Skip to main content

Intelligence Briefs

Connected TV: Any impression served to non-human traffic is worthless to an advertiser, so let's stop it now

Industry Contributor

Dylan Dharmadasa, Head of Product and Operations
Finecast

21 October 2020 4min read

Connected TV (CTV) is apparently the latest battleground for ad fraud. Who knew that Australians are watching TV via 1,300 CTV apps at scale? Perhaps now is not the best time for industry to relax controls in the pursuit of lower CPMs.

 

Key points

  • Recent claims from DoubleVerify about “sharp rises” in fraud traffic rates, are alarming and we should always aim to stamp out any fraudulent behaviour.
  • But they aren’t necessarily representative of all buying in the ad tech ecosystem and the claims should be regarded in context of who is making them and why.
  • Most fraud opportunities exist in open exchange buying where it is possible for CTV devices to be falsified through ad networks, but much of this nuance hasn’t been explained in depth, with ‘fraud’ claims levelled at the whole CTV ecosystem.
  • Fraud risk increases dramatically when path to the source of media is unclear, which opens up an avenue for payment to bad actors. If a vendor has a direct billing relationship to the inventory, or buys within a closed ecosystem, the ability for a fraudster to partake in the supply chain is greatly reduced.
  • There are always inherent problems with new and emerging tech platforms, but the simple act of buying media in a closed ecosystem, particularly Connected TV, can drastically reduce the risks for any client. 
  • It’s crucial to weed out players that are passing off a sub-standard version of Connected TV to their clients, which are inevitably the ones that lose out.

 

My Takeout

Alarming stats around ad fraud often emerge, but understanding what they mean, and how it impacts the overall supply chain is important - as are individual decisions to prevent it.

Recently an article on trade press website AdNews claimed, “Fraudsters are flocking to CTV” citing a DoubleVerify report. 

Mi3 explored the issue in more depth talking to DoubleVerify, some programmatic firms and media agencies about the issue and its impact in Australia, and surmising that “premium publisher deals are key to response”.

The DoubleVerify global insights report claims “sharp rises” in fraud traffic rates, an alarming 161% increase on CTV year on year. Whilst these statistics are very concerning, they aren’t necessarily representative of all buying in the ad tech ecosystem and should be regarded in context

Firstly, let’s dive into how fraud appears in the advertising ecosystem. Fraud follows where there is monetary gain for fraud to exist. Without the monetary benefit, there is no benefit to fraudsters to act within an ecosystem. For example, in linear TV, miscellaneous fraudsters should find it difficult, if not impossible to position themselves within the supply chain and gain financial benefit from the selling of media.

Conversely, when we look at the broader digital ecosystem, take for example, display banners or social media, there can be a benefit: If a fraudster can drive non-human traffic to a page, there is a potential to monetise the inventory. Another way might be to a masquerade a display banner placement as a valuable BVOD impression.

There are some key questions: Does this mean that all display or social or BVOD is therefore tainted and that fraud is a fact of life? What is different about the linear TV ecosystem that reduces its risk profile compared to the digital world, and what is different about the display or social ecosystem?

The answer lies in the fact that the linear TV ecosystem is a closed one. The supply chain is clear. I can work with five, well-known broadcasters: Seven, Nine, Ten, SBS, Foxtel, there are no other intermediaries that I buy from that sit between the broadcasters and the agency. For every dollar I spend with each broadcaster, the broadcaster reports the same dollar on their side of the ledger.

So, is it possible to apply the same practices? The answer is of course, yes. It comes down to quality supply chain control. Who are your partners, do you know them, are they reliable, are they incentivised to act in the right behaviour? Can unnecessary intermediaries be removed?

In social, buyers typically rely on the social platform’s capabilities to mitigate and manage fraud, through their own automated detection systems, anti-fraud policies to recoup any expenditure on fraudulent activity after-the-fact. Buyers should assess the policies to ensure they satisfy a client’s requirements.

CTV is an emerging channel – it’s growing fast and offers huge opportunity for advertisers. But like anything new, buyers should exercise caution and apply risk mitigation practices.

At Finecast, like linear TV, we work in a closed ecosystem, directly with all five Australian broadcasters, namely, Seven, Nine, Ten, Foxtel, SBS. We work directly with our DSP partner and the broadcasters work directly with their SSP partners to integrate all inventory and data buying. For every dollar spent on CTV inventory, the broadcaster can see the opposite side of the ledger, down to the exact ad impression. There are no networks or resellers or outside traffic from unnecessary intermediaries.

But the reality is that some of the industry doesn’t buy in the same direct fashion. Some buyers are happy to buy from questionable CTV apps, and happily relax strict supply chain controls in favour of the ever-lower CPM. They may buy inventory from intermediaries or resellers that may not even have offices in Australia. Let’s be clear, any impression served to non-human traffic is worthless to an advertiser.

The result of relaxing these controls may be a ‘lower CPM’ but it also provides a monetary incentive for fraudsters. And therefore, the need for anti-fraud vendors [like DoubleVerify] grows substantially as you will need automated detection across all your activity to monitor fraud levels and assess supply partners who regularly offend.

DoubleVerify cites in its report that is has blocked 1,300 fraudulent apps since March 2020. Another alarming stat, but one that flags questionable media buying practices in the first place, leaving clients highly exposed to fraud. It seems unreasonable to think that Australians are watching TV via 1,300 CTV apps at scale. Buyers should be considering their own business rules and practices whether it’s via their agency, direct buys or via their DSP.

There are industry-wide initiatives to improve the ecosystem for buyers that do need to work in an open fashion. For example, in channels like display there isn’t always the option of buying hundreds of sites directly, and the IAB has made great strides to reduce and mitigate the risk of fraud. Ads.txt and apps.txt, which ensure that a publisher’s inventory is sold through authenticated supply chains, are fantastic examples of initiatives employed to mitigate fraud risk for buyers.

The fight against fraud is not as simple as any one specific act, it’s about understanding the risk profile of how you buy media and applying rigour in business standards and principles, through a combination of anti-fraud policies, standards, technology or any other means.

Let’s go. What do you think?

Industry Contributor

Dylan Dharmadasa, Head of Product and Operations
Finecast

Market Voice

Linear TV is neither dead nor dying: But it needs to evolve faster than VOZ

Audiences have already converged. Now the TV, digital and media planning industries are racing to catch up. But some routes are quicker than others. Choose wisely, says Amobee’s ANZ Head of Sales, Andrew Dixon.

Go deeper 3min read

Andy Dixon

Head of Sales ANZ, Amobee

Amobee

23 November 2020 3min read

Linear TV is neither dead nor dying: But it needs to evolve faster than VOZ

Audiences have already converged. Now the TV, digital and media planning industries are racing to catch up. But some routes are quicker than others. Choose wisely, says Amobee’s ANZ Head of Sales, Andrew Dixon.

Go deeper 3min read

By Andy Dixon - Head of Sales ANZ, Amobee