Share with your friends


Analytics Magazine

Executive Edge: The devil is in not having details, so get granular

Analytics data science news articles

Averages can distort your understanding of your business, particularly if there are shared costs and revenues.

Matt Lindsay Advertising pricing strategiesBy Matt Lindsay

Averages lie to you. One of our publishing clients looked at the average sell-through rate of its online advertising inventory and noted it was 70 percent. “We can launch a metered paywall, and as long as we do not lose more than 30 percent of our inventory, the lost advertising revenue should be minimal,” they thought.

What the average sell-through rate did not tell them was their best inventory was sold out while its lower quality inventory was only 30 percent sold. The 10-article metered paywall reduced its inventory across the site, affecting both high- and low-quality inventory, by about 20 percent, which reduced its digital advertising revenue by several hundred thousand dollars. They removed the meter and re-evaluated its paid content strategy.

Averages can distort your understanding of your business, particularly if there are shared costs and revenues. As an example, a company had business and residential customers. Its business customers had high-density deliveries, meaning several items were ordered and delivered at the same time.

Residential customers typically had one or two items per delivery. In estimating customer profitability, they divided the cost of a delivery stop by the number of packages and determined that residential business was not profitable. “Let’s stop making residential deliveries to save the costs associated with that activity,” they thought.

Before they implemented the plan, they realized that the costs associated with those residential deliveries did not go away if the business was no longer there since the costs were shared with the commercial deliveries. The relevant cost metric for the residential deliveries was marginal cost, not average cost. Once the delivery driver was on his territory, the incremental costs of a residential delivery were minimal.

The solution to the problem with averages is to seek granular insights. Granularity, the level of detail present in a data set, is important for understanding the dynamics of a business. In economics, we focus on the margin, the last unit sold, hour worked or dollar earned. In practical application, marginal effects can be approximated by looking at a business in greater detail and moving away from grand averages. In digital analytics, data-capture tools are often designed for overall audience measurement for advertising applications and not for customer analytics. Audience measurement for advertising does not require the level of detail that customer analytics does, and the application of audience measure data to customers can lead to wrong conclusions and bad outcomes.

The economics of baseball fans customer analytics & knowledge

The economics of baseball fans are a good example of the importance of granular detail in customer analytics.
Photo Courtesy of | Rolf Svedjeholm

In one of our favorite examples of the importance of granular detail in customer analytics, we found the economics of baseball fans in a city can vary dramatically. We worked with a major market newspaper and digital publisher to develop the business case for a sports-only digital product in their market.
We initially grouped customers by their preferences for certain sports and estimated advertising and digital audience revenue of each customer segment. We estimated the likelihood of each customer group to subscribe and the potential for lost digital advertising revenue if the sports product was paid versus free.

When we focused on the baseball fans, we noted that the results for the two major league baseball franchises in that city were very different. One team’s fans were national in their distribution, while the other fan base was predominantly local to that market.

The advertising value of the fans outside the market was much lower since local advertisers, who purchased advertising through the direct sales force at much higher effective CPMs relative to programmatic channels, did not value non-local digital impressions as much as in-market advertising inventory. In addition, local fans were much less likely to subscribe to the digital sports product due to their ability to read coverage of the team from other local outlets without paid access.

As a result, the optimal business strategy was for one team’s fans to receive the product free and the other team’s fans to purchase subscriptions. This type of insight would not have been possible if customers were grouped by all sports fans or even all baseball fans.

An important level of detail that is lacking in most digital data is the combination of digital advertising revenue with content consumption by individual customer. This is a result of the way advertising impressions and content consumption data are captured on most websites.

Google Analytics, both paid and free versions, capture content consumption, typically the number of page views and unique users. Both versions of the product do not offer complete user-level detail for all visitors to the site, but the premium version will offer a sampled set of this level of data if requested.

Google DFP and other advertising servers will capture and report data on delivered impressions, CPMs, and click-through rates. The challenge for analysts is that these data sets do not merge easily, if at all, at the level of the individual visitor. The data must be merged at a “lowest-common-denominator” that is typically aggregated to a day-site section or hour-site section.

As the baseball story above demonstrates, this level of data aggregation can lead to some significant mistakes in determining the best revenue strategy for a digital publisher. Do not make the mistake of using aggregated data to make detailed decisions. Instead, get granular! Your baseball friends and residential delivery fans will thank you.

Matt Lindsay, Ph.D., is president of Mather Economics, a global consulting firm that applies proprietary analytical tools and hands-on expertise to help businesses better understand customers and, in turn, develop and implement pricing strategies. Lindsay has more than 20 years of experience in helping businesses improve performance and drive revenue through economic modeling. He holds a doctorate in economics from the University of Georgia.

Analytics data science news articles



Related Posts

  • 72
    Does advertising work? Few will deny that advertising plays an important role in building awareness. The idiom, “out of sight, out of mind,” speaks to the importance of being seen in order to even be thought of. Looking back over the years, however, there’s a strong case to be made…
    Tags: advertising, percent, analytics, marketing
  • 61
    The growing effectiveness of digital media has changed the way marketers interact with customers. New age customers are informed and empowered. The challenge for organizations is to be a part of the customers’ conversations involving their product and to influence their choices. Hence, the adoption of digital channels such as…
    Tags: marketing, customers, digital, customer, analytics, product
  • 60
    The consumer goods industry thrived for years on its ability to please the average shopper. From toothpaste to soap powder, it knew how to give people what they wanted – and how to sweeten the purchase with the right price, a tempting discount or a great deal. But for the…
    Tags: analytics, percent, marketing
  • 58
    Companies across industries admit to the growing importance of data analytics to improve sales and marketing effectiveness and decision-making. However, many struggle to piece together siloed data, properly define the problem or design the solution. As a result, they often fail to realize widespread business impact from their efforts, according…
    Tags: analytics, marketing, business, data
  • 58
    The convergence of marketing technology (Martech) and advertising technology (Adtech), event-triggered and real-time marketing techniques, personalization and the use of contextual clues are the four key forces that point to a data-centric future for marketers, according to a recent report from Gartner.
    Tags: marketing, analytics, customer, advertising, data


Using machine learning and optimization to improve refugee integration

Andrew C. Trapp, a professor at the Foisie Business School at Worcester Polytechnic Institute (WPI), received a $320,000 National Science Foundation (NSF) grant to develop a computational tool to help humanitarian aid organizations significantly improve refugees’ chances of successfully resettling and integrating into a new country. Built upon ongoing work with an international team of computer scientists and economists, the tool integrates machine learning and optimization algorithms, along with complex computation of data, to match refugees to communities where they will find appropriate resources, including employment opportunities. Read more →

Gartner releases Healthcare Supply Chain Top 25 rankings

Gartner, Inc. has released its 10th annual Healthcare Supply Chain Top 25 ranking. The rankings recognize organizations across the healthcare value chain that demonstrate leadership in improving human life at sustainable costs. “Healthcare supply chains today face a multitude of challenges: increasing cost pressures and patient expectations, as well as the need to keep up with rapid technology advancement, to name just a few,” says Stephen Meyer, senior director at Gartner. Read more →

Meet CIMON, the first AI-powered astronaut assistant

CIMON, the world’s first artificial intelligence-enabled astronaut assistant, made its debut aboard the International Space Station. The ISS’s newest crew member, developed and built in Germany, was called into action on Nov. 15 with the command, “Wake up, CIMON!,” by German ESA astronaut Alexander Gerst, who has been living and working on the ISS since June 8. Read more →



INFORMS Computing Society Conference
Jan. 6-8, 2019; Knoxville, Tenn.

INFORMS Conference on Business Analytics & Operations Research
April 14-16, 2019; Austin, Texas

INFORMS International Conference
June 9-12, 2019; Cancun, Mexico

INFORMS Marketing Science Conference
June 20-22; Rome, Italy

INFORMS Applied Probability Conference
July 2-4, 2019; Brisbane, Australia

INFORMS Healthcare Conference
July 27-29, 2019; Boston, Mass.

2019 INFORMS Annual Meeting
Oct. 20-23, 2019; Seattle, Wash.

Winter Simulation Conference
Dec. 8-11, 2019: National Harbor, Md.


Advancing the Analytics-Driven Organization
Jan. 28–31, 2019, 1 p.m.– 5 p.m. (live online)


CAP® Exam computer-based testing sites are available in 700 locations worldwide. Take the exam close to home and on your schedule:

For more information, go to