At the turn of the year, we at Decipher wrote a piece on these pages asking whether the advent of future ad-tech – against the backdrop of ad-tech itself, the web and television becoming ever more converged – required a new form of (at present, extremely limited) regulatory oversight. This was in reaction to a story, first published in the Financial Times and subsequently elsewhere, that over the course of a one-month investigation 72% of ad impressions being offered on open exchanges under the FT.com were fraudulent. Such was the scale of the problem that the FT chose to write directly to advertisers, warning them not to trust inventory being traded on open exchanges and to only use the FT’s trusted seller.
A month has passed, and in the intervening period several figures within the industry have responded to the view we put forward – one of whom we’ll return to later on. In the meantime, we promised to provide readers with a more detailed look at what exactly online “ad fraud” is (it’s a broad term that can mean a number of different things at any one time), as well as to examine what the industry is doing to tackle the problem. Thanks to one response we received to our piece, we’ll also be able to bring readers right up to date with what is being done in the UK through a new self-regulatory mechanism.
Types of ad fraud
Online ad fraud is well-defined by a Brightroll report as false views, clicks and/or visits to a publisher’s desktop or mobile site, and can be driven either by non-human, automated actors (“bots”) or humans. This illegitimate traffic can be initiated by a person manually, by an unknowing user (via an infected computer), or fully controlled by a bot or botnet (a network of “bots”). Online ad fraud is not new (think click fraud in the context of search advertising), though it is taking on new shapes and forms. Bots, and by extension botnets, can generate ad views while users are unaware, hijack user controls to generate “fake clicks”, simulate consumer activity or compromise cookie data. iFrame “stuffing” (when an ad is made unviewable by stuffing it into a “1×1” pixel), and ad “stacking” (when multiple ads are played at the same time, with only the top one being viewable), add to the mix. All of this can happen across display or video advertising, and the problems posed by measurement and tracking of such activity make it difficult to achieve a true sense of the scale of the problem.
What can be done?
Faced with this, what can ad-tech do? To use Brightroll, a video ad platform, as an example: aside from providing its customers with advice on best practice, it works with anti-fraud technology, developed by itself and third-party independent companies, to provide accredited measurement and validation, human curation, and what it describes as “industry leadership and certification”).
Alongside the ad tech companies themselves sit a small number of third-party companies working to reduce ad fraud. Such a wide definition takes in those that specialise in detection (WhiteOps, Spider.IO [acquired by Google in February],Forensiq and MdotLabs), viewability and brand analytics (Moat, Integral Ad Science and Pixalate), and protection (DoubleVerify).
What is being done?
The industry is encouragingly not sitting on its hands, though the actions it is taking are fragmented and piecemeal. Some are treating the symptoms and not the cause. For example, in January the US Interactive Advertising Bureau’s “Traffic of Good Intent Task Force” revealed its Best Practices for reducing “risk to exposure” of traffic fraud. While the move was welcome, the advice provided was largely practical and lacked little fresh insight: buyers should set specific goals for campaigns, manage relationships with “trustworthy” sellers, and address traffic fraud by licensing anti-fraud software; networks should “make a stand” by building collaborative industry relationships, and block both the ads and the payments of so-called “bad actors”. So far, so obvious. This was followed in June by the launch of the Trustworthy Digital Supply Chain Initiative, a joint US effort between the Interactive Advertising Bureau (IAB), the Association of National Advertisers (ANA) and the American Association of Advertising Agencies (AAAA) complete with a “road map” for achieving certain objectives (one of which is to “eliminate fraudulent traffic”).
The UK picture is little better. In late 2013, the Digital Trading Standards Group (DTSG) introduced its Good Practice Principles to bring greater transparency into the UK digital display advertising market. The DTSG comprises trade bodies including the Association of Online Publishers (AOP), ISBA (the Voice of British Advertisers), the Institute of Practitioners in Advertising (IPA) and the UK Internet Advertising Bureau. Some companies have achieved its seal of compliance as a means of reassurance to brands.
While this form of proactive industry collaboration is naturally welcome, it doesn’t come without its limitations. Following reported early wrangling between members, much of the DTSG’s early activity has been concentrated on display, and not video, advertising. A notable, display-related achievement has been the agreement and adoption of industry standards in relation to ad placement next to inappropriate or illegal online content.
Bringing us right update to date: JICWEBS and its online ad anti-fraud initiative
And so to how the industry in the UK is seeking to tackle online ad fraud in this and the years ahead. The UK’s Joint Industry Committee for Web Standards (JICWEBS), the independent body that defines best practice and standards for online ad trading, met in mid-December at the Institute of Practitioners in Advertising (IPA). Among JICWEBS’ members are the Association of Online Publishers (AOP), Internet Advertising Bureau (IAB) UK, the News Media Association, ISBA, and the IPA itself.
December saw JICWEBS hold its first cross-industry technical group meeting designed to reduce online ad fraud. Industry attendees included representatives from the Audit Bureau of Circulation (ABC), comScore, the Financial Times, Trinity Mirror, Reckitt Benckiser, DoubleVerify and Integral Ad Science – a reflective cross-section of all of those across the industry affected, in one way or another, by the issue of online ad fraud.
Coming out of December’s meeting, this group now aims to publish anti-fraud good practice principles for the UK market by July and, beyond that, to announce the first companies to be accredited for meeting industry-agreed standards to reduce the risk of fraudulent ads being served. An interim announcement on progress is expected later this month.
Some final thoughts (for now)
As we argued in our last piece, such an issue is a truly international one, and it’s encouraging to learn that JICWEBS intends to work with relevant US industry bodies to produce a complementary anti-fraud framework to the US market. Entities with an international presence require globally-focused standards, and Decipher’s hope is that these will be developed and enforced with sufficient oversight where required.
We’re unsure about how national and international regulation should work in a marketplace that so easily crosses borders. In financial services, a company has to be licenced to operate in any territory where it takes money from clients, or buys on their behalf. So regulation is driven back to a national territory. We may need to think about something similar for companies wishing to engage in the ad-tech ecosystem.
On these pages, we’ll be keeping a close eye on how this issue is developing. This is simply too big and too important an issue to be ignored, and we hope that these steps will be sufficiently robust to allow the industry to begin fighting back.