.
- Blacklisted traffic sources: these are set in advance with the advertisers before starting a campaign. We ensure that we don’t serve blacklisted sources.
- Unique identifier: depending on the clients’ SDK, we might also add an additional security feature to authenticate the postbacks by adding a unique identifier. This ensures that the conversions are valid and we avoid any potential fraud
Why choose WeQ SHIELD?
As you have seen, mobile advertising fraud is growing day-by-day without any sign of slowing down. It is becoming increasingly difficult to detect which is made more difficult by the fact that here is no unified agreement on fraud and compliance benchmarks. Advertisers are putting a huge amount of their budget at stake every time they run a campaign.
While advertisers bear most of the financial costs, everyone in the industry suffers from fraud. As an ad tech player, it is necessary to be educated and stay ahead of all the different types of fraud that appear on the market. As the next step, it’s important to be vigilant by paying more attention to the patterns in owned data. User acquisition managers should work closely a trustworthy partner who can provide the expertise and tools to identify fraud patterns, alert them about any risks and suggest strategies for optimizing campaigns. WeQ SHIELD is a comprehensive anti-fraud solution that combines machine learning technology and human expertise, which provides all the indicators to identify and combat fraud.
Lastly, as it often happens, while the client sees the full picture, mobile advertising partners see only part of the data: open data sharing and transparency among attribution and tracking firms is necessary to combat fraud together!