How to use FouAnalytics

I'll be brief ;-)

If you buy large quantities of CTV ads

CTV is the fastest growing, and most expensive type of digital ads today. That's why bad guys focus their attention on stealing as much of these budgets as possible. Fraudsters use mobile apps to pretend to be CTV apps (2020, Grindr CTV fraud); they use smart refrigerators to fabricate fake CTV bid requests (2022, Smart Refrigerators Impersonating CTVs); or they use javascript code in ad slots to generate 12 billion fake CTV bid requests per DAY (2023, VASTFLUX Takedown). Even if you buy direct from CTV sellers, you may still be exposed to MFA sites, other forms of fraud, or simply your expensive CTV ads NOT showing on connected TVs, but on crappy sites and mobile apps (see example below - CTV ads running on websites). That is why smart advertisers use FouAnalytics CTV tags to measure where their CTV ads actually went.

If you buy CPC search ads

Google and Bing don't allow anyone to measure the search ads themselves. So we have to measure the clicks coming from these ads, and arriving on the landing page. By having the FouAnalytics on-site tags on the landing pages, we can see whether the clicks came from bots AND we can see where the clicks came from. In the example FouAnalytics data below, you can see a CPC (cost per click) search campaign running on Bing, where the ads and clicks came from a long list of suspicious sites. This tells us that "search partner network" was not turned off and a large portion of the dollars were lost to fake/fraud sites, search arbitrage sites, or other undesirable places outside of Bing .com or MSN .com. This is why you should have FouAnalytics as your independent way of checking where your ads went and where your clicks came from.

If you buy ads on social media/walled gardens

Since Facebook, Instagram, YouTube and TikTok don't allow anyone to measure the ads themselves, we place the FouAnalytics tag on the landing pages. This way, we can check the quality of the clicks that arrive from various paid media sources. See the 8 donut charts below and note the relative ratios of dark red (not good) and dark blue (humans, good) from the various sources.

In the TikTok example, the advertiser did not turn off Pangle, the audience network. So most of the ads were not loaded on TikTok, but on other crappy apps, AND the clicks we got on the landing pages came from those bad apps too (i.e. the apps had falsified the clicks too, in addition to loading the ads). In the FouAnalytics data below, you can see the clicks that arrived from utm_source=tiktok, but came from mobile apps that were NOT TikTok. This is why you need to measure your landing pages with FouAnalytics on-site tags, when running campaigns in social media and walled gardens.

If you run affiliate marketing

If you pay for affiliate marketing, be sure to use FouAnalytics to detect if affiliate partners are not just loading your affiliate landing pages in popunders on pr0n sites, hidden iframes, ad slots, or recommendation widgets to stuff the affiliate cookie and falsely claim credit for driving sales for you. You are paying them affiliate commissions that you didn't need to pay (because the sales would have occurred anyway, or had already occurred). In the following example, the affiliate page was loaded into a 300x250 ad slot on pr0n sites, so the affiliate cookies could be set on the browser without the human knowing (since they were watching pr0n).

More details: Detect AFFILIATE fraud with FouAnalytics, save millions

If you run retargeting and remarketing campaigns

If you believe that retargeting a user that has visited your site is useful, consider that bots deliberately visit your site first to collect a cookie, and then visit cash-out sites to earn higher retargeting CPMs. The FouAnalytics data below shows evidence of this. Note the retargeting and remarketing rows in the data grids below. They have the highest amount of dark red (bots) compared to the other campaign lines. If you do remarketing campaigns (showing ads to users that have purchased from your before), note that fraudsters are not falsifying the purchases. Those did occur. But they are falsely claiming credit for them -- i.e. claiming that their remarketing campaigns caused those sales, when they did not actually have anything to do with those sales.

See this article by Keri Thomas with one such case of remarketing fraud -- the vendor wrote false data into their own customers' Google Analytics to make it appear that they drove a lot of ecommerce sales.

If you run programmatic display or video ads

Be sure to use FouAnalytics in-ad tags to measure where your ads actually went and whether bots loaded your ads. Placement reports and even log level data contain errors because they record the domain or app passed in the bid request. DSPs and ad servers do not run postbid javascript tags to detect where the ad actually went. So if breitbart or any other fraudulent site lied and put someone else's domain in the bid request that is what you will see in the log level data. The fact that you don't see breitbart or fakesite123 in the log level data is because they lied and put some other domain in the bid request. By running a postbid javascript you can actually see where your ads went. You will also see all the forms of fraud used to load your ad and steal your budgets. Once you see this, you can block bad sites and apps, or remove them from your inclusion lists. Even if you buy from PMPs (private marketplaces) you need to measure whether the sellers in that PMP are cheating (and using bot traffic, chumbox traffic, etc.)

More details: How Advertisers use FouAnalytics to Optimize Digital Campaigns

This is how you use FouAnalytics, so you can "see Fou yourself" what you couldn't see when using legacy fraud verification vendors' tech. Time to upgrade your analytics.

See: Advertisers Replace DoubleVerify, IAS, and Moat with FouAnalytics

If you know someone who would benefit from using FouAnalytics to measure their campaigns, feel free to forward this to them.