Viewability vs. fraud 2016

Next year there has to be a commitment that the industry will trade ethically, writes Forensiq’s David Sendroff and Julia Smith.

The vibe around DMEXCO earlier this year was that ad fraud has overtaken viewability as the number one issue to tackle. This is not to say that viewability no longer remains a key issue affecting ad quality, but if an ad impression is fraudulent, the viewability issue becomes irrelevant.

There is a significant difference between malicious ad fraud and ads simply appearing below the fold, and yet confusion still seems to reign. In fact, while there is an overlap between fraud and viewability – especially in qualifying traffic quality – they are not inseparable.

As a starting point, we need to look at the fundamental differences between fraud and viewability before we can decide which one will take centre stage in 2016.

So what are the key differences?

1. A fraudulent impression is an ad impression that a human could never have seen. This impression is delivered with malicious intent.

2. A non-viewable impression is an ad impression that the user never saw, but this could be due to the site construction or design and/or user behaviour. For example, did the user stay on the page long enough? Did they have a word document open in front of the web page?

3. A sub-standard impression, such as one on a page with unsafe content, is an ad impression that a human may have seen, but that the marketer would rather they hadn’t. This impression has negative ad quality and brand safety implications.
[advert position=”left”]
The difference between viewability and fraud can also be summarised in two distinct ways. The first differentiator is that fraud should be the first level of defence – if the impression is fraudulent, it cannot be viewed, irrespective of the ad quality or position.

The second, and probably most important point, is that measuring viewability on fraudulent inventory can be inaccurate as botnets can fake viewability for all known measurement methods.

However, there is one key development that will increase the importance of viewability within our industry – the increased use of Cost per View (CPV) as a trading metric and the growing trend for publishers to sell and trade against it.

To add further weight to the debate surrounding using CPV to trade, AppNexus now offers the metric on every impression sold on the platform.

Utilising viewability technology from recently acquired Alenty, AppNexus scores every impression and offers advertisers the ability to only pay for viewable impressions. This will lead to publishers having to understand the quality of ads on their sites and use tools to assess viewability.

Publishers have so far been resistant to this new CPV metric for trading, but in 2016 they will have to start accepting it instead of fighting it. If publishers are only getting paid for viewable impressions, they’ll need to start actively looking into the performance of their ad formats and re-building sites to ensure ad formats perform for both the advertiser and publisher.

There is a naive expectation that there should be 100% viewability across the board. This – of course – is unreasonable. The IAB recently proposed a 70% viewable target, however, there are calls within the digital ad industry for a much tougher threshold for the tolerance of acceptable fraud levels, with figures of less than 10% being cited.

In 2016, there has to be a commitment that the industry will trade ethically and deliver low levels of fraud and high levels of viewability as a given. To achieve this, we have to tackle fraud first and improve ad quality second.

Companies that address both of these issues will reap the benefits of leading the way in brand-safe advertising solutions, and as an industry we will continue to gain the trust of advertisers.

David Sendroff is founder and CEO and Julia Smith consultant at Forensiq.