When you see 90% fraud, what do you do?

When you see 90% fraud, what do you do?

If you were the marketer in the following case, what would you do?

For months (or perhaps years), your legacy fraud verification vendor reported 1% IVT ("invalid traffic") and you and your agency thought everything was fine, so you continued to pour money into these campaigns.

When we did the no-cost pilot and placed a FouAnalytics in-ad tag into an ongoing campaign, we found the following. Eyeballing it, it looks like 80 - 90% dark red, across the board.

What do you do? What do your instincts tell you? If the marketer didn't fall out of their chair already, their first instinct is to cover it up. They don't want their boss to find out, so they try to bury the evidence. Is that what you would do too?

What if I told you, you don't need to be embarrassed? You don't need to panic. And don't worry about the 90% dark red that we found with FouAnalytics? Is Dr. Fou out of his mind? (Yes, he's been out of his mind for the last 10 years already.)


A path forward, a process to clean your campaigns.

Why would I say "don't worry about it" when the dark red is clearly 80 - 90%? It's because NOW you can see clearly; and there is a rational path forward. There are even concrete and discrete steps you can take to clean the campaign, that you didn't see before when the legacy fraud verification vendors just told you a number, like 1%. Now, we obviously cannot chop off 90% of the volume of impressions overnight. You won't be able to spend all your budgets, and CPM prices would go through the roof. Nobody wants that; and no one can explain that to their bosses.

Instead, we put in place a process to clean the campaign progressively. How do we do that? We pick off the worst offenders first. In the FouAnalytics dashboard, we surface for you the top 10 bad sites and apps so you can review them and understand why they were labeled bad. That means, they use bot traffic to juice their numbers, or any one of 3 dozen other forms of fraud that inflate their revenues artificially -- like ad stacking, sticking 3,000 ads on a page, popunders, forced redirects, ad slot refreshing every 1 second, etc, etc, etc.

If you understand why they are a bad site, you can choose to turn them off by adding them to your blocklist. You simply check the checkbox next to the domain or app name, and those get added to the boxes at the top of the page. This makes it convenient for you to copy and paste the list to send to your agency to turn off -- add to block list. Note that you don't have to justify to your media agency why you want to turn of any site or app. They are your agent and SHOULD do what you tell them to do, right?

Some advertisers expect to see a large drop in dark red when bad sites and apps are added to the block list. Yes, we do see that in some cases, like the one below where the dark red was cut in half -- from 48% to 23%. But, depending on what you are buying, we may or may not see as drastic a reduction in dark red. And that is OK. We are still blocking the worse offending bad guys so more of our ad budgets can flow to better sites and apps.

Be aware that adding bad sites and apps to the block list will result in lower volumes (fewer ads to buy). You can clearly see that in the green volume bars above. They dropped significantly after the change was made on April 26.

You are buying fewer ads, so you can also raise your CPM bids. This sets you on the path to buying better ads from better publishers. Real publishers with real human audiences can't afford to sell ads at sub-$3 CPMs. If you are seeing those CPMs in programmatic channels, those are likely to be spoofed (bad guys pretending to be good publisher domains).


You were limited by the tools before, time to upgrade your tools and do better

I will wrap this right here. You don't have to panic or worry if we find 90% dark red in your campaigns when we do a no-cost pilot with FouAnalytics. That's because you honestly didn't know. The tools you were using -- the legacy fraud verification vendors -- failed you. They failed to detect the fraud and protect you from fraudsters.

See: They didn't even measure it, and told you there's no fraud

It is honestly not your fault for not taking action before. After all they kept telling you fraud was 1%. If it's that low, why would you need to do anything, other than keep spending? With better analytics for your digital media, you can see better, so you can do better. And this is not just about fraud. By simply having better analytics for your digital media, you can see that most of your ads were used up between midnight and 2 am, and you had none left to serve for the rest of the day when humans are awake and going online. You won't see this if you're just getting monthly reports, in excel spreadsheets.


Whatcha gonna do?

Are you going to cover up? or are you going to chin up and take action to make your digital campaigns better than they ever were?

You take the first step. I am here to help, as are a group of seasoned FouAnalytics practitioners who have looked at data for years and can advise you on what moves to make. Reach out, and I will put you in touch.

Have a good long weekend. We'll pick this up next week after the break.











Ryan Purkey

Managing Consultant @ LexiTech Consulting & rQuadrant Limited

1 年

Or from the Analytics side, "Why are our Display campaigns many orders of magnitude lower than any other type of traffic coming to our site?"

回复
Marc Dhalluin

Would Your BRAND ask you to LEAD it NO MATTER What? THINK. Carefully.

1 年

Spero Patricios you may be interested

Dominic T.

Senior Data Science-Marketing Professional

1 年

Great article and solid recommendations on the process side! ?? (But I am now hearing the Cops TV intro!) ??

回复

要查看或添加评论,请登录

Dr. Augustine Fou的更多文章

社区洞察

其他会员也浏览了