Browse All Articles

One of our retail clients presents two slides at every board meeting: the first shows all of their marketing initiatives that aren't measurable with ROI and the second shows all of the programs that are. He noted that his performance review is tied directly to the number of programs that he can move from the first slide to the second. Clearly, in an increasingly efficient and accountable economy, the old Wanamaker adage that ‘half the money I spend on advertising is wasted; the trouble is I don't know which half’ is no longer acceptable.

So, how did we get here? And, how can we move toward deterministic measurement to ensure we are investing in channels, publishers, creatives, and placements that drive the highest return?

Art vs. Science

Twenty-five years ago, targeting and measurement were more art than science – with a considerable margin of error. We were doing the best we could with the data and analytics available, but it was challenging to find the right metrics and tools to demonstrate the impact of our marketing spend. Over time, more data sources flooded the market, presenting an overwhelming amount of options. Now we had a new challenge, sifting through the data sets to determine what had relevance.

Technologists stepped in with sophisticated, and fast, systems to help deliver data in more usable forms, and a shift in marketing began to take hold. Early loyalty and CRM systems launched. Many households got check cashing cards. Probabilistic mix models became the source of truth. Once a primarily creative process, targeting, and measurement now had a layer of science. Billions of operational dollars began flowing into data centers with the promise of precise targeting and measurement at scale. CFOs were no longer satisfied with measuring what might be true. They wanted certainty in the quantitative impact of their company’s marketing spend.

But, as marketers, we were challenged to provide this level of measurement. We have gained more insight for sure but getting to absolute measurement is still a challenge. New marketing technologies offer a more sophisticated view of consumer data with more robust analytics, but very few connect the data points in a way that provides us with accurate ROI. For example, geolocation helps us better understand brick-and-mortar sales by linking someone’s location with an assumed purchase. While driving foot traffic is an important part of marketing goals, it is still an ROI ‘guess’ at best. Online, clicks have long served as a proxy for consumer intent to buy, but causality remains a challenge, particularly offline. Both of these technologies got us closer to the end goal, but neither were truly deterministic.

The new currency is currency

Deterministic consumer spend data is becoming more readily available, allowing us to more precisely target consumers and, more importantly, measure campaign effectiveness. As an industry, we are at the early stages of integrating this insight into our digital marketing capabilities. The promise of CRM systems combined with bank transaction data provides us a full wallet picture of spend. This view provides the unique and exciting ability to target consumers and measure campaign effectiveness based on actual spend, not surveys, models, panels or proxies.

The path to deterministic targeting and measurement lies in integrating actual purchase data, with the ability to measure true incrementality. In many of the largest channels, we still measure campaign effectiveness with archaic metrics like effective CPMs, clicks, and ratings points, when we have the ability to measure effectiveness in what matters most to CFO’s…actual dollars.

This blog was adapted from a contributed piece I authored for MarTech Advisor. Visit http://bit.ly/1RLe5Zs to read the full article.

There is no correlation between a click and actual in-store or online spend. That is not a typo, it is a truth that we continue to see across all categories with few exceptions. Unfortunately, while as an industry we have become experts at driving efficient means to drive clicks, these clicks can’t be used to pay down overhead, invest in R&D or increase gross margin. Clicks are not a currency. Currency is a currency.

As marketers, we crave more deterministic methods to demonstrate that our marketing executions work (and, better understand how to fix them when they don't). We believed that clicks, as indicated by a consumer action, were a significant improvement versus the obvious limitations of measurement using circulation, placements, and estimated audiences. A click provided us a way to measure consumer action taken on ads across digital platforms. From these clicks, the first true pay-for-performance metric – cost-per-click or CPC – was born. But, CPC was deceptive. A click is only a proxy for consumer intent to buy, but it doesn’t tell us if a purchase is actually made.Related to CPC, we use ROAS (or Return on Ad Spend) as a way to measure the return of our online campaigns. But, ROAS has two key flaws:Attribution: Who gets the credit for the sale? Is it the last click? Do all clicks get a share? Should we divide sales among impressions and clicks? There are countless attribution companies to help marketers answer these questions, buttheystruggle to deliver a definitive answer.Online to Offline Impact: Since digital marketing is often linked to online sales, companies tend to group them together – that is, digital marketers are evaluated on online sales. But, 92 percent of sales happen in stores, so how do we understand the impact these digital campaigns have on offline sales?Digital marketing teams are held to the KPI of online sales because there are few alternative measures of effect and ROI. As a solution, I’d offer that there is a new currency for measurement: currency.Working with our advertiser clients, we see that campaigns optimized on clicks – as with most programmatic buys – do deliver thousands of cost-efficient clicks. However, as counter intuitive as it seems, our experience is that clickers are not spenders. In fact, they are just as likely to be spenders and non-clickers. Therefore, optimizing on clicks will not provide the same level of actual purchases as that same advertising dollar could generate if targeted and measured based on spend likelihood. We help marketers optimize campaigns based on true sales. Our patented method links online consumer behavior - like exposure to an online display ad, frequency, creative - to online and offline purchases to help marketers accurately evaluate the true sales impact of their efforts. Digital advertising works. If we optimize for currency, returns will increase significantly.Want to read more from about why sales trump clicks as the best metric for marketing effectiveness? Read our full white paper at http://bit.ly/21ReKhH.

We are now in full swing of the New Year and heading into the summer, which means we are at the point of the year where people have actually gotten into a rhythm of committing to their resolutions -- or they have completely dropped them. One of the main resolutions that people commit to is fitness. Through our purchase insights, we found a new fitness trend gaining steam – on-demand and online fitness.

Payments to on-demand fitness services jumped to 7.7% of total spend on workouts last year. This is up from 4.8% two years earlier. We also conducted a migration analysis, looking at users who spent most of their fitness share of spend on big-box gyms and only dabbled in on-demand fitness in 2015, but left their big-box gym in 2016. Users who spent most of their share of spend on big-box gyms and only 12% on on-demand, increased their share of spend to nearly 32% for on-demand fitness after they left their big-box gym.

While we show that big-box gyms still held a considerable share in 2016, approximately 73%, we do show that percentage is dropping year over year.

We shared these fitness purchase insights with The Wall Street Journal and The Baltimore Sun. See The Wall Street Journal story, here, and The Baltimore Sun story, here.