World Association of Newspapers and News Publishers


Why are some advertisers wasting money on unseen campaigns?

World News Publishing Focus

World News Publishing Focus
Your Guide to the Changing Media Landscape

Why are some advertisers wasting money on unseen campaigns?

When you look back over the last 10 years, the lack of standardized metrics has caused the biggest loss to advertisers (not ad agencies), and to quality content publishers. Yes, it is shocking – these are the parties most interested in the process, who should be the most influential and cooperative.

But for some reason, the introduction of standardized metrics has never happened. Advertisers have spent billions on campaigns that have never been seen by anyone, or have been only partly seen. In the race for more page impressions, quality content publishers have been punished by falling prices – and as a consequence, falling content quality – to satisfy a growing Page Views (PV)-measured audience.

Only now has the process related to ad viewability and engagement actually started. We now have the ability to measure user attention, engagement and other more sophisticated metrics much more precisely than the standards used by TVs – which still run on the Gross Rating Point metric built on a statistical audience sample. 

The world we knew was set on Page Views (PVs), Unique Users (UU) and sessions – despite the fact we all knew that those metrics were giving us a distorted and misleading picture.

Page Views – all mistaken clicks, all impressions that were bounced even before the page content was loaded (due to crazy overload of trackers, ads etc.), as well as all bots – all of these formed a significant part of that metric.

Unique Users – frankly, it was brave or blind to call it unique. If you have three devices, you will be recognized as three unique. If you are clearing your devices of cookies, you might be recognized as six or even nine unique. So, how many real unique users do we have?

But today, with much better analytical tools able to collect and process huge streams of data in real time at extremely low cost, publishers and advertisers should fight for much better standards and constantly adjust them to the changing environment. That would strengthen the argument to justify the higher cost per impression for a well-performing content or ad.

Suggestions for text content

Text content is the easiest to monitor and to get a precise picture of visitor engagement, because it requires visible action. Namely, action generated in the form of scrolling, using the cursor or another signal that the window is active.

The Financial Times in May 2015 suggested in a white paper the introduction of a trigger of five seconds on the page, before counting it as a page viewed or as effective time spent on the site.

At the same time, we should think who the users of the metric are:

  • If the metric is aimed at the advertiser, perhaps it should be set at five seconds after the ad is loaded (unless the ad is embedded into publisher content, not the third-party load)
  • If the metric is aimed at the publisher, perhaps it should be set at five seconds after the content is loaded. Of course, each publisher having the appropriate tools can set another metric for its internal use.

ATS – Effective Attention Time Spent

One suggestion is to start the timer only after the visitor has stayed on the page for at least five seconds. Then, if for another 15 seconds there are no movements (cursor, mouse, scroll etc.), or if another window is open and is active, the timer should stop. When the visitor returns, the timer restarts and the process repeats.

That should give us a pretty good sense on visitor engagement, and tell us with relatively high probability how far he or she went with the content.

Of course, to get an even more accurate picture, one can add the speed of scrolling and combine it with the attention time spent (ATS).

For advertisers, the key element should be PV and ATS with the ad. The key element here should be that 80-100% of the ad will be in view for at least five seconds.

That event should trigger the time counter, which would add the time spent on the ad starting from 50% viewability before the 70-100%, and up to 50% viewability after. That would give a pretty good picture of true user exposure to the ad.

Any exposure that won’t generate at least 70% viewability for at least five seconds should be disregarded.

Publishers can and should build their own additional metrics to monitor and test any business or editorial thesis.

Audio and video metrics are much more complex for the measurement of precise visitor engagement. The visitor could be listening to audio (music or interview) while working in another window, or watching video for a long time and therefore reporting “no action.” But even with those limitations, publishers and advertisers should still get a much better picture than the data generated by Nielsen reports.

Just because publishers are posting video doesn’t mean advertisers will necessarily be interested, Horizon Media’s director of digital Alex Stone says. “I don’t think most media planners will bat an eye at the strategy of some of these guys. Prove that you can do it well. Prove that people are engaging and staying with the video.”

Metrics for audio content consumption

Because of the higher complexity of possible actions, any metric should be divided at least into two groups:

  • Regular PV and ATS – for the case when the visitor has spent some time on the site exposed to text content or an ad, but has decided not to press the Play function
  • PV Audio and ATS Audio – for the case when the visitor has pressed the Play function.

PV – it should be counted only if the visitor has spent at least five seconds on the page – whether the visitor opened the audio or not.

AutoPlay – if the page is AutoPlay, the five-second trigger should be implied to count the PV Audio as successful. The same should apply for ATS. In the case of audio only, the mute, other audio or video file open or stop functions should stop the time counter. Metrics such as FullAudio or % of audio should be the standard metric.

If the ad is in text or video format – the in-view metric should apply as well.

The above actions should apply to any audio material there – advertisement or content.

Publishers should collect many other events here, such as turning the volume up or down, Play/Pause, ATS vs AudioTime, and actions after AudioTime has finished etc.

Metrics for video content consumption

Because of the higher complexity of possible actions, any metric should be divided at least into two groups:

  • Regular PV and ATS – for the case when the visitor has spent some time on the site exposed to text content or an ad, but has decided not to press the Play function
  • PV Video and ATS Video – for the case when the Play function was “on” and the video was at least 80% in view. 

PV – it should be counted only if the visitor has spent at least five seconds on the page – no matter whether the visitor opened the video or not.

AutoPlay – if the page is AutoPlay, the five-second trigger should be implied to count the PV Video as successful.

For ATS – it should be counted with the five-second trigger and stop in case the video is not 70% in view or the video window is not the active window. In case of video-only, the pause and any other audio or video file open or stop functions should stop the time counter. Metrics such as FullVideo or % of video should be the standard metric.

In case the video page has a text ad – the in-view metric should apply as well.

Publishers should collect many other events here, such as turning the volume up or down, Play/Pause, ATS vs VideoTime, and actions after VideoTime has finished etc.

There is an obvious truth: that no perfect, ideal, serving-all-purposes metric exists. The point here is to limit the scale of fraud, and to make the playground more transparent to key players – publishers and end advertisers. With Big Data processing and analytics with rational pricing available, as well as last year’s Financial Times and Nikul Sanghvi white paper on the “Cost Per Hour” initiative, there are no more reasons to withhold the minimum standardization process. With that, publishers can reclaim power over their data and the revenue they are generating with their loyal readers.

About the author:
Jarek Gora is an experienced publishing industry executive and entrepreneur. He is an adviser to the Media Development Investment Fund CEO and is a World Association of Newspapers and News Publishers policy experts team member. He is the co-founder of Deep.BI, a Big Data analytics company focusing on real-time insight for digital media companies.

Author

WAN-IFRA External Contributor

Date

2016-05-06 14:24

Author information

The news publishing industry is experiencing transformation at an ever-growing pace, with new policy issues arising as the landscape changes.

We will be examining policy discussions that will define the news publishing environment of the future, the key topics being internet governance, privacy and copyright. Click here to learn more about our work.

WAN-IFRA Media Policy team and experts.


© 2017 WAN-IFRA - World Association of Newspapers and News Publishers

Footer Navigation