What Happened?
Last month, BuzzFeed news shared some scary findings for digital marketers around the globe. Their research revealed that some of the world’s most influential brands’ websites have fallen victim to large-scale digital fraud. Including this incident, JPMorgan Chase CMO Kristin Lemkau’s predictions for this year’s total loss to ad fraud are $16.4 billion, which is more than double the loss seen in 2016. However, other sources predict this year will see a decrease in total ad fraud loss by 10%, though still a substantial amount.
Ad fraud is a rapid crisis in the advertising world, and investigations like these are showing how seemingly credible players in ad supply are potentially active, and profiting, from digital deception.
…but this is not new. At Obility, we’ve dealt with poor quality display placements since we started – but with this experience, we also know how to overcome it – which we’ll share later in this blog (feel free to skip ahead to the Display Tactics section).
Ad Fraud History
The issue at hand is that the internet has become the modern day “Wild West”; the opportunities are everywhere – good and bad. Similar to the BuzzFeed report, reports of similar fraud date back to 2013, when Mike Shields warned about ghost site creation and ad-selling on them via programmatic channels, with “human visitor” bot simulation to rapidly increase views. Revisiting the topic in response to Buzzfeed’s investigation, Mike theorizes that some of the blame for this ongoing problem can be found through Chris Anderson’s Long Tail concept – a way that some in the ad industry continue to target people with relevant ads for cheap. We’ve all heard about the Facebook scandal.
Agencies earning revenue from these websites’ fraudulent views are claiming to be unaware of such happenings, as pressure builds to answer questions about this digital pandemic. While there are 5,000 ad tech companies out there claiming they can help solve the problem, we wonder – where are those examples?
The experts of Obility have discussed the causes and effects of this issue, and now we’re sharing how we can, and do, effectively combat these scams and disadvantages for our clients.
How Does Obility Stop This?
At the end of the day, there’s no excuse for allowing display ads to show on poor placements – and especially fraudulent websites. Just like we monitor search query reports & add negative keywords with our “search” campaigns, there’s both pro-active and re-active work that needs to be done to properly manage display campaigns. As marketers, we also need to use our brains, and not simply buy into buzzwords like “predictive” or “programmatic” and assume we can set our campaigns and forget them. Lastly, we need to ensure that we’re properly reporting on our display campaigns. Are we driving positive ROI? If not, there’s good reason to perform an audit.
So, if you’re worried about the quality of your display campaigns, there’s 4 areas that your marketing team or agency must focus on:
- Have the right mindset (Marketers should question everything)
- Be Pro-Active
- Be Re-Active
- Report, Report, Report
Let me explain…
Marketing Mindset
It makes me sick to think that there are marketers who think that just because they bought platform “X” that they don’t have to manage their campaigns anymore.
By default, a great marketer will ask questions, put themselves in the prospects’ shoes, go left when everybody else is going right. This is true for not only addressing prospects, but also tools. Just because a software is “AI-driven”, “programmatic” or “predictive” doesn’t mean that it will work. As a marketer, we need to also question the tools that we use, test them and come up with our own conclusion.
Programmatic and predictive tools are awesome, and I can’t wait to see them mature, but that trust has not yet been built…obviously. Until that does happen, we must be and do the following things.
Be Pro-Active
Just like we don’t launch search campaigns without negative keywords (right?), pro-active initiatives need to be implemented with display campaigns – and this is where I feel many people drop the ball.
Since we know our target audience we know many of their preferences; websites they don’t like to visit, topics that upset them and interests that they simply don’t care about. We also know certain websites that they visit when they’re in the “business” mindset, as well as websites they frequent when they’re looking to get away from business (Facebook or game website).
What we may or may not know (apparently most of us don’t…) is that we can both target, or more important for this conversation, add exclusions based on these (1) websites, (2) topics and (3) interests.
What this means is we can exclude the websites that we know our brands don’t want to be associated with, and ensure that our ads don’t get placed on poor pages by adding topic / interest exclusions that we know our prospects don’t like.
Website Placement Exclusions
At Obility, we’ve managed hundreds of B2B companies and thus have been able to compile a long list of website placements that work well for B2B businesses in the Tech space (websites that lead to qualified leads). Similarly, we have been able to identify a list of websites that perform poorly for our clients. We exclude these placements for every display campaign we launch. You should have a list of websites that don’t work for your brand, and also exclude them from the get-go.
*Reach out for our “Display Whitelist” for top tech website placements, or exclusions list.
Website Topics / Site Category Exclusions
If Hatchbuck wanted to avoid their ad being shown next to Louis CK’s sexual assault article, all they needed to do was add a few site category exclusions: “Sexually Suggestive” + “Sensational & Shocking” topics – maybe even “mature audiences”.
But they shouldn’t stop there, usually we launch our display campaigns excluding this type of topics (they didn’t):
- Sexually suggestive
- Sensational & shocking
- Profanity & rough language
- Military & international conflict
- Tragedy and conflict
- Crime, police, & emergency
- In-game
- Mature audiences
Website Interests Exclusions
Admittedly, many digital marketers are familiar with site category exclusions, and even more are familiar with negative website placements, but less are familiar with “Interest-based” exclusions.
Interest-based exclusions allow us to get even more specific. This is where marketers who know what they’re doing hang out.
Instead of just negating politics, we have the ability to exclude things like “Far-Right” or “Far-Left” views, subjects like “Birth-Control” and even stranger [Things] areas like “Occult & Paranormal”. “Erectile Dysfunction” is probably an area most of us would like to stay away from as well. At least for work purposes.
Here is our go-to exclusion list:
- Christian & Gospel Music
- Occult & Paranormal
- Right-Wing Politics
- African-Americans
- Left-Wing Politics
- Online Games
- Celebrities & Entertainment News
- Ethnic & Identity Groups
- Dating & Personals
- Campaigns & Elections
- Birth Control
- Religion & Belief
- Reproductive Health
- Seniors & Retirement
- Government
- Erectile Dysfunction
- East Asians & Diaspora
- South Asians & Diaspora
- Indigenous Peoples
- Corporate & Financial Crime
- Discrimination & Identity Relations
- Human Rights & Liberties
- Scandals & Investigations
- TV Talk Shows
- Political Humor
- Opinion & Commentary
- Political Polls & Surveys
By being pro-active with the above tactics, we restrict the possibilities of our ads showing on questionable placements, and it increase the likelihood that we (a) get in front of a qualified prospect and (b) get in front of them when they’re on a relevant (tech) website.
However, being pro-active will only lessen the risk of showing banners on poor placements. We still run the risk of showing up next to Trump or Louis CK if we’re not Re-Active as well.
Be Re-Active
This is the easy part, and maybe that’s why it’s been ignored. Maybe these programmatic and predictive websites make being reactive difficult. I doubt it though. I think we’re just lazy.
On a weekly, and definitely on a monthly basis, marketers need to be analyzing (manually!) the placements that their ads are being displayed on. You might be thinking this is a lot of work, and it can be, but the truth is that you don’t need to go through every single placement.
I like to prioritize the placements that are gathering the most impressions, so that my limited time is spent most effectively. I know that I can’t go through every placement, so I start with the most important (where I’m spending the most money), and go from there.
I manually comb through the placements, often visiting sites I’m not familiar with which allows me to gauge relevance. I’ll also identify outliers when looking at click-through rates, as placements with very high or low click-through rates are often suspect.
Additionally, if something is too good to be true, it often is. On numerous occasions, we’ve seen sites like Wikihow.com drive conversions in volume, then later find out that all of these leads were garbage – we now exclude these sites. This brings me to the final recommendation.
Reporting
At the end of the day, clear reporting will tell you whether or not you should continue to push money into your display campaigns. With clear reporting, you only have to ask yourself:
Are my campaigns reaching their goals?
Or more specifically, we might ask “are my campaigns driving more revenue than they cost?”. If we’re pushing awareness campaigns we might ask “are my campaigns driving clicks that have above average bounce rates?”
I’m not trying to “victim blame”, but with the number of companies claiming victim to ad fraud, I have to put some of the blame on these companies, as they clearly were not analyzing their display ads’ performance.
Take Away
Ad fraud is upon us, and we will likely never be able to eliminate it completely, but we can take both pro-active and re-active steps to mitigate the intrusion. You’ve been been both armored and armed, and now you’re ready for the great unknown that is the Display Network.
Note: These strategies are geared towards the Google Display Network (GDN), however these strategies can be applied to any ad platform.