Happy family watching TV, just one form of audience measurement.
In the old days, Nielsen tracked the broadcast TV audience by asking select viewers to write down what they watched in physical watch diaries. That TV audience measurement was inexact, but it was all there was, and the industry relied on it for decades.
Things are different in this age of streaming video. Unlike the broadcast TV market, the OTT audience is large and diverse, spread across a variety of services and platforms. How do you measure that audience to track their size, demographics, and engagement?
Key Takeaways
There is no single company or metric you can use to measure the OTT audience—you have to use multiple metrics to get the big picture
You can track OTT viewership, demographics, engagement, and consumption habits
When measuring the success of OTT advertising, consider the completion rate, attribution tracking, online visits, and brand reach and lift
You should also track OTT streaming quality in terms of availability, bitrate, playback errors, rebuffering, startup time, and video start failure
The Challenge of Effective TV Audience Measurement
Measuring the streaming video audience is essential for anyone developing OTT content, services, and advertising. You need to know as much as you can about the streaming viewer to fine-tune what you offer to them.
Unlike the days when Nielsen ruled the ratings roost, however, there is no single source of truth for OTT audience measurement. Each service has its own internal data, and some release it for public consumption. Still, the OTT industry lacks a single independent source of viewer tracking.
Not surprisingly, Nielsen has tried pivoting from broadcast ratings to online viewer tracking but isn't nearly as dominant with online programming. Some experts point out that Nielsen under-represents streaming viewers today, measuring only around 41,000 households—a tiny fraction of the overall OTT viewing population. Even with the recent launch of its Nielsen ONE measurement service, it's still just one of several companies trying to track online viewers.
The rapid growth of the OTT industry has opened the floodgates for a host of companies offering viewer measurement services. NBCUniversal developed a measurement framework that reveals that, along with Nielsen, there are dozens of companies tracking OTT content and advertising viewership, including:
NBC Universal's measurement framework for OTT.
How to Measure OTT Audience Size, Demographics, and Engagement
Whether you engage one of these companies or undertake in-house audience analysis, there are some key metrics you want to measure. Let's look at some of the more important things to track.
Viewership
The first and most useful OTT metric to measure is viewership—the number of people watching a given service, program, or advertisement. Viewership is often measured directly, as in a certain number of views. (This is how YouTube tracks viewership.) You can also measure viewership in terms of hours viewed, such as Netflix's Stranger Things season four registered 1.5 billion viewing hours in its first month of release.
Another popular viewership metric is monthly active users or MAU. This metric measures the number of unique viewers who engage with a service in a given month. MAU lets you track the change in viewership from month to month, which can be informative.
Subscription services can track their total subscribers at any given point in time. Netflix, for example, had 222 million subscribers at the end of 2019. This is not a good metric for non-subscription services, however.
Demographics
Raw viewer numbers are helpful, but they don't tell you anything about who is watching a given service or program. For that, you need to track viewer demographics, such as:
Age
Gender
Location
Education level
Looking at demographics helps a service better target it's programming. It also lets advertisers better target their ads.
A demographic breakdown of Netflix viewers.
Engagement
Knowing your viewers is one thing. Engaging your viewers is something more. You want viewers to engage with your content, and there are several ways to measure that engagement.
If your content is on a web-based platform, such as YouTube, you can track viewership and engagement on a second-by-second basis. For content distributed via smart TVs and streaming media devices, you must rely on other behavior that indicates engagement, most of which involve social media. These include:
Star ratings
Number of subscribers
Number of comments
Number of likes vs. dislikes
Number of social shares
Consumption Habits
It's also valuable to how viewers are watching your content. To measure viewers' consumption habits, look at the following metrics:
Device usage
Minutes watched
Average session duration
Churn rate
Traffic sources
How to Measure OTT Advertising
Both viewers and advertisers are embracing the advertising-based video-on-demand (AVOD) model. If you advertise on or sell advertising for OTT services, you want to know if people are watching your ads or if they're fast-forwarding through them. You want to know if viewers remember your ads and if they ultimately make a purchase.
Not surprisingly, many companies focus on tracking OTT advertising and marketing. Here are just a few of the many things you can track regarding OTT advertising.
Best practices for measuring OTT marketing campaigns.
Completion Rate
Advertisers want to know if viewers watch their ads or quickly click away. That's why tracking the completion rate is essential. The completion rate for OTT advertising is typically higher than for ads in other media. Still, you want viewers to stick around and see as much of your message as possible. The higher the completion rate, the better.
Attribution Tracking
While a high completion rate is good, you also want to make sure that consumers are engaging with your ads and ultimately making a purchase. Unfortunately, unlike web-based ads, OTT ads aren't clickable, making it difficult to link purchases to ad views directly.
Attribution tracking attempts to make this connection. You can't do this directly, but many programmatic advertising agencies can provide indicative data. For example, you may be able to track ad viewers via the GPS location data on their phones, which could indicate that someone viewing your ad later visited a physical store.
Online Visits
You can also track when consumers visit your website and compare that data to when your ads run on a streaming service. If you see an uptick in website visits within minutes of an ad running on an OTT service, that's a good indication that the ad drove the visits.
Brand Reach and Brand Lift
Other OTT advertising metrics are more traditional and require additional market research. For example, brand reach studies can determine if an OTT ad campaign is reaching an audience relevant to your brand. This type of study uses traditional market research to ask viewers what they're watching and what they've seen. Similarly, a brand lift study can determine if OTT advertising changed the perception of your brand among key consumers.
How to Measure OTT Quality and Performance
In addition to tracking who you're reaching and how it's important to measure how well you're reaching them in terms of streaming quality. It's unfortunate that 92% of streaming viewers express frustration with OTT video—and more than two-thirds get frustrated once a week or more.
Fortunately, you can track several key factors affecting OTT quality and performance—and, ultimately, viewer satisfaction.
Availability
This metric tells you if the content is available when viewers request it. Poor availability means viewers can't stream the programs they want when they want.
Bitrate
Higher bitrate streams provide better quality viewing than lower-bitrate streams. Viewers may experience grainy or pixelated pictures if the bitrate is too low.
Playback Errors
A large number of playback errors can significantly impact viewer satisfaction. Low bitrate, rebuffering, or other technical issues can cause this problem.
Rebuffering
Rebuffering happens when the stream is slowed or interrupted, thus using up the typical 30-40 second buffer and causing stuttering or pauses in playback.
Startup Time
The startup time tracks how long it takes for a streaming program to start after the viewer clicks the play button. Too long a startup time can cause viewers to abandon a program before it even starts.
Video Start Failure
This metric measures the number of viewer attempts to start a program that ends in failure before playback begins. A high video start failure rate can significantly affect viewership.
Bottom Line: Audience Measurement is Essential to OTT's Success
To be successful in OTT distribution, content, or advertising, you need to know as much as possible about the OTT audience. The more you know, the better you'll be able to target your content and advertising and improve your performance. You can also improve performance by embracing Penthera PlayAssure and other solutions designed to minimize technical issues and improve viewer satisfaction. Use all the tools at your disposal to measure and analyze your audience—then use those insights to deliver a superior OTT product.
Contact Penthera today to learn more about our solutions for improving the streaming viewing experience.
