"This is the biggest advertising story of the decade, and it's being buried."
So wrote Ad Contrarian Bob Hoffman, the retired CEO and chairman of Hoffman/Lewis Advertising, in June 2013 on a $7.5 billion scandal that has been developing under the digital radar in the advertising world for the past few years. The three main allegations, according to those who are making them:
- Half or more of the paid online display advertisements that ad networks, media buyers, and ad agencies have knowingly been selling to clients over the years have never appeared in front of live human beings.
- Agencies have been receiving kickbacks and indirect payments from ad networks under the guise of "volume discounts" for serving as the middlemen between the networks and the clients who were knowingly sold the fraudulent ad impressions.
- Ad networks knowingly sell bot traffic to publishers and publishers knowingly buy the bot traffic because the resulting ad impressions earn both of them money—at the expense of the clients who are paying for the impressions.
These charges have not seen much discussion within the online marketing community. But the allegations have the potential to affect everyone involved in online advertising—ad agencies, in-house departments, agency and in-house digital marketers, online publishers, media buyers, and ad networks. An entire industry—billions of dollars and thousands of jobs—is at stake.
And it all starts with a single impression.
The impression that you make
(Online advertising is based on an "impression"—without the impression, then an advertisement cannot be viewed or clicked or provoke any other engagement. The Internet Advertising Bureau, which was founded in 1996 and "recommends standards and practices and fields critical research on interactive advertising," defines "impression" in this manner:
a measurement of responses from an ad delivery system to an ad request from the user's browser
In another words, an "impression" occurs whenever one machine (an ad network) answers a request from another machine (a browser). (For reference, you can see my definition and example of a "request" in a prior Moz essay on log analytics and technical SEO.) Just in case it's not obvious: Human beings and human eyeballs have nothing to do with it. If your advertising data states than a display ad campaign had 500,000 impressions, then that means that the ad network served a browser 500,000 times—and nothing more. Digital marketers may tell their bosses and clients that "impression" is jargon for one person seeing an advertisement one time, but that statement is not accurate.
The impression that you don't make
(Just because a server answers a browser request for an advertisement does not mean that the person using the browser will see it. According to Reid Tatoris at MediaPost, there are three things that get in the way:
- Broken Ads—This is a server not loading an ad or loading the wrong one by mistake. Tatoris writes that these mistakes occur roughly 15% of the time.
- Bot Traffic—Whenever hackers write these automated computer programs to visit websites and post spam or create fake accounts, each visit is a pageview that results an an ad impression. According to a December 2013 report in The Atlantic, 60% of Internet traffic consists of bots.
- Alleged Fraud—In Tatoris' words, "People will hide ads behind other ads, spoof their domain to trick ad networks into serving higher-paying ads on their site, and purposefully send bots to a site to drive up impressions." Noam Schwartz described in TechCrunch two additional methods of alleged fraud: compressing ads into a tiny one-by-one pixels that are impossible to see and using malware to send people to websites they never planned to visit and thereby generate ad impressions. AdWeek found in October 2013 that 25% of online ad impressions are allegedly fraudulent.
Tatoris crunches all the numbers:
We start with the notion that only 15% of impressions ever have the possibility to be seen by a real person. Then, factor in that 54% of ads are not viewable (and we already discussed how flawed that metric is), and you're left with only 8% of impressions that have the opportunity to be seen by a real person. Let me clarify: That does not mean that 8% of impressions are seen. That means only 8% have the chance to be seen. That's an unbelievable amount of waste in an industry where metrics are a major selling point.
Essentially: If you have an online display ad budget of $100,000, then only $8,000 of that ad spend has the chance to put advertisements in front of human eyeballs. (And that's not even taking into account the poor clickthrough rates of display ads when people do see them.)
If you are paying $0.10 per impression, then the $10,000 that you will pay for 100,000 impressions will result in only 8,000 human views—meaning that the effective CPI will actually be $1.25.
How bot traffic affects online ads
(In a Digiday article, Jack Marshall interviewed a supposedly reformed fake web traffic buyer on how the scheme works. Here are three excerpts:
How and why were you buying non-human traffic?
We were spending anywhere from $10,000 to $35,000 a day on traffic. My conversations with [these ad networks] were similar: They would let me decide how much I was willing to pay for traffic, and when I told them $0.002 or below, they made it clear they had little control over the quality of traffic they would send at that price. Quality didn't really matter to us, though. As a website running an arbitrage model, all that mattered was profit, and for every $0.002 visit we were buying, we were making between $0.0025 and $0.004 selling display ads through networks and exchanges. The biggest determinate of which traffic partner we were spending the most money with was pageviews per visit. Since we were paying a fixed cost per visit, more pageviews equaled more ad impressions. Almost none of these companies were based in the U.S. While our contacts were in the US and had American names and accents, most of the time we found ourselves sending payment to a non-US bank.
In other words, the publisher would allegedly pay an ad network $0.0020 for a visit from a bot, and the resulting ad impression would garner $0.0025 to $0.0040 in revenue—that's a gross margin of 25% to 100% for the publisher for doing nothing! It's no wonder that so many websites around the world may be allegedly involved in this practice.
Do you think publishers know when they're buying fake traffic?
Publishers know. They might say "we had no idea" and blame it on their traffic acquisition vendor, but that's bullshit, and they know it. If you're buying visits for less than a penny, there's no way you don't understand what's going on. Any publisher that's smart enough understand an arbitrage opportunity is smart enough to understand that if it was a legitimate strategy that the opportunity would eventually disappear as more buyers crowded in. What we were doing was 100 percent intentional. Some articles revolving around bot traffic paint publishers as rubes who were duped into buying bad traffic by shady bot owners. Rather, I believe publishers are willing to do anything to make their economics work.
Do networks, exchanges and other ad tech companies do anything to stop this from happening?
We worked with a major supply-side platform partner that was just wink wink, nudge nudge about it. They asked us to explain why almost all of our traffic came from one operating system and the majority had all the same user-agent string. There was nothing I could really say to answer that question. It was their way of letting us know that they understood what was going on. It wasn't just our account rep, either. It was people at the highest levels in the company. Part of me wished they'd said "You are in violation of our TOS and you have to stop running our tags." I would have been happy with that. But they didn't; they were willing to take the money.
If these stories are true, then ad networks do not care that the impressions are from bot traffic and publishers do not care that are getting bot traffic because they are both making money. Who gets hurt? The companies advertising their products and services.
The worst part of it all
(It's not only that online display ads are alleged to be amazingly useless and that many publishers and ad networks are allegedly involved in sleazy deals. A March 2015 investigative report in Ad Age found the following:
Kickback payments tied to U.S. media-agency deals are real and on the rise, according to Ad Age interviews with more than a dozen current and former media-agency executives, marketers' auditors, media sellers and ad-tech vendors who said they'd either participated in such arrangements or had seen evidence of them. The murky practice—sometimes disguised as (undisclosed) "rebates" or bills for bogus services—is being motivated by shrinking agency fees and fueled by an increasingly convoluted and global digital marketplace. "It's really ugly and crooked," said one ad-tech executive who described receiving such requests.
Some arrangements go like this: A large media shop, poised to spend $1 million with that ad-tech executive's firm to buy digital ads last year, asked for $200,000 to be routed back to the agency's corporate sibling in Europe. The $200,000 would pay for a presentation or presentations by the sibling's consultants. But these types of presentations aren't worth a fraction of the price tag, according to numerous executives dealing with the same issue, who spoke on condition of anonymity for fear of losing business.
Essentially, here is what is allegedly happening:
- Clients give money to agencies to purchase online display advertising
- The agencies give the money to the ad networks
- The ad networks give a portion of the money back to the agencies
- The clients' display ads are only 8% viewable
- The 92% non-viewable impressions still earn money for publishers and ad networks
I think we can see who the loser is—everyone is making money except for the clients.
During the same month as the Ad Age report, former Mediacom CEO Jon Mandel reportedly told the Association of National Advertisers Media Leadership Conference that widespread "media agency rebates and kickbacks" were the reason that he left the agency business.
Heads in the digital sand
(I have yet to hear about this issue being addressed in any talk, panel, or session at a digital marketing, martech, or adtech conference. Prior to today, I have seen only one article each in two major publications in the online marketing industry. (Mozzers, please correct me if I am mistaken and have missed something major on this topic.)
Why is no one talking about this?
No marketing agency wants clients to know that 92% of its display advertising spend is wasted. No advertising manager wants the CMO to know that only 8% of the company's ads are reaching people at 100% cost. No CMO wants the CEO to know that 92% of the entire ad budget is being flushed down the digital toilet.
I myself would probably have not been permitted to write this article when I held various agency positions in the past because I managed clients' online advertising and some PR and digital marketing clients of the agencies were advertising networks themselves.
(Today, I am the director of marcom for Logz.io, a log analytics startup, and I have the luxury of being accountable only for the results of my in-house work—and I do not plan to use online advertising anytime soon. Still, I was a journalist in my first career years ago, and I wanted to write this report because I think everyone in my beloved industry should know about this explosive issue.)
Hoffman, the retired ad agency CEO who I quoted at the beginning, puts it better than I can:
How does an agency answer a client who asks, "You mean more than half the money you were supposed to be custodian of was embezzled from me and you knew nothing about it?" How does an ad network answer, "You mean all those clicks and eyeballs you promised me never existed, and you knew nothing about it?" How does a CMO answer his management when they ask, "You mean these people screwed us out of hundreds of thousands (millions?) of dollars in banner ads and you had no idea what you were buying?"
Everyone is in jeopardy and everyone is in "protect" mode. Everyone wants to maintain deniability. Nobody wants to know too much. If display advertising were to suffer the disgrace it deserves, imagine the fallout. Imagine the damage to Facebook, which at last report gets over 80% of its revenue from display. Imagine the damage to online publishers whose bogus, inflated numbers probably constitute their margin of profit.
If the comScore findings are correct and projectable, it means that of the 14 billion dollars spent on display advertising last year in America, 7.5 billion was worthless and constituted some degree of fraud or misrepresentation.
But clients, CMOs, and CEOs are going to read one of these articles one day and start asking uncomfortable questions. I would suggest that Mozzers—as well as all digital marketers and advertisers—start thinking about responses now.
Responses to the scandal
(Google, to its credit, has disclosed that 56% of its digital ad impressions are never actually seen—of course, the report was also released with the announcement of a new ad-viewability product.
Ginny Marvin summarizes at Marketing Land:
Google's viewability measurement tool, Active View, is integrated into both the Google Display Network and DoubleClick. Advertisers can monitor viewability rates and buy ads on a viewable impression basis rather than by served impressions.
Google also announced an update to DoubleClick Verification last week, which includes viewability monitoring, ad blocking, a content ratings system and spam filtering capabilities.
The goals of the Media Rating Council (MRC), an industry organization founded in the United States in the 1960s following congressional hearings into the media industry, are:
- To secure for the media industry and related users audience measurement services that are valid, reliable and effective
- To evolve and determine minimum disclosure and ethical criteria for media audience measurement services
- To provide and administer an audit system designed to inform users as to whether such audience measurements are conducted in conformance with the criteria and procedures developed
The MRC has certified "viewable impressions" as a legitimate metric (as opposed to "served impressions"). The Interactive Advertising Bureau (IAB), mentioned earlier, issued guidelines in December that online advertising networks should aim for at least 70% viewability.
Facebook, for its part, announced in February 2015:
We are working with the MRC and a consortium of advertisers and agencies to develop more robust standards for viewable impressions. Our goal is to work with the MRC, our partners, and industry leaders around the world to help apply further standards for feed-based websites like Facebook, mobile media and new ad formats.
The American Association of Advertising Agencies, Association of National Advertisers, and IAB announced last year that they would create a new organization, the Trustworthy Accountability Group, to fight problems in the online advertising market and do the following:
- Eliminate fraudulent traffic
- Combat malware
- Fight Internet piracy
- Promote greater transparency
TAG now consists of representatives from Mondelez International, JCPenney, Omnicom, Motorola, Google, Facebook, AOL, and Brightroll.
Canada's latest anti-spam legislation aims to fight Internet malware and bots—but a big stumbling block is that most of the problem comes from outside the country.
Will these corporate and organizational responses be enough? For the following reasons and more, it's impossible to know:
- Industry guidelines depend on voluntary compliance. Industry recommendations do not have the force of law—any business that thinks it can still make a lot of money by ignoring the guidelines will likely continue to do so.
- Possible penalties for past behavior. Regardless of what reforms may occur in the future, should those who knowingly engaged in such alleged fraud and deception in the past be held criminally or civilly liable? (I'm not a lawyer, so I cannot comment on that.)
- IAB's 70% viewability goal. Should advertisers accept this metric as simply the nature of the medium? One estimate of the total display ad market amounted to $14 billion. If the 70% viewability goal can even be reached, should and will advertisers accept that $4.2 billion of their collective ad spend will still be lost before their advertisements are even viewed by human beings?
I have no answer—only time, I suppose, will tell.
But others are coming up with their own answers—those large corporations that are spending billions of dollars a year on online display advertising. As Lara O'Reilly wrote in May 2015 at Business Insider, $25 billion in ad spend is now under review in what Adweek is calling "Mediapalooza 2015." O'Reilly gives one possible reason:
Media reviews let brands reassess their ad spending, often by offering those contracts out in a competitive bidding process. The companies include General Mills, Procter & Gamble, Volkswagen, Visa, Sony, Coca-Cola, Citi, 21st Century Fox ... the list goes on. Some of these — P&G, Sony, and 21st Century Fox — spend more than $1 billion on advertising each year...
It could be that marketers are finally getting fed up with the apparent lack of transparency about where their budgets are actually being spent and why.
What marketers can do
(Image of an Indian online-marketing team I used with rights in a prior Moz essay
on the future of marketing departments)
Regardless of what the future will hold, here are my recommendations on how digital advertisers can respond:
- Stop doing cost-per-impression (CPI or CPM) campaigns. Traditional digital advertising strategy recommends that people use CPM campaigns for brand awareness, cost-per-click (CPC) campaigns for traffic, and cost-per-action (CPA) campaigns for sales and conversions. In light of this scandal, I see no good reason to do CPM at all anymore.
- Revise advertising KPIs and metrics in human terms. Earlier in this article, I calculated the following change to a hypothetical CPI value: "If you are paying $0.10 per impression, then the $10,000 that you will pay for 100,000 impressions will result in only 8,000 human views—meaning that the effective CPI rate will actually be $1.25." In addition, half of all clicks in CPC campaigns might also be bots. As a result, a $2 CPC result may actually be $4 when reaching human beings is taken into account. Ad campaign analysts may want to take alleged bot and fraudulent activity into account when calculating ROI and whether display advertising is worthwhile.
- Demand full disclosure. Clients should ask agencies and media buyers if they are getting paid directly or indirectly by ad networks. Agencies and media buyers should ad networks how they are combating bot activity and any fraudulent behavior. Ad networks should not turn a digital blind-eye to publishers who intentionally use bots to make profits off of advertisers. If anyone gives vague answers or otherwise disparages such questions, then that is a red flag. Advertisers should demand and receive full, verifiable information in light of what has allegedly been occurring.
- Block certain countries from campaigns. According to a report in Ad Week, China, Venezuela, Ukraine, and Singapore have "suspicious traffic" rates of between 86% and 92%. (The rate in the United States is 43%.)
- Use ad-fraud detection platforms. Companies such as Forensiq, SimilarWeb, Spider.io (which was bought by Google), Telemetry, and White Ops compare visit patterns with industry benchmark behavior as well as check for malicious software proxy unmasking, verify devices, and detect any manipulation.
- Run manual campaigns as much as possible. The only way to reduce wasted impressions significantly is to research and implement digital ad campaigns manually rather than use programmatic ad buying. Digital advertisers should research potential websites on which they want to run advertisements to see if they are legitimate—potentially even running ads on only the largest, well-known sites but doing so continuously. This way, it might be best to focus your ad campaigns on quality viewers rather than trying to maximize the quantity of viewers by also including lesser-known sites.
Beyond the current responses of the ad industry and my present recommendations for marketers, I do not know what will happen. My goal here is simply to explain to digital marketers what has allegedly been occurring. What the future will hold—well, that's up to we marketers and advertisers.
Additional resources
- The Ad Fraud Wiki
- Bob Hoffman's blog post with a partial list of his writings on this topic
- Ad Week's Mike Shields, and Ad Age's Alex Kantrowitz—two of the reporters who are following this issue closely
Bob Hoffman has been a lone (and sane) voice on this topic for years. But he himself has pointed out part of the reason why no one wants to talk about it - namely, just about everybody involved stands to gain from the status quo. The only people being rooked here are the advertisers (and by implication every customer of those advertisers = because we as consumers ultimately foot the bill for inefficient brand marketing expenditure).
Things are only going to change when advertisers en masse demand greater transparency and rigor in the process. Bob Hoffman has already pointed out the huge amount of media accounts currently in review ($25bn at last estimate).
Perhaps Kraft helped lead the way, rejecting 85pc of the digital ad inventory they were offered because it was "crap" (technical term):
I've blogged previously my take on the digital ad controversy.
And it isn't just digital display advertising. Social media also suffers to an extent from fakery.
And let's not get started on the plague of referral spam which is playing havoc with site traffic data at the moment.
In summary, great article. We need more people like you and Bob Hoffman bringing these issues out into the open. Assuming we are genuinely interested in making real things happen in the real world with real people.
Hello, Mozzers! I'm the author of the article and will be happy to answer any questions and clarify any statements. Feel free to post comments, and I'll respond as soon as I can!
Is this likely to also be leaking over into organic? i.e. If I look at my webmaster tools is the impressions/clicks count being driven up by bots?
That's a great question! To be honest, I have only explored the paid side of the issue. I looked into the organic side, and here's what I found:
- Google filters out known bots -- but which ones are known and unknown seems not to be publicly available. Plus, one obvious question is, "How long does it take for a new bot to become 'known' by Google?"
- Moz has a great post here on filtering the relatively-new headache of referral spam
I'm afraid that's all I know at the moment about questionable organic and referral traffic. This essay just focused on paid. If any Mozzers can address this more in-depth, please feel free! :)
I see crap like this all the time.
4webmasters(.)org
buttons-for-your-website(.)com
best-seo-offer(.)com
buttons-for-your-website(.)com
free-share-buttons(.)com
There are a few premade custom reports you can use to filter those out on Google, or just do it yourself (remove those spam referrers)
Yes, this does leak over into organic. Usually you can spot it in analytics as suspicious. In referral traffic you should be able to tell which sessions are legitimate.
You can resolve this by blocking the spam IPs in your .htaccess file and checking that your robots.txt file is disallowing known spam bots. It's also a good idea set up a domain exclusion list in analytics.
Yes... you can block them. But in all reality, you can't block them. We end up adding a new spammer domain to our list daily. On most days, there are 2 or 3 new ones.
That's a crap ton of time editing files considering we have to do it for nearly 100 .com's every time a new spammer shows up and starts screwing our data. Not really a plausible solution if you're running more than one or two websites.
"...in all reality, you can't block them."
Yep, particularly if you have dozens or hundreds of GA accounts to monitor. The only viable options are relying on advanced segments, switching analytics providers, or waiting for GA to come up with a fix.
Good Point. But do other analytics providers really have this problem solved?
Hi Samuel, thanks for this enlightening article. Are chances of fraudulent traffic same with both search network ad and display network ad?
Almost all of the research I have found has focused on display network ads (including but not limited to the Google Display Network), which can be priced in CPM and not only CPC and for whom fake impressions are a big issue.
I've seen a lot less criticism of search-based ads (those that appear next to organic SERP listings), which are typically priced on a CPC basis. However, there is a long-running issue of fake clicks (not only fake impressions) that ad networks have tried to address.
Someone submitted this article to the news aggregator Fark.com (which has a geeky user base). The comments are interesting. Here are just a few:
That high [reference to the 8% viewability figure]? When I was using Adwords, I was billed for impressions on one site in one month that exceeded the total number of views for the site by a factor of 100. Also, had zero click-throughs registered on my site from that site, Adwords said there were over 1000.
Then:
I'm in the Ad business and clients have become so cheap with this stuff anyway. There is no budget for online ads anymore. I do see some of what our interactive dept does online a lot though, but I'm sure our media-buy partners pull this scheme all the time. Thats why they occupy 10 floors of a 5th avenue high-rise.
Then:
4 years ago Google gave me back 80% of my adwords budget due to fake clicks. Just a few months ago I started up again. Everything was fine for about 3 months and i regularly got referals for a whole $30 a month in adword charges. Was paying very little for my ads as my target area is super small. Then out of the blue I got socked with several hundred dollars in clicks.
The adwords were working but when I go from paying $30 a month to $500 a month with no changes on my part AND get zero leads out of it then it's utter bullshiat...again. This time I got no reimbursement.
It's interesting how Google has repeatedly carried out exercises and pay-backs that essentially prove that there's a massive, massive problem with bots and fake clicks on the GDN. If Google is aware of this (and they obviously are) and if it's becoming such a popular phenomenon, I'm surprised that there aren't already a lot more stringent measures in place for the measurement of clicks and weeding out fake clicks. Why should people have to go and report these fake clicks to Google rather than Google actually weeding them out automatically?
Right on! I've spoken publically at conferences around some of the issues around display. The industry is definitely going the right way with the likes of Active View but we're still not where we need to be yet. Not least because what actually constitutes "viewability" is still a moving feast. Hopefully MRC/IAB can bang the drum to support the push for regulation here.
I'm not convinced about display & CPA bidding. Mainly because the "A" bit doesn't sit right with me. Something else I talk about whenever someone will listen.
The whole problem is the assumption that correlation means causality or "post hoc ergo propter hoc" (after this, therefore because of this). It's such a dangerous assumption to make.
This has been perpetuated further by measuring conversions on a "post-view" or "post-impression" basis.
Given the fact that ads are everywhere online it is safe to assume that at some point it is highly likely that someone who might become a customer over the next 30 days could have been exposed to a display ad at some point. What that doesn't mean (as the OP has so eloquently explained) is that the user has seen the ads nor does it mean they were understood or relevant. However, using post-impression, media agencies & media owners would have clients increasing their budgets on the basis of display ads being an "influencer".
I am absolutely of the mind that display advertising can be an incredibly powerful tool. Some of the innovations in the UK market in particular are astonishing. However, the way we talk about it as an industry needs to change. Display isn't going away so we need to be the ones that help navigate the many pitfalls the uneducated client may be drawn into. That's ultimately what clients pay us to do.
On LinkedIn over the past few weeks I've seen several white papers being pushed out by media owners & agencies around display delivering ROI. One of them went as far to say their product performs better than search. In every single instance I've spotted I have asked 2 questions:
In every single instance, I've been ignored. Very frustrating as I treat client budgets as if they were my own. I give honest answers even if it means not securing additional business. regardless of the short-term loss, I know that those clients will happily ask me to advise on future projects as they know I'm not out to make a "fast buck".
For an advertiser, it seems like the old adage is true:
40% of something is better than 100% of nothing.
Thanks for sharing.
But what if you should be getting 80% of something if the impressions that you paid for would actually be seen by human beings?
I'm not trying to defend this practice in the least, but I think there's a little bit of mixed metrics going on here. Ultimately, the clients don't care about impressions. Oh, sure, their in-house marketing team might make a lot of noise about wanting to see all the numbers and dig through the data, but that's mostly so that they can keep justifying their own paychecks. What the clients care about is if you can make a reasonable case that they are making mmoney - more specifically, are you making them more money than they would have made without you. After that, everything else is just window dressing.
So when you say "What if it should be 80% of something?", that's the wrong question to be asking, because they aren't really paying you for impressions. They're paying for revenue. Period. So if they are getting returns that they are satisfied with, they aren't losing anything. You can argue that they are being misled and manipulated, and they are, but ultimately the only real manipulation is that they are paying more per impression than they think they are, and ultimately if you're reporting the CPI in any meaningful fashion, you're doing reporting wrong. Fake traffic and non-existant impressions don't affect ROI, revenue lift, brand mentions, or whatever bottom line metric you SHOULD be reporting. The only reason this is a story is because of how many agencies still use intermediate metrics for anything but internal optimization, and the REAL scandal is that anyone actually cares about raw CPI.
+1
Exactly, very interesting read but also much ado about nothing. So the ad networks get rid of the fraud and then raise prices to cover lost revenue for the publishers? In my view the CPI isn't that important.
I may only be willing to pay 0.002 on your fraud laden network because it doesn't convert for me. Get rid of 90% of the fraud and I'll be willing to pay 0.02, but either way the amount I'm spending and value it brings my company doesn't change. Am I missing something?
Ultimately, the clients don't care about impressions... What the clients care about is if you can make a reasonable case that they are making money
Of course, that is true. But what about when the clients realize that they could make even more money and have greater ROI if more of their impressions were actually seen by human beings?
I have always believed that paid click advertising across all platforms were fishy.
-Chenzo
This supports our study we did with ad fraud 88%+
Over 88% of Digital Ad Clicks Deemed Fraudulent, New Study by Oxford BioChronometrics Suggests
We had a problem with Yahoo PPC in Australia. We noticed this incredible and unnatural jump in traffic and we traced it back to Yahoo PPC and it turned out to be all from sites owned by one guy.
One keyword in each targeted campaign. I tried turning off the keyword and a couple of days later he'd pick another one. Thousands of supposed clicks at $0.02 each. Turned the whole thing off.
The scam wasn't even very well done and a blind man could see what was happening. Yet Yahoo pretended like it was normal. After threatening to take it to the ACCC (Aussie version of the FTC) they agreed to assign it to their fraud investigation team.
The report was supposed to take a few weeks but we never heard back and were still screaming months later. We finally got a partial refund and walked away.
This forum link talks about it and clarified what was happening:
"What kind of scam is Yahoo Search Marketing in AU trying to pull?"
https://www.blackhatworld.com/blackhat-seo/black-ha...
This article sounds like the same thing:
https://www.seerinteractive.com/blog/beware-of-a-ne...
We have a new technology called BioChronometrics which eliminates ad fraud. Very new technology.
Thought provoking and insightful article. Thanks.
Not GA/SEO expert but this might help: Error #27 – Traffic Reports Inflated By Spam & Other Bots
https://blog.kissmetrics.com/google-analytics-data-errors/
Robots can inflate and distort your real traffic numbers, so it’s important to know how to filter this traffic out. Google made some changes last year which allow filtering known bots out of your traffic reports.
Here’s their post on G+ announcing it last July.
It’s funny how product owners always dress things up to sound super cool isn’t it? The reality is it won’t stop spam bots or unknown search engine bots from inflating your traffic. There are ways however of refining your data to see where spikes come from.
Dave Buesing does a nice job of going through the process of filtering spam from your traffic. Check out his article here.
To activate the bot filtering in your Analytics Dashboard do this:
Go to: Google Analytics >> Admin >> Account >> Property >> View >> View Settings >> Bot Filtering (tick the box).
In last one week I have spent Rs 5000 in google adwords but no single responce :-/
Every time I run a display campaign via Google Adwords Display network I get 100% bounce rate. These run on name brand sites, latimes.com, cnbc.com, etc. It can't be a coincidence. When I bring it up to Google they just ignore me. I think there is massive fraud going in the display network. We found our own ads clicked on them and then could see our actions in Google Analytics, so we know the analytics is working. So if all the ads achieve 100% bounce rate that is really a serious issue. How do we expose this non-sense?
Again, this practice only hurts the little guy who can't afford to take part in "under the table" activities. It's the industry's dirty little secret that everyone fears but no one wants to talk about.
Thanks for bringing it into the sunlight where it belongs!
Thanks -- I'm glad you liked the article! :)
They are also "double dipping" charging credit cards two or three times for the same fake marketing. Might as well really just go right to the heart of the swindle, Sam! They'll charge the Client's Credit Card 2,008 dollars on Monday, 2007 on Wednesday, and 2003.18 on Friday, all for the same bogus 2,000 ad spend. This is more than just a loopy fuzzy math "impression" scam, it's a flat out Class Action Lawsuit all over it, and criminal conspiracy.
However, what is the solution? If we bring the blue hairs in Congress into it, we get regulation of the Internet. If we bring in the Justice Dept? Probably the same thing. When the biggest names in the business like Facebook and MSFT it is simply unbelievable by the average to think that these big firms are rip off artists. Some of us have to sleep at night, and actually have morals and scruples. Is that where we have all gone wrong? Are we playing fair while everyone else has a license to steal.
The funniest and almost stupidly ridiculous of these scams are by the new: "31 Pics of Famous Child Actors" or "See the Hottest Singers without Makeup" and they click off more actual unique visits because they compose these slide shows, and stupid consumer is just click click clicking the revenue. I did some research into the writers and companies, and usually the writers are nobodies, and the companies are based in Palo Alto.
So yea, it's literally a completely amoral system run by people with no scruples, and all they care about is ripping people off. They are NOT BUILDING ANYTHING! They are not out to change the world. They don't give anything back to Causes, they are just out to steal. It's criminal, it's disgusting, and I for one would rather the truth come out instead of this state where we are using decade old marketing language to prop up their scams.
I've vented enough. I tweet at PronetworkBuild
They are also "double dipping" charging credit cards two or three times for the same fake marketing. Might as well really just go right to the heart of the swindle, Sam!
I had never heard of that -- thanks for the comment! If you could post links to any reports showing that this is widespread in the industry (without divulging any private or proprietary information), I'm sure the community would appreciate it!
I am just surprise, why this sort of important news was not been in the mainstream. I Agree there are huge monoply.
Assuming the allegations are true, I'd argue that the reason this has not been in the mainstream press a lot is because a lot of people stand to lose a lot of money.
Excellent article! We actually just posted a little Q&A about ad fraud on our blog at Virool . This is a very important topic that our consumers need to be educated on. We also recently announced that we are TAG certified! It's a good move by the iab to start pushing for more transpereancy and a cross platform effort to reduce ad fraud.
yes there are a lot of bogus right now fake ads and fake stats
Hey Samuel,
Thank you for sharing for what exactly I was looking for couple of months . I raised these issue in inbound community https://inbound.org/question/view/what-are-thoughts... but surprised no one is interested in below topics :
Click farms: These are large groups of individuals who are paid to falsely manipulate the performance of ad campaigns. These organised groups use a variety of mobile devices to click on ads and systematically change devices and SIM cards to avoid detection. Click farms are set up with malicious intent to generate fraudulent impressions.
Incentivised ad networks: These networks reward users for completing specific tasks, such as clicking on an ad, reading an article, or downloading an app, all of which boost CTR (Click-Through Rate). Rewards could include coupons, reward points, or virtual currency. Users of these networks may or may not understand the impact of their actions.
We sell digital and print advertising in the recreational cannabis industry in Washington and Colorado. When we make sales calls we get to hear all of the horror stories directly from the business owners. I wonder if certain industries are targeted specifically due to certain factors. For example the cannabis industry might be a good target because of the amount of cash flow these businesses see and the fierce competitive environment that has developed with the sudden legalization. A higher number of "eyeballs" would help advertising avenues increase sales, but it would encourage those advertising avenues to promote illegitimate traffic. It might make for a great case study!
It would be nice if there was a magical tool developed that keeps an updated list of spam referrers and a transparent list of known bots as they are discovered. If it's use was the industry standard in reporting legitimate "eyeballs," that would help curb the problem. Is there anything like that out there currently? Finally, Moz you guys kick A**! Articles like this are ground breaking and only lead to positive changes. Especially for young companies like ours. Do you guys give tours? We're located right down the street!
I wonder if certain industries are targeted specifically due to certain factors.
Yes, certain industries are definitely targeted. If there is a lot of money to be made, then there is greater potential for ad fraud, black-hat SEO, and countless other actions that I've personally seen and heard that companies do. Specifically, I'm referring to forex, gambling, porn, pills, and a new thing that is becoming popular in my part of the world, binary options.
I always tell people who are getting into online marketing -- avoid those industries! If you want to learn real marketing, join a company that is selling a specific (hopefully good) product to a specific target market. Learn everything about product marketing and building brands. Those other, bad industries rely on tricks and spam that, in just one example, will get them killed by Google.
Finally, Moz you guys kick A**! Articles like this are ground breaking and only lead to positive changes. Especially for young companies like ours. Do you guys give tours? We're located right down the street!
I just wanted to highlight this comment in case Moz didn't see it. I concur -- not that I'm biased because I contribute here! :)
Hi Chase,
We do give tours! You can email tours[at]moz.com, and we'll work out a time for you and your team to stop by.
Email sent!!
Thanks for the great article Samuel. I too have been concerned with transparency of display impressions v. clicks etc. I know that the key "villains" here have been the ad networks and media agencies but where does Google itself shake out in all of this? As opposed to the big advertisers, I pay Google directly for my display ads; if I'm not getting what I pay for, I think Google needs to be held accountable too. I've seen a lot of comments on this string about how Google is trying to cut down on false impressions etc. etc but they are also directly profiting from this as well right? Shouldn't Google be held partially responsible for all of this?
I really don't want to point any fingers and state categorically that certain companies are doing certain things -- I have not done any original research and have only cited the investigations of others in this article. I only mean to present the allegations because the reports and facts seem credible -- it's up to we marketers and advertisers to decide who, if anyone, is responsible and what to do next.
That being said -- it seems that Google specifically is aware that most of its display ad impressions are not viewable. As I wrote, they announced that finding in conjunction with the unveiling of a new ad-viewability product. So it seems that Google recognizes and is attempting to address the issue. But has Google also been responsible for the issue? That, I cannot say one way or the other.
I have run hundreds of digital campaigns, from $5 a day to $10,000 a day budgets, through many of the top DSP/DMP portals including Adwords, I have had many conversations with top adtech sales professionals and attended many adtech functions and I can tell anyone who cares to listen that the digital advertising industry is the biggest scam ever perpetrated on the paying public. I cant think of one campaign that ever produced a positive ROAS for the brand!
But of course I will be on recap call today touting how wonderful we are doing for the brands we "service"
All adtech budgets should be listed as branding on the company chart of accounts, because thats all it will ever be!
PS
God help us if cookies ever go away, thats if you think DMP's give any better visability
Contextual = BullS....
Audience = Bigger BullS
Keywords, if you can find a platform that supports them may, and I repeat, may help. But considering this industry, I doubt it.
One more thing, our platform has the ability to track geo targeting. The day we went live with it, just for fun, we audited some of our adwords accounts (mcc). We were surprised at how many campaigns were getting charged for clicks outside of our geo-lacation. Argued with Google for 3 months, they finally refunded. We now have an agency rep that manages our accounts, refunds come much faster now.
Thw point is, until atribution is 100% we will keep getting robbed.
Hello, everyone -- just another update! This article is getting a lot of attention on Hacker News (Ycombinator). The thread here is also full of insightful and interesting comments. If anyone wants to get additional insight into this issue, I suggest checking that out as well!
How does this affect video ads?
I looked into the video side of the issue and saw these reports:
Millions of fake video views by bots
A report on a fake video impression fraud scheme (but it's by a company that sells ad-viewability products, so it's not an objective source)
A Wall Street Journal report on "drone pools" and video ad fraud
Yep, we're seeing a pick in humanoids that were never seen before. Most networks don't even pick these up yet.
Particularly worrying esp since Singapore seems to top the list. And everyone is saying do video ads esp on Facebook. And agencies are charging a bloody arm and leg for social media marketing and marketing automation because 60% of it is claimable under a government grant - it's a 3 way conspiracy/inefficiency in Singapore!
Shocking article but amazing content! I'm an AdWords specialist but I was thinking about an opportunity in starting with display campaigns via ad networks. Let me think again :)
Thanks for sharing!
Sander
I just got a rather large refund from BING. It was open and shut they made the refund in less than 7 days so I am happy with their service. I bet most unsavvy people would not catch it. I don't want to say what it was because I am considering passing it off to a class action law firm however I will need a $% as a finders fee. I did screen cap everything should be a slam dunk. From what I have researched so far is that a finders fee is frowned upon by many states.
How does this impact retargetting ads? If these bots are reaching our clients' sites organically or otherwise, it's possible they're being served retargetting ads. I've seen a surge of fraudulent sessions in Google Analytics through the referring domains tab recently. I'm wondering if this is related.
Thanks for the comment! To be honest, I have not explored the retargeting side of the issue that much -- so, I'd rather not comment at all than say something that I cannot support with evidence.
So, if you don't mind, I'll open the question to the Moz community! Anyone have any thoughts on the effect of this issue on retargeting?
I've worked in global ad agencies over the years so I was aware of this and I have read articles on this topic in the past. Tho personally never run display campaigns for large clients as I've been in Search Marketing for 9 years. Display guys were always getting far larger budgets aswell in the past :( As noted above stopping bot traffic is impossible as new bots operate daily. Even if you run huge lists of blocked sites and geo block you can still get hit. Agencies just need to be very transparent and honest to be successful in the long run, no point cutting corners.
Maybe if anything thus creates an opportunity for business that audit online advertising agencies. I know that I have always wondered how you can be completely positive that all of the numbers behind the impressions reported in day advertisiNg are accurate.
what a great article, we usually go via the direct route, but whats to say the bigger sites that charge on a CPM still use bot traffic in their calculations. I might be sending a few emails out this afternoon and doing a bit of digging.
Thanks for the compliment! This is the type of comment that makes me love to contribute to Moz. It's one thing to write about theory -- but if I'm able to help someone in some concrete, specific way, that makes the effort worthwhile! :)
Interesting and full article!
As you point out, some of the clicks on ads are false. Bots act freely throughout the network, wasting part of the investment by advertisers.
There are people who click on a page, not the page that interests you, but to cause damage to the seller.
This problem is difficult to remove. There is no vaccine against this fraud. But all you can do to avoid this is welcome.
I appreciate your recomendacuibes to try to avoid this problem.
Sam,
Thanks for bringing this up. No doubt that advertising and web monetization is a topic that isn't discussed too often among online marketers.
But as someone who managed ad operations in the past, and as an affiliate in my spare time, I sure do feel close to the topic.
I knew I'll have to comment after you wrote to stop CPM campaigns. Because it's the main way my company makes money.
So while you're raising a very good point for offline brands (in fact, 8% is very disturbing). Online brands on the other hand, shouldn't really worry about CPM in my opinion. And let me explain why.
You see, when Coca-Cola, Volkswagen, or 21st Century Fox - want to raise awareness for a brand new product through CPM, they have a very big problem. They can't really track the results of the campaign.
What they can do, is compare a campaign of one product to another, or try and target specific locations and analyze the impact. The problem is, that two products could really be apples and oranges, plus there are so many other things (including offline ads) - that could affect the results.
But for online brands, when all clicks are trackable--things are way simpler.
Years ago, when I started with ad ops, I was shocked to find out that impressions are counted for banners below the fold, even when users don't scroll down. But then I realized, that in the end of the day the advertiser only cares about the final results, or in other words, the Cost Per (whatever is my) Goal. If the goal is to get leads, then it's the total expenses--for the number of leads. If it's actions (sign-ups, purchases), then once again, the total expenses--for the revenue made through the campaign.
CPM is the main way we sell media and it works very well for our advertisers. We learned that all the advertisers want to know is how much money they spent, and how much money they made through our traffic. That's it.
Other than that, I did hear about ad networks fraud, but never thought that numbers are so high. We all wish that more advertisers would work directly with publishers, and that data would be more transparent. Thanks for sharing!
Igal, thanks so much for the great comment! Just mind if I throw out a counterpoint?
(Obvious caveat: This example is all hypothetical and does not reflect on you, your company, and anything you do as an affiliate. :)
Say you report to a client that they spent $100,000 from your ad network and got 1 million leads at a cost of $0.10 per lead (CPL). Sure, the client might be satisfied at that result. But what about the day when the client reads an article on this topic, calls you, and says something like this:
"Half of the impressions were never seen? Well, how many more leads would I have gained and how much lower would my CPL have been if all of the impressions had been seen?! How much money did I waste because of your faulty ad network?"
That's what ad networks -- and the media buyers and ad agencies that use them -- are going to be hearing.
You are making a good point and clients should ask these questions. They should also compare between agencies, ad networks, and sites to advertise on to see what brings better results.
Obviously you'll agree that not all publishers buy bot traffic, but of course, it is important for the advertisers to be aware of the fact it exists, and I'm sure that many aren't aware.
I'm just saying that if an advertiser is familiar with the industry, the industry's CPL/CPA prices, and what the competition has to offer. CPM should work really well for both sides.
Samuel,
I agree with Igal here and don't understand your counter-point. The client is paying for the outcome, so as long as everyone is getting "duped" equally, I don't see how it affects the price. If all of his ads were being seen by human eye-balls, he would simply have to pay a LOT more for those impressions.
In other words, if I peel my bananas before weighing and paying for them, I won't save any money. The store will simply charge a higher rate for peeled bananas.
Do you disagree?
(Of course, this doesn't make it ok to practice scandelous behavior. I'm just pointing out that it's not so simple to conclude that the client is losing money.)
By the way - this is a fascinating piece of journalism. Well done (and thanks to @Moz for the refreshing genre!)!
Here's the point: The outcome for the client would be better if all of the impressions were visible. As I wrote in another comment:
But here's the issue: Say a $5,000 ad campaign resulted in a cost per lead of $1 and only half of the ad impressions were visible. If 100% of the ads were visible, then the number of leads would double and the cost per lead would decline by half to $0.50.
That's how advertisers are losing money to non-viewable impressions.
I understand. But if 100% of the ads were visible, the campaign would cost $10,000, not $5,000. The price is not based on thin air.
Here's the thing:
If you pay $0.10 per served impression, then you would pay $10,000 for 100,000 served impressions. However, at an assumed rate of 8% viewable impressions, that $10,000 ad spend will result in only 8,000 human impressions—meaning that the effective CPI rate will actually be $1.25. based in terms of how many actual humans see an ad.
Samuel, you're not understanding my point. The prices for advertisements are not random. They are based - at least partially - on supply and demand, and the ROI produced by the advertisements purchased. If I'm willing to pay $100 for 10 leads/conversions, and it takes 100 impressions to produce those conversions, then I will pay $1 per impression. If it turns out that only 10% of those impressions were viewable, that means that I paid $10 per viewable impression (when I thought I was paying $1). But if this is the "industry standard" - that only 10% of impressions are viewable, and then one day advertisers start paying for viewable impressions only, then the price will change. The price per impression will go up to $10, since viewable impressions have 10x the conversion rate. So the advertiser will continue paying the same amount of money.
I hope I'm making myself clear.
Yair is spot-on - The CPM is market driven. If an advertiser isn't getting value for money than they simply refuse to pay the amount and the inventory remains unsold. If it's on an auction based platform the price will just drop to a point when it does become valuable.
My client doesn't care how many views an ad received, just how much a lead or sale costs and whether that's profitable when placed into their sales funnel.
If an advertiser is not tracking their performance then they are perfectly suited to the marketing agency taking them for a ride. The fake impressions are a non-issue for any client or marketing team doing their job properly.
Hey Samuel, great work. A few paragraphs in I was thinking you'd make a great journalist. Your background definitely shines through here.
I gave a talk roughly about this issue last fall at some point. It wasn't nearly as in-depth as this, but generally went into viewability, fraudulent ad serving, lack of transparency, etc. Ironically, it landed my agency some display-only work. The only way I approach display campaigns now is with clients we can collaborate with and educate enough to run their own campaigns and ad buys.
Thank you for the nice comment! Yes, it was nice to write a Moz essay for once that was more "journalistic" in nature rather than one that discussed marketing theory and tactics and whatnot.
But, I also need to give a shout-out to the Moz team for publishing this article! Obviously, this issue is very explosive and the implications are enormous for the digital marketing and advertising community. Needless to say, not every outlet would publish something like this.
I don't work for Moz and cannot speak for them. But my observation is that Moz is devoted to publishing good, objective, and important content on anything relating to online marketing and advertising. In the end, publishing such information -- without bias or fear or favor -- helps the community here to get the best information and do their jobs the best.
Hey Samuel,
Thank you for sharing. I've been burned more than once this year from the boutique to enterprise size agency. In the end I have diverted all display and programmatic advertising to in-house team and direct buys with publishers. The process isn't pretty, however the ROI difference for assisted and last-click is tremendous. Sadly, outside of Paid Search, I can no longer trust or refer an agency.
I recently went through a couple of products, threat matrix and distill networks. Not sure whether these could help in keeping the bad bots away.
It depends on what type of marketing you are doing and in who's network. Keep in mind, bots are changing very rapidly and who ever you are using must have highly advanced bot detection capabilities like Secure Ad (Oxford BioChronometrics). We are seeing humanoids that are so advanced most players will not detect them.
Programmatic works and works great, the fact that some ads were never seen means as much as some people change the channels over TV commercial, miss a page skipping a print ad or talking on the phone when a paid radio spot hits . it’s just the nature of the business , if your not showing a good ROI stop what your doing and move to another midia .. RTB has worked very well for us.
Given Google's extraordinary ability to solve incredibly complex mathematical problems, it's extremely unlikely that: (1) they are not aware of the scope of inaccurate (or fraudulent) clicks or impressions, and (2) that they cannot fix the problem or account for it using a mathematical model. "It is difficult to get a man to understand something, when his salary depends on his not understanding it." -Upton Sinclair
If Google can't figure out reCAPTCHA, then they definitely won't figure out ad fraud.
The numbers that you mentioned are incredibles. Really this amout of money are fraud in online advertising??
Ad fraud is hitting $20B already. $18.5B here and$3B here.
This is a great article thank you. I have learnt a lot about digital marketing. My background is finance.
I would say that this level of corruption will lead eventually to a series of scandals and exposures and a correction. I read some views here that the correction would just involve advertising prices going up (for better quality advertising), meaning 'advertising spend' and 'revenues from sales' for clients would remain the same. So little difference for the client.
This is not usually how a correction in an inflated industry works, the banking industry is the most recent example, but there the correction was muted by regulation protecting the wider economy. The only restraint from a more severe correction in the marketing industry is competitive pressure amongst clients to maintain advertising spend. In an effectively unregulated market a series of scandals/exposures would remove this competitive restraint for a period of time, enough to make a major correction.
So it is much more likely that a series of scandals/exposures will mean that the marketing industry will have to accept smaller margins (smaller salaries). Marketing companies would have to reduce prices due to a severe cut in client spend from a number of companies together (they can afford to cut advertising if their competitor is also doing it as a result of the scandals/exposures) which puts the most egregious advertising companies out of business.
Margins will then stay lower even whilst client spend recovers to more normal levels with the client likely to be the winner - spending slightly less on higher quality advertising. This may or may not generate greater revenue from their product/service, but they will at least be back in control of the advertising service they are paying for, and that's a good thing for the client.
I don't think it's a serious problem at all - just a nuisance, at best.
Also, I don't think anybody loses much in the process - including clients. That's part of the reason nobody is trying to do "anything" (or much) with it. Why? Because professional marketers evaluate campaigns based on the ROI.
In other words - if a company still sells enough products or services to justify the expense, then in what respect is it losing? Sure, you may argue that it paid a lot more for false clicks that it needn't have paid for to convert as many customers - but I don't think that's necessarily the case. Because this system can be abused, millions of websites have been created to take advantage of it - in other words, the pool of potential placements has greatly increased, bringing the average cost per click down.
In other words, on one hand bots have been created to inflate the number of registered views and clicks - but because they exist, the number of places where ads can be displayed has also increased. There would have to be a very complex - and probably impossible - statistical analysis made to actually evaluate the impact it has on both supply and demand sides.
If the system could not be gamed, the number of sites - and, thus, placements - would likely be much smaller, bringing the average CPC up, because you'd have equal number of companies competing for far less display space.
Therefore the actual cost could very well be on par with what the company is paying now.
In this light the companies which actually stand to lose more are the relevant, reputable publishers, who earn less due to lower average CPC, which is taken away from them by fraudulent bot activity and millions of placements artificially created to take advantage of it.
But even this argument is slippery, because those reputable sites have to balance their interest in having you click the ad, with an even bigger interest in keeping you on their site - and have you come back. In reality many - if not most - clicks happen by accident anyway. If the content is good you don't want to click the ad - you want to read the text.
So for the end user ads next to pretty good content content they actually enjoy (e.g. Cracked) are actually detrimental to user experience - hence the rise of ad blocking software.
In fact - and ironically - bot activity could have actually created FAVORABLE conditions for advertisers. As long as bot traffic exists and as long as you can utilize fraud detection software, you can actually get ahead of your competitors by paying less, avoiding artificial traffic, while still enjoying the CPC rates lowered by the fact that it exists. For the playing field to be level, all advertisers would have to use detection software - and we know that's not really the case yet.
In a bot free reality, existence of millions of websites - together with their placements - would also be rendered useless, so the cost of reaching a single willing buyer could very well be just as high as it is today.
Dude, you're delusional. $20B/yr of fraud is a big factor. Most will not admit the issue because of the conflict of interest at hand. It really comes down value creation. Spend the same budget with real ads and convert a higher percent. That is the new frontier...
I'm sorry but that's no factor at all. You cannot say that the companies would pay less for reaching their real customers, because with cutting away fraud you're also cutting away the display supply side, which drove cost per click to new lows. In other words, companies pay a lot less per click, on average, just that more of these clicks are rubbish.
Remove the rubbish clicks and the average CPC goes up, because you end up paying more to reach each genuine person. In the end you are likely to end up paying roughly the same to generate your sales.
So for advertisers it's not fraud, it's just noise. The only companies really losing out are reputable publishers, who get a smaller share of the advertising spend, simply because some of it is seeping away to low-end, crappy websites, designed mostly to leverage fraudulent clicks. That's all. But even that is hard to estimate.
It reminds me of software or music piracy figures, which have always been completely bonkers. People who didn't buy music or games, simply wouldn't buy them at normal prices at all, even if piracy didn't exist. It's similar here - the cost of reaching a real human being would likely be the same with or without fraudulent traffic. The only difference is in how the advertising spend pie is divided between those who drive traffic.
Ah, I see you're in the bot prevention industry. That explains a lot.
Interesting article and a subject that does not get a great amount of coverage. I've found that it often comes down to the webmaster of the site to manage spam traffic on their site. Using a content network like the GDN removes element that control. Usually you should be able to spot spam traffic in your analytics, highlighting the importance of regularly reviewing the quality of your traffic.
Since this is 2 years old would you have any updates on the statics or measures that may have changed, good or bad? As a Certified Fraud Examiner, I would like to share the article with updated information.
Hove you guys looked into BlueLink Marketing or AdLander.io?
Bots are the tip of a huge problem when using 3rd party ad-tech providers. get control of the programmatic technology in-house and you can have clear insight and accountability.
Hi - really insightful article, thanks for sharing. I work within marketing effectiveness so this kind of insight is very useful for day-to-day conversations with colleagues.
I'm struggling slightly with the 8% theory. I understand the concept of broken ads, bots and what you've described as alleged fraud and, running through the math, I'm having to make the assumption that these factors aren't mutually exclusive - if you add tem together you get 100%.
The bit I didn't quite understand is how the above leaves us in a position to start a notion that 15% of impressions are possibly seen by a real person - are bots and alleged fraud 15% overlapped?
My second question is as follows: what are we are then removing a further 54% from? - 46% of 15% isn't 8% i.e. the viewable proportion of what's left once you remove broken ads, bots and alleged fraud.
Final question - how does "below the fold" banner placement and consumer installed ad-blocking software factor in? Is this the 54%?
Thanks in advance.
Awesome blog!!
Check out my blog on Michigan Programmatic Search Test for additional insight. https://thebestdigitalmarketingcourses2016.wordpress.com/2016/02/09/programmatic/
Just like Big Oil and Big Pharma, no one likes to have anyone shed light on their mafia-like cartel. Either a massive expose and/or a Congressional investigation will need to ensue to see these smarmy fat cats change their ways. But we live in a new era of transparency, and word will spread fast. Keep writing and exposing their fraudulent ways util we see a change.
Samuel
I understand your concern. but I think not all the time fake traffic conflicts with impression. I was working on DFP for last couple of months and find many issues with unfilled impressions and un-rendered impression, where ads are never seen. Which can be measured by page-views. We all know that Bots And Spiders Causing Unusual Spikes in Analytics Traffic but it can fixed through Google Analytics’ Bot & Spider Filtering. In that way we can also exclude the number of fake impression. I am not sure if there is good tool which exclude the bot impressions.
That is why I like CPC mode ad campaign. Where we can measure ROI with perfect math.
I certainly understand the clarity that CPC campaigns can give with regards to ROI calculations. When I've run ad campaigns in the past, I'd always wanted to say our average cost per click (or per lead or whatever) was $X.
But here's the issue: Say a $5,000 ad campaign resulted in a cost per lead of $1 and only half of the ad impressions were visible. If 100% of the ads were visible, then the number of leads would double and the cost per lead would decline by half to $0.50.
That's how advertisers are losing money to non-viewable impressions.
Samuel,
Thanks for sharing! Good to see someone else is shedding light on this.
Because of this issue, after 18 years buying and selling media (across various industries), I recently jumped to the agency side to pursue technology that would eliminate this waste and fraud, while also connecting online and offline data, and enabling advertisers to directly attribute conversions to an online display campaign. In essence, solve the cookie problem (fraudulent clicks/impressions, lack of data, ability to be blocked/restricted, etc...) by removing the cookie.
So, the point is, there is a better way to run an online display campaign WITHOUT fraudulent clicks and wasted impressions - by targeting individuals (or businesses) through their IP address. This typically performs well above industry standards, and makes it easy to attribute sales to a specific online display campaign, since we know exactly which individuals (or businesses) received the ads.
Just thought I would share for the benefit of other readers that are looking for a way around the fraud that seems inherent to online display campaigns.
All the Best!
-T
Thanks for the comment -- and congrats on moving away from the alleged "dark side"! :) If you could provide more information on how advertisers can avoid waste and fraud, I'm sure Mozzers would love that.
Sure! As the article above thoroughly covers, there is rampant waste and fraud that is inherent in the online display environment. However, through the use of patented IP targeting technology, we are able to target specific users by identifying an IP address that corresponds to a specific physical address. We can then serve multiple ads to these addresses, and ONLY these addresses, without the use of cookies - think of it as direct mail in a digital format. Therefore, we can ensure that advertisers are reaching ACTUAL humans and not bots or other non-human devices. This also means that advertisers no longer need to waste impressions targeting a block or zip code to reach a specific individual. IP targeting enables advertisers to serve ads to only those addresses they want, without wasting impressions on their neighbors.
Furthermore, because this is not cookie-based, it is immune to browser-based filtering methods or other problems inherent with cookies. Audience targeting that focuses on behavior and content has become problematic for advertisers because browser-based methods are unable to reach unique users or provide verifiable accuracy. A cookie is supposed to represent a unique user, but every unique user is generally connected to 3, 5, or even more cookies at a time. Therefore, there is no way to really know whether a unique user that is connected to one cookie is the same unique user that might be connected to other cookies. Other problems include being blocked at the user level, having the cookie time-out and disappear (along with all the data it collected) after a brief lifespan, and other issues which make it nearly impossible to consistently identify and track individuals online. By targeting unique users at a specific household or business without the use of cookies eliminates the need for browser-based tracking by referencing geographical location in connection with web entry location. In short, it has solved the "cookie problem" by eliminating the cookie.
Hello Mozzers -- I forgot to add one additional resource to the article. Here is Ad Weeks's full archive of ad fraud posts, for those who want to do more digging. Thanks for reading, everyone!
Hi,
I'd like to ask you further detail about Facebook and other social media.
To be honest: I've never seen Facebook data matching with the data collected by G.A. Usually, Facebook reports much more visits/clicks towards my sites. So I sued to think that Facebook used to fake this number on purpose. Do you think that is due to this issue?
Speaking of social networks ads, I have seen that two ads are "chasing" my Facebook and twitter news feeds. I've noticed them, as their services sound nice, but they don't cover my local area. So what I don't understand why are they not able to target these ads in good way? By they, I mean these business, their services supply only London city: why do they push ads to me that I am not within London postcode area? OK, a human being can make errors, but to be honest it sounds more like pushing ads randomly. I say this because they are new star-up, and I think it's quote unlikely they invest money to cover areas they can't supply. it won't brand awareness when you are not sure that tomorrow you won't go in bankrupt! Do you
think this kind of wrong targeting can be due to an issue, bad management or real ads fraud?
Cheers
PP
To be honest: I've never seen Facebook data matching with the data collected by G.A. Why should it? They are two completely different systems and each will track traffic ir visitor metrics according to their own internal specifications. Like here you have PRO designation by your name other places maybe not.
Great article Samuel. Out of interest what's your opinion of post-impression conversions?
Thanks for the comment -- unfortunately, I'm not sure exactly what you mean. Do you mean someone sees an ad for a website and then visits the website later either directly or via another medium?
thanks for sharing this article. Usually you should be able to spot spam traffic in your analytics, thanks zugtech for helping me to join the moz
[link removed by editor]
Excellent article. I noticed today that some Adwords ads were coming up at the bottom of the page. I happened to scroll down and see them but I realised that anything below the fold isn't necessarily going to be seen.
I am glad that someone show interest to uncover the truth behind internet ads and money. These scams are the way to make money quickly and there concern is all about your account and other document details. They want your money and for that they convince you anyhow.
Please be aware of these Scammers.
Another update for Mozzers from Ad Age on another plague of online ad fraud:
Thousands of so-called zombie apps are infecting mobile phones, expending data usage and battery life at an alarming rate while costing advertisers nearly a billion dollars annually.
I agree that on-line advertising is especially susceptible to fraud. If you look on the web, you will see many sites where people are using click fraud to make lots of money. This is a fraudulent activity wherein people will pay other people to click on links with no intention of buying anything. Then the advertiser has to pay the people who are hosting the on-line ads, even though the click through activity was entirely fraudulent. The way this activity earns money is that they are earning more from the advertiser than they pay to the people who click on the ads. What is hard to understand is how these people are able to morally live with themselves when they are stealing money from others.
Hi,
I'm a very small advertiser/publisher but I see far more impressions than I think is possible.
Can I e.g. extract suspicious IP's out of logfiles and add them to e.g. a robot.txt file to block them from accessing my websites and share my blacklist with other websites? Or is this not possible at all?
I'm looking for a low-cost solution to weed out the fraudulent impressions before they cost me money...
You've summed up in one article why people still prefer to advertise on Google than on other ad networks - and it's because G is the cleanest of them all.
Excellent article. Learned a lot of things. Thanks for taking time and writhing the article.