TV sampling techniques pointed at New York as the location of BritBox’ biggest users. However, direct measurement and analysis told a different story and is allowing the service to run its business much more efficiently.
The power of real user data was brought home to me recently after a conversation with Efe Akbulut, Head of Analytics at BritBox. BritBox is a US SVOD joint venture by the UK’s BBC and ITV and recently announced it had reached half a million subscribers. Mr. Akbulut highlighted the difference between Nielsen’s sampled techniques and direct measurement, and how sampling can miss some of the most valuable information of all.
Nielsen data is only an estimate
Nielsen panel data has been the standard by which the television industry has measured itself for decades. The data provided by the company made it possible for the annual broadcast TV ad business to grow to $70 billion in the US. Nielsen obtains its data from a panel of 40,000 homes scattered across the top 210 US television markets. The panel is a proxy for the viewing behavior of all 120 million television households in the US.
According to Mr. Akbulut, the idea of sampling in this way is:
“We can’t connect with each individual, so the best thing to do is to create a sample, diverse with different voices, tastes, and viewing habits, to represent the overall market.”
However, he says there is no perfect estimation and measuring in this way misses important data essential to the success of SVOD services.
Online TV is accurate
Mr. Akbulut says the direct measurement of subscriber behavior by SVOD services has two primary advantages over those relying on Nielsen sampling. It provides:
- Accurate data on the behavior of every service user, and
- The opportunity to turn the data into actionable information to run the business more efficiently.
He says BritBox is enjoying success taking a data-oriented approach to managing its U.S. market strategy, rather than struggling with trial-and-error estimates that might result from traditional TV sampling methods.
Using sampling to find where the customers are
To illustrate his point, Mr. Akbulut described how he used Nielsen-style sampling on the BritBox data to create a heat map of where the service’s subscribers are. The high demand market areas were New York, Los Angeles, Chicago, Philadelphia, and Dallas-Ft. Worth. However, the heat map is affected by what Mr. Akbulut calls the population metric. Nielsen focuses its sampling around the top 210 TV markets in the US. So, the data obtained gives good resolution in those areas but says little about anywhere else.
If Britbox were to try and establish where its super-fans were using this sample data, the population metric would be a serious limitation. As Mr. Akbulut says:
“I have more customers in New York, so when I add up usage, I see more total minutes there.”
Finding out where the most engaged fans are is critically important to services such as BritBox because more engaged fans are much less likely to leave.
Direct measurement shows where the super-fans are
To find where BritBox’ biggest fans truly are, Mr. Akbulut and his data team analyzed individual user’s daily watching routine and calculated average streaming minutes for 2018 in each census region. The results were eye-opening. Even though the East and West coast regions include more subscribers, non-coastal regions of the U.S. – such as Mountain and West South Central – have more engaged fans. BritBox subscribers in these regions watch for twice as long as New York viewers.
If BritBox had relied on the sampled heat map, it might have concluded that New York is its “best” market. However, this regional analysis shows that unexpected markets like Amarillo, TX have the “best” fans.
The ability to leverage accurate user data and analyze it identify key viewing groups is a critical tool for SVOD services. Moreover, as the number of services multiplies, the data becomes even more valuable in the battle to attract and retain subscribers.
Why it matters
Nielsen’s traditional sampling methods introduce distortions in the resulting TV viewing data that obscure important details.
Direct measurement of subscriber behavior by an SVOD service allows it to avoid these distortions.
The resulting information allows a service to run its business more efficiently.