Ad fraud: mobile advertising accounts for nearly half of digital spend, but it comes at a price

/IMG/198/337198/google-interstitial-crackdown-370x229

US Advertisers are spending US $2.6 billion on mobile ads each month, $0.4 billion in the UK, they understandably want to know that their ads are seen by real people, not robots, ideally the people most relevant to their ad message.

Advertisers are now investing up to half of their digital advertising revenues – according to the latest stats – in mobile ads. They want/need to know that their money is invested wisely and that the ads they are paying for are seen by real people, not robots, ideally the people most relevant to their ad message.

Advertisers need to be able to trust that the various parties that create, manage and deliver etc. their ads have put all measures in place to ensure their ad campaigns deliver maximum return on investment ROI. That is entirely reasonable.

Richard Foan, Chairman, JICWEBS (the UK-based Joint Industry Committee for Web Standards) tells ClickZ:

“The level of concern among advertisers about click fraud – that includes mobile click fraud – worldwide has now reached a point where the industry – agencies, ad networks, DSPs, exchanges – needs to accept there’s an issue and continue pulling together, through cross-industry initiatives to rebuild trust.

The industry must get behind initiatives from bodies, such as JICWEBS in the UK and TAG (Trustworthy Accountability Group) and MRC (Media Rating Council) in the US, make sure their business practices – and those of their suppliers – are compliant with the best practice guidelines and get themselves certified/registered as being compliant.”

Research by InMobi (via eMarketer), July 2016, shows that at 48% concern over fraud/viewability is the biggest barrier to the take up of programmatic purchasing (buying ads on the fly via an ad exchange) of mobile inventory.

mobile_programmatic_barriers

We will look more closely at the industry initiatives led by JICWEBS, TAG and MRC in part two of this column.

What is mobile ad quality?

Advertisers (publishers too) are beginning to require from their adtech partners an assurance that they are buying good quality mobile inventory.

The following definition of mobile ad quality is an adaptation of one supplied by David Ip, mobile, cross-device and healthcare advertising at Conversant Media.

Mobile ad quality is a combination of:

  1. No evidence of ad fraud, or other invalid traffic – ads are seen and clicked by humans, not robots.
  2. Optimum viewabilty – ads are displayed on the mobile page in such a position and for long enough that a human registers it.
  3. Content adjacency or brand safety – ads are displayed on a mobile site/app and next to content that is appropriate.

This is the first of two columns on mobile ad fraud. Part 2 will focus on spotting and combatting mobile ad fraud. Viewabilty and content adjacency will be discussed in subsequent columns.

How big is the mobile ad business?

By any measure, mobile ad revenue is growing very rapidly, worldwide – particularly in the US.

US – mobile ad business was worth $15.5 billion in the first six month of 2016, up 89% YoY. That is 47% of online ad spend (Source: IAB, November 2016).

UK – mobile ad worth £1.7 billion (US$ 2.1 billion) in the first six month of 2016, up 56.1% YoY. That is 36% of online ad spend (Source: IABUK, October 2016)

How big is mobile ad fraud?

When you have a prize of this size it should be no surprise that it’s attracting the attention of the fraudsters. The problem is no one knows exactly how big mobile ad fraud is or how much it is growing.

The internet advertising business has only recently started trying to measure and discus how to combat desktop ad fraud – which has been around a lot longer than mobile ad fraud – arguably, it has been plaguing the business for decades.

If things continue as they are, the WFA (World Federation of Advertisers) projects:

[Internet] Ad fraud is likely to exceed $50bn globally by 2025 on current trends second only to the drugs trade as a source of income for organized crime.

In fact things have got so bad that in July 2016, in the US, two senators wrote this letter to the FTC (Federal Trade Commission) asking what steps the FTC was taking to:

  1. Mitigate digital ad fraud.
  2. Reform opaque advertising exchanges.
  3. Combat criminal organizations involved in ad fraud (together with e.g. the FBI).

But it isn’t clear what proportion of ad fraud is mobile, as industry stats do not seem to be available. However anecdotal evidence from agencies – those that have the systems in place to detect it – suggests that mobile ad fraud is on the rise or, if not, they are getting a lot better at detecting it.

Shailin Dhar Ad-Fraud Consultant, Dhar Method:

“Mobile click fraud is uncharted territory in comparison to desktop because it has not been monitored with the same scrutiny and not for nearly as long. Estimates with fraud are always tough but I would say there are more invalid/accidental clicks on mobile 

That’s the universal problem with fraud statistics: there’s no such thing as accurate fraud numbers. It’s always estimates, regardless of what anyone tries to claim. You are trying to count something that actively doesn’t want to be counted.”

The US and UK already have significant volumes of click fraud but we will see, or at least detect, an increase in emerging advertising markets like India and China.

So how does mobile click fraud work?

Mobile ad fraud or click fraud works in multiple ways, including:

  1. Fraudster places hidden code in a) browser extensions; b) native applications, which are then download to the user devices, or c) on publisher sites.
  2. The code generates fake views or click throughs a) as the user of the compromised device browses the web, or b) or simulating browsing while the device is supposedly dormant; or c) ads are served invisibly behind legitimate ads or screens on the compromised publisher app or site.
  3. The views or click throughs are attributed to the fraudster’s network, which takes a cut of the click-through revenue.

For a more comprehensive description see Anti-fraud best practices (PDF) from the UK based JICWEBS (Joint Industry Committee for Web Standards).

According 2015 research by Forensiq, an ad verification vendor, mobile devices are particularly susceptible to being hijacked by native apps, innocently downloaded, that contain malicious code that serve hidden ads and simulate click-throughs to landing pages that are never seen or by the user.

Over a ten day period, the vendor spotted 12 million unique devices running run at least one of the 5,000 apps flagged for ad fraud. The research estimated that these hijack apps would cost advertisers US$857 million in 2015 in hidden ads. An example of a hijack app is shown in the image below.

hidden ads

Why is mobile ad fraud on the rise?

Alex Hewson, head of media EMEA, M&C Saatchi Mobile:

“Simple, risk and reward. Without data analysis processes and technology in place, it is difficult to detect and highly profitable. Legal recourse is very rare as, owing to the evolving nature of the industry, and standards are unclear.”

The area of most concern according to agencies and ad verification vendors alike is ads to drive downloads of mobile applications.

Hewson, again:

“As cost models in mobile move towards cost per install for a large proportion of spend, the fraud we see is therefore largely associated with app install campaigns.  And this is predominantly driven through display channels.”

Badly designed ad formats to blame – yes it’s those [EXPLETIVE] interstitials again.

Sometimes what appears to be ad fraud, may not be. Sometimes unintentional clicks are caused by mistakenly tapping an ad when attempting to close it.

One of the plagues of mobile web and apps is full screen ads, such as interstitials, that interrupt page loading with a plea to “download our app” or “sign up for our newsletter”, but so often – no matter how accurately this is done – when tapping the cross (exit) to close the ad, it instead clicks through.

Whether this is just poor design or malicious intent on behalf of the advertiser, publisher or ad network, these bad ads deliver erroneous or fraudulent clicks. But we aren’t just talking a few clicks here and there.

These bad ads are so prolific that they actually cause spikes in mobile ad fraud levels. As Jason Cooper, general manager, mobile at ad verification vendor Integral Ad Science, explains:

“Fraud of all kinds fluctuates, so we detect both rises and falls at select times. We mostly attribute any increased occurrence of mobile click fraud to the rising popularity of full screen ad units such as interstitials. The IAB requires that these ad units incorporate an exit button at the corner of the ad, meaning any miss click from the user while exiting the ad could be causing skewed CTR (click-through rate) results.”

In the next column we will consider how the industry, agencies and advertisers can help prevent being victims of mobile ad fraud.

This is Part 36 of the ClickZ ‘DNA of mobile-friendly web’ series. Here are the most recent chapters:

Read the reports:

Andy Favell is a ClickZ columnist on mobile. He’s also a London-based freelance mobile/digital consultant, journalist and web editor. Contact Andy via LinkedIn or Twitter at Andy_Favell.

Related reading

flat design mobile payments
Concept for mobile apps
man's hands with smartphone
cost per sale per city
Simple Share Buttons