Warning: A non-numeric value encountered in /home/customer/www/ticevision.com/public_html/wp-content/themes/Builder-Cohen/lib/builder-core/lib/layout-engine/modules/class-layout-module.php on line 505

Scenes from the ARF 2019 AudiencexScience Conference – Day 2

AxS logoDay 2 of The 2019 AudienceXScience conference from the Advertising Research Foundation (ARF) was held April 16th in Jersey City. This annual fixture in the media research industry calendar – a rebrand of the ARF’s 2006-2018 Audience Measurement conference – again brought together many luminaries to shed light on the current state of media measurement.

ARF logo

Detailed Notes for Day 2
See Day 1 Notes here

Below are notes from each of the panels/presentations I attended. These notes are by necessity distilled down based on how quickly I could take notes, so they do not reflect the totality of the presentations or discussions. I apologize in advance to any presenters who feel short-changed, misinterpreted, or misquoted.

Marketing Effectiveness in the Digital Era – Les Binet, adam&eveDDB

Presented results from a long-term analysis of UK data, which has head to four books. Fundamental principle: Brand (building) vs activation (sales). Activation can be high efficiency and high ROI – but while it creates sales blips, it does not build growth. Brand building creates long-term memories – broader reach, different attention, more memorable activities; its decay is slower which leads to long-term build.

How to maximize effectiveness

  • Penetration is always the main driver of growth; reach is king
  • Maximize mental availability; brand awareness/salience/fame
  • Messages vs emotions; rational is used for activation, emotional for brand effects

Invest in share-of-voice; if share-of-voice is greater than share-of-market, then growth. An optimal budget should be 60% brand, 40% activation. The fundamental rules haven’t changed with the emergence of digital. Digital increases efficiency and makes activation easier, but brand building is still more important in the long run.

The Race to Own the Future of TV – Julie DeTraglia, Hulu; Natasha Hritzuk, WarnerMedia; Ali Rana, Snapchat

— AT&T’s sale of Hulu?
JD: no change in near term at Hulu
NH: WarnerMedia are treating their upcoming DTC service as a CPG product, not a tech product. Doing UX research for consumer features. Content discovery and personalization are key attributes.
— Snapchat issues?
AR: they have Discover for storytelling from respected partners. Both scripted and unscripted shows. Content for mobile is very difference from regular TV.
— Future of appointment TV?
NH: Appointment TV now is when people get together to watch, not a set time based on broadcast schedule
JD: viewers want their content on every screen. Most of Hulu viewing is a connected TV in the living room
— Ads?
JD: Hulu wants to offer choice and flexibility like they do for viewing their programs. The ad load is less than regular TV; viewers can choose ads or have interactivity. All this leads to more effective ad environment. Their choice to place ads in “pause” screens was another space they could use without interrupting viewing (since viewing was already paused)

— Measurement?
NH: are we putting the cart before the horse by focusing on developing current measures, when they are working on ad experiences that bypass traditional ads; shouldn’t the measurement match the new experiences, rather than trying to fit new experiences into the old measures?
JD: Measurement needs to be part of the ecosystem. She has to do attribution with different vendors depending on measure needed. Hulu does some attribution now directly so to bypass.
AR: In 5 years, all advertising will be “performance” advertisers (eg, only pays on results, not exposures or impressions)

— Importance of diversity
NH: It’s a given. May need to offer multiple service options to serve all consumers.
JD: Same. Their research covers all types of persons.
AR: Snap has diverse user base and staff.

Seeking a Framework for Measurement – Radha Subramanyam, CBS

Media measurement has historically been about counting, in the future it will need to add outcomes as well. The current state of attribution research is that there are no consistent outcome measures or standards; and the impact of linear TV is underestimated.

The state of counting is it is too complex and still siloed. She wants simplification: a total audience count across all devices that gives total program and commercial audiences.

Philosophy for the future

  • Data comes in all sizes
  • Consumer analytics need to be aligned (survey and passive measures)
  • There is an art and a science to interpret meaning of data – the art focus on storytelling

And apparently, if she’s on your team and she yells at you, it’s a sign she cares.

Exploring the Multiple Dimensions of Attention – MediaScience & Google

What is attention? Desk analysis of existing literature revealed there is attention (in a continuum from Passive to Active) and inattention. There is much academic research on attention but little on inattention.

Attention is the absence of inattention, and inattention can be accurately measured. In the lab, blink duration and eye fixations per second had the highest accuracy in measuring attention/inattention

Within attention the best measures may be dependent on the content viewed, or the intended outcome of the stimulus.

Next steps are 1) a pilot to see if measures of attention translate to ads and 2) confirm the best measures for ad attention.

The Future of Audience-Based Buying  – Comscore

This session was really just a review of OpenAP without any new insights. It was also somewhat ironic as WarnerMedia (Turner), one of the founders of OpenAP, announced three days later it was dropping out of the OpenAP system.

OpenAP is helping network sales teams and their buying partners utilize new datasets. These can be used for planning, buying, posting, and auditing.

Demand for OpenAP has been “limited” but expanding. Despite the free access, the presenters quoted there are about 1,000 individual users signed up.

Consistent segment definitions can be used across network groups with secure segment sharing. It also allows independent 3rd party posting.

From Proxy-Based Optimization to People-Based Optimization – Survata

The problem of proxies. Today optimization is typically against viewability, CPMs, and reach but not against outcomes (such as brand lift).

To enable auto-optimization, need to move from campaign level to persons-level reporting (the latter being modeled). Also need single KPI to optimize against (such as funnel impact).

Can’t use traditional survey research, need “programmatic scale”
Can’t use tradition panel accuracy, need superior data accuracy
Don’t use look-a-like respondents, need causal AI

Cross-Platform Insights Every Influencer Will Cite This Year – Nielsen

This was pretty much a recitation of relevant results from the latest Total Audience Report from Nielsen.

There has been a 182% year-over-year increase in connected TV (CTV) impressions
There is currently about 10 billion(!) hours per month viewing time of CTV in the USA, translating to about 75 hours/month of CTV time among CTV users.
CTV adds about a 16% increment to a P18-49 audience.

Erwin Ephron Demystification Award

Congrats to Leslie Wood!

Brand Purpose and Cinema – NCM, ScreenVision, MESH

Many brand experiences are perceived as neutral, whereas consumers and brands both want “purpose”. This study used Real-Time Experience Tracking (RET), a one week brand experience diary.

Paid brand touchpoints are seen as less engaging and persuasive than owned or earned touchpoints. But paid can be a first step to drive people to the better-received owned/earned experiences.

Cinema cuts through neutrality [as one would expect from an NCM/ScreenVision presentation]. Two thirds of cinema brand experiences were positive, more than any other touchpoint, and was particularly helpful among 18-24 demo. TV & cinema together work even better.

A Levi’s case study was presented. Cinema exposures were 2x more engaged than TV alone; 93% found cinema memorable compared with 71% of TV.

Can Data Privacy Be Good for Brands? – Dan Linton, W2O Group

The risk of harm is real. Examples are physical (such as when FitBit jogging data revealed secret military/CIA bases) and emotional (such as when a woman miscarried but still was followed by baby advertising online).

The California Consumer Privacy Act will have a large impact, and is being followed by similar laws in WA, VT, OR, CT, IL, and TX. GDPR is already impacting the EU.

But GDPR did not kill off digital advertising in the EU. In fact, privacy ethnics are not detrimental but can be a positive differentiator for a brand/ad tech service. There are many positives that can result. These include:

  • Getting ahead of the curve in terms of what data are collected and how – and if any will fall foul of new laws
  • Becoming aware of, and organizing, data streams. Where are they from? Why do we use them? Are they really needed? Where are they stored? Is there PII to worry about?
  • Being transparent will build trust
  • Give consumers a reason to engage and share their data

Presenting the ARF Code of Conduct – Paul Donato, CRO of the ARF

Donato discussed the recently announced ARF Code of Conduct. What makes it different?

  • A focus on research not activation-type data
  • A commitment requires research of terms and agreement
  • A chain of trust between elements of the research process
  • Includes automated, location, and AI-driven research
  • There are monitoring KPIs; the ARF can see how many times the terms have been read and agreed to
  • There is a required annual compliance report

Companies can apply online and it is voluntary. It was made voluntary to avoid company lawyers resisting a more structured commitment.

[Donato completely sidestepped the whole issue of compliance. The code is a nice idea but it has no teeth – there is no active enforcement by the ARF and it’s dependent on someone being a whistle-blower. And the penalty of having the ARF seal rescinded may likely have no effect other than temporary embarrassment]

Too Much Math, Too Little Meaning – Rishad Tobaccowala, Publicis

We are in the 3rd connected age (1st = initial computer/browser based; 2nd = computer + smartphone; 3rd = internet of things, all is connected)


  • Erosion of trust
  • Close-mindedness – we need to do “A/B testing” in our own beliefs, ie consider other viewpoints
  • Rising inequality
  • These are all the dark side of the first two connected ages

Data isn’t missing about what to do to solve many of these issues, it’s the will to implement solutions

Purpose – what are we doing all this for?
Poetry – where is the art/beauty in what is being done?
People – you need to change people or keep them and change their mindset

See Day 1 Notes here

David Tice is the principal of TiceVision LLC, a media research consultancy.
Don’t miss future posts by signing up for email notifications here .  
– Read my new book about TV, “The Genius Box”. Details here . 

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.