Summary

Being one of the largest publishers of India there is bound to be requirements for large traffic acquisitions on the websites. But, what happens when you go out in the market to fulfill this requirement? You will be reached out by every Network partner, Agency, Ad platform across Indiato get a piece of your budget. Doesn’t matter what is delivered and how it is delivered, all that matters is does it meet your tracking systems requirements. Is there a way to find out what's genuine and who among all is dependable? What to answer when the advertisers ask on low engagement rates?

‍

The Problem

It all started with a common complaint over a period of time from different advertisers that they were seeing large number of suspicious clicks which were not converting for them. Having large traffic requirements also needs multiple network partners to fulfill. The publisher at one time was working with more than 160 network partners. With so many partners it became difficult to determine which one/s was delivering low quality traffic.

There was a lot of manual effort taken up by the internal teams to identify the fraudulent source sand they could identify 4 sources contributing to around 6% of the traffic, more was to come.T his process worked to an extent but it was a very difficult and a uphill task to scrutinize all the sources.

Major challenges faced by the publisher during this process,

1. Which network is actually working for them.

2. Identification did not necessarily mean the networks would accept it as there was not much data to present to the network partners.

‍

The Solution - Botman®

The publisher engaged with Botman® and the solution provided was the Ad Fraud Detection module of Botman®. One big advantage the publisher got with Botman® was that it did not have to check for fraud manually anymore as the entire process was automatic now.

On activation, Botman® immediately started screening traffic across all sources and bucketing them against different traps like Spamnet, Data Center, Proxy Traffic, Botnet, Crawler, Tor traffic,etc. Second level of filters were also added to catch traffic under combination of traps like Spamnet+Click Spam, etc. Result of an month long exercise of screening was that the suspect traffic ranged between 40%-45% across 148 network partners. Everything unusual or caught under the traps was flagged suspicious and notified immediately. Sufficient data points were shared in terms of IP addresses, User Agents, Geo Location etc for the publisher to present a case to their network partners.

‍

The Result

Reduced suspicious traffic to less than 5%

It’s important to acknowledge that a 100% solution to ad fraud is never possible. There will always be some amount of bot traffic coming to the websites.

Datasets shared with the publisher helped them to reach out to their network partners to block malicious traffic. It was able to blacklist the malicious IP’s, Domains, networks, etc and reduce the fraud rates to sub 5%. Though there are small spikes of unwanted traffic at times but with Botman® present it gets identified immediately.

This reduction in ad fraud has ensured that the publisher is getting the most out of their digital ad spends and their advertisers are also happy to see the steps taken by the publisher to curb ad fraud and increase their revenues.