COVID-19 Triggers Innovations In Executing Attribution & Analytic Models

  • by , Op-Ed Contributor, November 20, 2020

Day 4 of ARF’s “Attribution & Analytics Accelerator” focused on “Accelerating Recovery” and concluded an intriguing and insightful event.

The five case studies presented supported the various concerns with attribution models -- notably Multi-Touch-Attribution, MTA, Marketing Mix Models, Analytics and the tsunami of data these techniques feed on that were evident from the first three days. 

However, innovations -- especially the speed of delivery of analytics in this extraordinarily complex and technical arena -- have been driven by brands in the COVID ecosystem.

These “real-time” or close-to-real-time results provide the cornerstones to adjusting any marketing dimensions to more rapidly support optimizing ROI and/or brand equity. 

The Session Leader -- Thomas Bauer, Partner at McKinsey & Company -- introduced five brands to explain how they are leveraging models, attribution, experimentation, and other analytics to guide them to a successful recovery during and following COVID-19.



As he concluded, “speed matters” and the presenters confirmed that analysis and insights are being delivered much faster.  This raises the question as to what are the tradeoffs against speed in terms of the validity and accuracy of results especially when there are often serious underlying data quality and availability issues?  

Bill Cramblit, senior manager, precision marketing analytics & optimization, MolsonCoors, and Vijoy Gopalakrishnan, senior vice president product development, media COE, IRI shared a terrific presentation: “Analytics During the Pandemic — Unearthing Inherent Consumer Behaviors and Business Building Opportunities.”

First -- and no real surprise -- the beer category has not followed the slower sales pattern experienced by many CPG categories during COVID.  It is outpacing the rest of the CPG categories significantly! 

Consequently, MolsonCoors needed to address three key questions. “Is advertising still relevant if products are flying off the shelves with reduced advertising?” Yes! With adjustments for the changed video consumption.

“Is this externality (non-normal ecosystem) helping them discover inherent consumer behavior thereby providing them segmentation and messaging opportunities within and post-COVID?”  Absolutely! They are adjusting their targeting and have made their creative brand messages reflective of consumers COVID environment. 

“Is optimization of advertising still relevant?” Yes! Based on responding to shifting consumer behaviors and a process of continuous measurement and optimization. 

Estée Lauder Companies face highly varied COVID-19 scenarios across their global businesses, each of which must take appropriate decisions by brand and by country on the ground.

In Italy, skin care and hair care showed strong increases compared to make-up and fragrances.

“Analytics Insights to business Actions” leaned on predictive consumer insights to leverage “intended future behaviors” via consumer research including social listening. These were triangulated with a variety of other readily available databases to forecast the beauty category and key sub-categories for the next five quarters and to underpin adjustment to “relevant” target groups, creative messages and media channels selected. 

Doug Jensen, senior vice president consumer marketing analytics & innovation insights, enterprise marketing, The Estée Lauder Companies, presented the multi product line, multi-brand, multi country story of this brutally competitive category. 

Scott Grenz, vice president media, GSK together with Steve Simpson, media analytics & insight management global practice lead, Publicis focused on the value of analytics to brand targeting opportunities and subsequent ad investment in, “New-to-Brand Buyer Insights: Advanced Analytics for Creating Engagement and Retention Strategies.” 

GSK profiled new-to-brand buyers that bought their products at the start of the pandemic.  Insights were gleaned regarding similarities and differences versus current brand and category buyers that were new-to the-brand in terms of demographic, product purchase and media use characteristics.  These insights enabled GSK/Publicis to reach those various audiences via programmatic trade desks and ultimately retain them with continually tuned custom creative.  Attribution was used to measure the effectiveness against three key target audiences identified as well as the tailored creative alignment for each to inform optimizations and each group’s conversions.  Preparation H and Advil were show-cased which underlined the value of the approach to drive quite different ad strategies for each brand. 

“Rapid Experimentation and Analyses Fuels Rapid Marketplace Response” for Chick-fil-A as presented by Paul DiMaria, marketing effectiveness lead, Chick-fil-A and Mark Garratt, partner and co-founder, In4mation Insights.  Digital sales tripled in the Quick Service Restaurant category during COVID resulting in an explosion of analytics. 

Adjusting MMM to account for, “so much complexity and dramatic shock” and long recovery in the baseline has been a tricky tour-de-force of analytics.  Understanding how media “might operate differently” to support changing logistics is still a work in progress.  For QSR, COVID has become a stress test for both advertising and modeling and the speed of delivering analytics insights have been accelerated.  For Chick-fil-A marketing had an enhanced obligation to inform, breakthrough and satisfy consumers new needs and wants.  It is taking a wide variety of avenues in-stores by daypart, in markets, and in media by market to deliver the bottom line with a close eye on competitive actions. ‘New Rules’ that will constantly evolve!   

Should we “Forget Everything You Thought You Knew”? Judd Eder, vice president of client success, OptiMine and Jamie Glover, director of client success, OptiMine believed so in their presentation, “Adapting Attribution for the Pandemic & The Rapidly Evolving ‘New Normal’ of Recovery - Lessons from the Front Lines.” They suggested that removing historical data from time series-analysis models is critical in such a rapidly changing ecosystem experiencing such massive disruption and change that led to “new” but fragile norms and recovery rollercoasters.

The new starting window?  May 2020.  As real-time analytics and insights based on market measurement and optimization become faster the foundational MMM and data sources will require constant “agile” refreshment.

This evolution will need to be accompanied by “cultural changes in organizations.”

The bottom line after four days examining the broad use of attribution and analytics by major brands? Tremendous progress that needs industry oversight on achieving acceptable transparency and quality to minimum industry standards (TBD) on every element of the approach.

I suggest this can be effectively, efficiently, and appropriately delivered by a collaboration by Sequent Partners, CIMM and the MRC.  

Look forward to industry comments on this position!



3 comments about "COVID-19 Triggers Innovations In Executing Attribution & Analytic Models".
Check to receive email when comments are posted.
  1. Mark Garratt from in4mation insights, November 23, 2020 at 2:57 p.m.

    There has been a lot of talk about how we have to start all over again from March or May 2020. You hear a lot of "past is not prologue" hype. This is not true. There have been tremendous changes, yes, and anyone who was blindly running turnkey industrial MMM is going to be in trouble. The truth is the past is still influential...just less so right now. Two things changed - the baselines and the media impact. Some baseline factors shrunk in importance - like standard macroeconomics or gas prices - others like mobility jumped in importance. Seasonality may have shifted but it still persists. In media, what was already rising up in digital, like CTV/OTT, found new niches and strengthened. Bayesian models have helped us mix the right amount of past data with the current gyrations. Slowly, the past will reassrt itself and will need to be upweighted.

  2. Tony Jarvis from Olympic Media Consultancy, November 23, 2020 at 5:27 p.m.

    Thanks for the valuable added insights into how key historical data will "return" to something like their former influence within MMM.  Per my report from Day 2 of this ARF/Sequent Partners event, the media data generally being used (notably, so called, "viewable impressions" which has no target audience exposure dimension) fundamentally misleads any model and will consequently provide spurious results concerning any "effects" of the media pressure compared to other more significant "effects" drivers, e.g., creative, trade promotions, pricing, brand equity, etc., etc.!  The headline was, "When Media 'Exposure' Is Not Audience 'Viewing' Or 'Hearing' There Can Be No Ad Outcomes!" 
    Perhaps Dr. Jim Spaeth of Sequent Partners will chime in? 

  3. Ed Papazian from Media Dynamics Inc, November 23, 2020 at 7:22 p.m.

    Tony, as you know, you and I plus a few others are on the same wave length on the issue of "impressions" versus actual ad exposure. However, the pressure to do something and do it in a reasonable amount of time underlies this and similar projects and I understand that. But, being fair, the data wont be totally wrong--so there is hope if some adjustment factors are introduced via modelling. As a quess, I would say that approximately 50% of the "ad impressions" whose impact---or presumed impact----will be evaluated are phantoms---meaning that half the time the person counted as exposed wasn't exposed. If this discrepancy could be factored into a parallel analysis of the data, it might reveal that for certain types of considerations---overall audience tonnage, total  reach or demographic targeting, as examples, that it didn't matter as much as it does when the question turns to what's the ideal frequency or how does ad wearout co-relate with impression weight?

Next story loading loading..