The Complexity of Reach and Frequency in Today’s Fragmented Media Environment

One of the most common questions we get is, “Can you tell us the total reach and frequency of this plan?” It’s a great question, but the answer is more complicated than it used to be. In today’s world, where people use many different types of media, it’s not easy to give a simple answer. Let’s explore why that is and understand more about reach and frequency in this modern media landscape.

Understanding Reach and Frequency

• Reach is the number of different people who see an advertisement.
• Frequency is how often these people see the ad.

In the past, it was easier to measure these because there were fewer types of media. Now, with digital, social, and streaming media, it’s much harder due to how spread out everything is.

Why Old Models Don’t Work Well Anymore

Traditional methods for estimating reach and frequency, like the Beta Binomial Distribution (BBD) model, were made for a time when print and TV were the main media. These methods assume all media types are similar and that ads are shown evenly across them, which is not true today.

Different Audiences

Older models assume all media types reach similar audiences. Today, the audience can be very different on each platform and channel. For example, the audience on a streaming service might be very different from the audience on a social media platform.

Uneven Ad Placement

Older models assume ads are shown the same number of times on each media type. In reality, digital ads are placed in many different ways, with some platforms showing ads more frequently than others.

Negative Numbers

Some old models can even produce negative frequency numbers, which don’t make sense and are not useful.

Newer, Better Models

To solve these issues, new models like the Multivariate Beta Binomial Distribution (MBD) have been developed. The MBD model is better because:

• Handles Different Media Types: It can work with many types of media, each with different audience sizes. This is crucial because media consumption is very diverse today.
• Flexible Ad Placements: It can handle different numbers of ads on each media type, which is more realistic. This flexibility allows for more accurate planning.
• Avoids Negative Numbers: It provides more accurate and sensible estimates, making the data more reliable.

Our Model for Measurement

We use a method called the Total Overlap Method (TOM) to estimate how many unique people see the ads without counting them twice. This method simplifies the process and helps us get a more accurate picture of the total reach. TOM calculates the overlap between different media placements, ensuring we don’t double-count viewers.

Real-World Performance

In a study comparing various models against a database of 440 web media schedules, the MBD model generally outperformed other models in terms of reach estimation accuracy. It had lower Average Percentage Errors in Reach (AER) compared to traditional models like the Binomial Distribution (BIN) and BBD. However, the MBD model still faced challenges in exposure distribution accuracy, performing similarly to models like CANEX and CBD.

Conclusion

Today’s media world is complex, and old methods for measuring reach and frequency don’t work as well anymore. New models like MBD and methods like TOM help us get better estimates. This allows us to make smarter decisions and create more effective media plans.

By understanding these complexities, we can help our partners reach and engage their audiences more effectively. So, next time you think about reach and frequency, you’ll know there’s a lot going on behind the scenes to get those numbers right!

For those interested in a deep dive into the development and performance of the MBD model, check out Yunjae Cheong’s dissertation. It’s a detailed read but offers valuable insights.