Using Facebook Topic Data to Optimize Real Time Social Marketing for TV Series

Richard Caudle
8th March 2016 0 Comments

You might have seen our recent blog post where we discussed how a TV channel used Facebook topic data to understand how audiences engage with TV series, and used this knowledge to optimize their real time marketing campaign. In this post we’ll take a look at how the research was carried out using our platform.

Using Facebook topic data to understand behaviour
The way we watch and engage with TV has changed. No longer do we need to wait until we get to work the next morning to express our opinions on last night’s TV, we can instead immediately tell the world our feelings using social media.

This real time engagement with content gives new opportunities to TV channels who can gather real-time feedback on their programming and promotional efforts.

In this instance a TV channel was keen to see which audiences engaged with their TV series, and how and when these audiences engaged. They wanted use these insights to optimize their promotional campaigns.

The TV channel used Facebook topic data to study their audience. Their investigation revealed:

  • which TV series generated the most and least engagement on social media.
  • which TV series resonated with which demographic segments.
  • most engagement took place during commercial breaks.

These findings allowed the TV channel to improve the targeting of their promotional content to the right demographic groups. The channel also began to make decisions in real-time, tailoring promotional content over the course of an episode to resonate with audience engagement on Facebook.

Let’s take a look at how the research was carried out.

Working with PYLON
Before we look at the detailed steps, here’s a quick reminder of how PYLON works in practice.

Screen Shot 2016-03-07 at 4.02.18 PM

You work with PYLON by:

  • Filtering the stream of data from Facebook to stories and engagements (such as likes and comments) you’d like to analyze. Filtered data is recorded into an index.
  • Classifying the data using your own custom rules to add extra metadata for your use case.
  • Analyzing the data you have recorded to the index.

You can learn more about the platform in our What is PYLON? guide. Now look at these steps in the context of this specific use case.

Filtering stories and engagements
The first step of working with Facebook topic data is recording data from a target audience for your analysis.

Using the DataSift platform you can capture stories and engagements on stories by creating a filter in CSDL. The filter specifies what data you’d like to be recorded from the Facebook data source to your index for analysis. The rules in your filter operate against the values of targets (data fields) of the stories and engagements.

For this study the TV channel recorded the audience discussing TV series broadcast on their network.

As an illustrative example this simple filter will capture mentions of some popular TV series:

Screen Shot 2016-03-07 at 4.05.00 PM

This filter makes heavy use of topics which are inferred from the content of posts. For well established TV series we can be confident topics will be inferred, but the filter includes a condition based on keywords to make sure as many posts as possible are captured.

Based on this filter if someone posted the following:

When are they going to make more Sherlock – it’s my favorite series by far!

This story and any likes, comments or reshares on the story will be recorded by the filter.

The TV channel started a recording using a filter similar to the example above to record data for their analysis.

Normalizing data through classification
Facebook topic data is already a rich data set but you can add additional value using classification rules. By adding classification rules to a filter the platform will record additional meta-data for each story and engagement. You can use this additional metadata in your analysis.

In this case the TV channel wanted to analyze the audience based on TV series they engage with. The TV channel used VEDO tags to group posts by the series being mentioned.

Taking the same approach with our example filter, the tags could be defined as:

Screen Shot 2016-03-07 at 4.06.11 PM

The tags help us to group posts by TV series and normalize the data ready for analysis.

When a story or engagement matches the filter conditions the classification rules are applied before the data is recorded to your index. So in this case if the content of a post reads:

When are they going to make more Sherlock – it’s my favorite series by far!

The story will be tagged with “Sherlock” when it is stored to the index.

Finding audience insights
Once you’ve recorded data to your index you can immediately perform analysis using analysis queries.

You can perform a time series analysis to see how an audience engaged over time. You can perform a frequency distribution analysis to quantify the engagement by segments of your audience. A more advanced form of analysis is nested queries where you can segment and quantify your audience by multiple dimensions.

You also have the option of using query filters to filter to a portion of your recorded data before performing analysis. So for example you could use the example tags above and filter to only stories and engagements relating to friends before performing a time series analysis.

With data being neatly classified and recorded into an index, the TV channel submitted analysis queries to investigate the recorded audience.

Which TV series generated the most and least engagement?
Firstly the TV channel looked at which TV series were being engaged with least and most by the Facebook audience.

To perform this analysis the TV channel made use of the tags they had created for the TV series and analyzed their recorded data using a simple frequency distribution query.

We could carry out the same analysis, for example in Python:

Screen Shot 2016-03-07 at 4.08.50 PM

Note the target being analyzed (beginning with interaction.tag_tree) is the result of our tagging rules we created.

Plotting the result of the query shows us how much engagement is taking place for each TV series.

Screen Shot 2016-03-07 at 4.09.24 PM

The example query analyzes the entire set of recorded data. We could ‘zoom in’ to precise time segments if we choose for more detailed results.

Which TV series generated engagement for which demographic groups?
Secondly the TV channel looked at which TV series were being engaged with broken down by demographic segments.

To perform this analysis the TV channel used a nested query which broke the audience down first by series, then by gender and finally age group.

We could carry out the same analysis, for example in Python:

Screen Shot 2016-03-07 at 4.10.01 PM

Visualizing the results gives the following chart:

age-gender-breakdown-series

This analysis reveals which demographic groups engage most with each series and provides insight for better targeting of promotions.

When did the audiences engage with social media?
Finally the TV channel wanted to learn more about when their audience was active on Facebook so that they could optimize real time promotions.

This investigation was performed using time series analysis. Keeping with our example we could perform the same analysis.

We can analyze the level of engagement during an episode by analyzing the count of interactions in each 2 minute interval:

Screen Shot 2016-03-07 at 4.12.46 PM

Notice we have used a query filter to select only interactions that mention the Sherlock series. submit three analysis requests, one for each series. We have also set a time period for the analysis using the start and end parameters so that we analyze only the period of the broadcast

Plotting the results of this analysis shows that audiences engage during commercial breaks, rather than during the program itself. This is evident from the recurring peaks and troughs in the timeseries result.

Screen Shot 2016-03-07 at 4.13.19 PM

Learn more…
PYLON for Facebook Topic Data gives analysts access to a vast new audience to test their assumptions and to inform better decisions.

To learn more about the platform take a look at our What is PYLON? guide.

Also, keep an eye on this blog for more Facebook topic data use cases which we’ll be posting soon.

 

 

Share This