Retailers and brands are using the wrong KPIs to make digital and social media buys

Photo: Getty Images/jacoblund
May 28, 2021

The “most used” KPIs (key performance indicators) for digital and social media investments are generally not the “most important” ones, according to a survey of U.S. marketers by the Association of National Advertisers (ANA).

Thirty-nine KPIs across six categories (Audience Measurement, Efficiency, Exposure Counting, Measurement Quality, Outcome, Other) were identified in a survey of 93 marketing professionals conducted in January and February.

The five most used KPIs were:

  • CPM (cost per thousand)
  • CPC (cost per click or interaction)
  • Unique reach
  • ROI/ROAS (outcome versus marketing investment) based on spending or lift
  • Site visits

The five most important KPIs were determined to be:

  • ROI/ROAS (outcome versus marketing investment) based on spending or lift
  • Exposed ROAS (spending and lift, only using valid measured exposures as a base)
  • Brand safety ketrics
  • Customer lifetime value
  • Conversion

The “Media KPIs That Matter” study found that the most used KPIs are primarily efficiency (CPM, CPC) and exposure counting KPIs (unique reach). ANA said, “This suggests that media KPIs that are the most used are mainly focused on the top of the marketing funnel, favoring measurement of the ‘stimulus’ over the measurement of the ‘response.’”

The most important KPIs are based on outcome and measurement quality. ANA said, “This suggests that media management today is being held directly accountable for driving business outcomes, and that the quality of the media exposure is more important than the quantity of the media exposure in driving results.”

ANA saw the findings as “highly encouraging” despite the most-valued metrics being less used.

The most used KPIs include many metrics that have long been industry staples, can be constructed from available data, apply to a wide variety of campaigns and goals and are relevant to almost every situation. The highest value KPIs put a premium on outcome measures that provide reads into more efficient and impactful campaigns.

“There is apparently demand for measures of quality (e.g., brand safety, customer lifetime value) that are seen as important, but not widely in use,” said Scott McDonald, ARF’s president in a statement. “In the meantime, everyone is making do with exposure KPIs that are available, but viewed as less desirable (reach, site visits, impressions, viewable impressions, etc.). It would seem that there is opportunity for companies that can address this gap.”

DISCUSSION QUESTIONS: Do you agree that there often seems to be a disconnect between the KPIs retailers and brands use for digital and social media and their actual effectiveness? What do you see as the most and least effective KPIs used in making digital and social media buying decisions?

Please practice The RetailWire Golden Rule when submitting your comments.
"This is the digital manifestation of an age-old disconnect between how brands and agencies/media companies view the role of advertising."

Join the Discussion!

16 Comments on "Retailers and brands are using the wrong KPIs to make digital and social media buys"

Sort by:   newest | oldest | most voted
Gene Detroyer

If you think this discussion is familiar, it is. It has been going on for the 50-plus years I have been in business. When I started, there was a bit of an excuse because trying to measure outcome was difficult and expensive and often flawed. The default measure was always CPM. It was as if the advertising agencies really didn’t want to be measured on performance. Oh I remember the arguments I had with the agency.

CPM told us nothing back in the Stone Age and it tells us nothing now. The only thing that makes sense to me is conversion. How much do I spend and how many customers do I connect with. It should be measurable to some reasonable degree.

Mohamed Amer, PhD

So true! It reminds me of the century-old anecdote of a drunk guy looking for his keys under a streetlight. A policeman joins him in the search, and after a while, asks the man if he’s sure he lost them there. The drunk responds that no, he lost them in the park but that the light is better here.

Ben Ball

This is the digital manifestation of an age-old disconnect between how brands and agencies/media companies view the role of advertising. Agencies view it as a way to get seen and noticed. Brands view it as a way to sell stuff.

Dr. Stephen Needel

Digital media experts like Joel Rubinson have been arguing for years that the traditional media KPIs (reach and frequency) don’t apply to digital and social. We’ve known forever that proper targeting of media with the right message always trumps frequency. Persuasive messages to potential buyers (Rubinson) and reinforcing messages to current buyers (Ehrenberg) generate behavioral results (sales).

Suresh Chaganti

Measuring full funnel is not easy, but quite possible. That measurement will not be exact because of the interplay of so many attributes. The key is not to get invested so much into accuracy of attribution – Because it will not be 100 percent precise – but look for a collection of metrics. When holistically looked at, they will provide pretty solid insights. Those metrics will start from ROAS, include ROI, CAC, and LTV.

It is far more important to understand the true cost of promotions and true cost of free shipping, than to obsess over just the top of the funnel metrics.

Mohamed Amer, PhD

Yes! CPM says nothing more than impressions and only gross impressions at best. Meaningless for any level engagement or value derived. There’s no way to calculate ROAS from CPM.

Jennifer Bartashus

The rise of retail media groups is closing the gap on KPIs. Many retailers now have internal advertising groups to work directly with brands and to tie outcomes more to the KPIs that matter versus the ones that are most common. Direct access to shopper data and the systems to track actual product uptake give them an advantage over traditional media shops. This trend is unlikely to end anytime soon.

Venky Ramesh

The broader question here is how important it is for marketing to work with other functions to drive integrated value to the consumer, and are KPIs in place to measure the consumer impact that marketing is able to influence?

For example in today’s world, where the supply chain is taking center stage, it becomes all the more important for marketing to collaborate cross-functionally with sourcing teams to communicate product origin, brand story, etc., with R&D teams to meet demands through rapid new product launches and with a supply chain team to ensure what marketing is promising is being delivered. This requires taking a look at marketing KPI from a whole new perspective.

Jeff Weidauer

The use of exposure metrics vs. response outcomes for media shows that marketing continues to be seen as some sort of alchemy with no logical connection between spend and result.

Ricardo Belmar

Exposure measurement has always been easier to generate than true outcome performance measurement. That doesn’t mean retailers and brands shouldn’t push for outcome metrics over exposure if they want to know what really works to convert more shoppers into buyers. Today’s tech and tools have improved dramatically, especially with the help of AI/ML and it is time for retailers and brands to adopt these outcome metrics more broadly. Those that embrace this challenge will be the winners, and we can see evidence of this by looking at the most successful brands today.

Trevor Sumner

I can’t help but equivocate using broadcast metrics like CPM and reach in the age of data and conversion science to the reluctance to implement retail technology, despite the clear ROIs. Fundamentally, those who embrace a culture of technology, data and transformation will rapidly outpace retailers caught in the past.

DeAnn Campbell

This article is right in flagging the importance of measuring outcomes and responses instead of top of funnel engagement. But it’s even more important to measure cross channel impacts. Understanding the lift to online sales that comes from brick-and-mortar, or the in-store sales that are initiated online is a critical and often overlooked metric.

Ananda Chakravarty

Yes, oftentimes there is a disconnect. When it comes to digital and social, views or visits have dominated and in places where outcomes are harder to measure accurately, e.g. marketing, costs have become the driving mechanism. The perfect example is watching a repeated ad on a streaming medium where we just watched the same ad moments ago. Which ad is worth more? The metrics being used don’t roll up to the higher level $ value of customers.

We all know metrics like conversion have so many nuances that only the change has any real impact. Business must take into account the ongoing context, which is often removed from the metric.

The best route is to build a set of metrics or an index based on these to measure performance — this performance must be tied to the set of levers that are being adjusted up or down. If not, either the metric or the data is faulty, and either will mean you can’t make a good decision.

Doug Garnett

There’s no question that the fundamental idea of the question is true. And, it’s no surprise. We have half a century of direct response marketing which already knew all of this — but which the digital enthusiasts ignored.

And yet, the better metrics are also the impossible metrics — the ones which cannot be calculated or estimated well.

In my DR campaigns, we always started from the solid metrics — CPO, CPC, CPM, etc — and made sure that clients knew they didn’t tell the whole story. Then, at specific points, we would attempt to estimate an ROI (for example). However this takes exceptional care. While we’d love to have firm ROIs they do not exist and cannot exist. So they must be approached with caution.

In other words, it’s right that they measure the 5 they DO measure. Where they may be making mistakes is IF they believe those 5 tell a nearly complete story.

James Tenser

The ANA report on marketing KPIs is an important acknowledgement of what I call The Marketing Metrics Fallacy. It states: “Just because a quantity is highly measurable doesn’t mean it is highly meaningful.”

Opportunities to see, views and clicks may be the most notorious examples of this as they are easily measured in CPMs, but very difficult to link causally to customer relationship value and conversions. On the B2B side especially, these are nearly worthless indicators. We report them to prove we have done some degree of work.

Attribution analysis remains the toughest nut to crack in marketing, despite or maybe because of the addition of more digital channels.

Rachelle King

Often, the budget holders that allocate funds for media are well-educated brand managers whose training can be more finance than marketing oriented. To that end, long held KPI’s like CPM and ROI are staples because that is how financial decisions are made. It’s not necessarily how marketing decisions should be made and, herein, lies the challenge.

Brand safety and customer life time value are meaningful but harder to quantify than ROI. While retailers and brands are aware of the need to expand to more holistic KPI’s, those responsible for allocating funds need a sound basis for decisions and there is yet to be a tried-and-true industry KPI for quality of media. There is, however, an opportunity for media investment decisions to better balance quantitative and qualitative KPI’s in the meantime.

"This is the digital manifestation of an age-old disconnect between how brands and agencies/media companies view the role of advertising."

Take Our Instant Poll

Which of the following do you think are the most important media metrics for brands and retailers?

View Results

Loading ... Loading ...