Last week in the Brands App
This week’s installment of the weekly blog by Whit.li comes to you from Alan Raphael. Alan is one of our top algorithm specialists. He’s going to take a look at how using baseline information for TV shows brings out hidden gems for potential advertising opportunities.
This week’s sections:
Visualizing Advertising Value: Finding Hidden Gems
In previous posts we touched on the importance of establishing baselines for your brand – an interest may be very popular among the engaged users of your brand, but it may also be just as popular with the whole population. So what can you do with that information? One possible way is to find value in advertising outlets. For this example, we will simplify a bit and assume tv shows are priced perfectly according to popularity, which is directly correlated to the number of people that follow a particular show. Of course, real pricing and correlations can be plugged in for more accurate results. Going back to baselines, if a show is more popular within your brand but still priced according to overall popularity, there will be value. So now, look at some real world examples gleaned from Whit.li:
We took tv shows from the major broadcast networks and looked the Coke brand. In the following graph, the area of each rectangle represents engagement of a show and the shade represents value. So the larger and greener, the better.
You can see some large rectangles representing shows that are very popular for Coke‘s engaged users. However, many of those are dark, indicating that they are also very popular shows overall and so Coke may pay the full price for advertising with them.
Some of the brightest spots for Coke, indicating possible value, are: ABC’s the Taste and CBS’s Vegas.
Of course, there are some questions regarding cause/effect that must be asked:
- Was there already a promotion with the show or is the relative popularity a result of some natural synergy?
- How successful has the show been in reaching and promoting a social audience?
Depending on those answers, this data can also be used to indicate how well certain partnerships have worked, specifically in the social media arena. Going back to the Coke example, let’s look at three somewhat similar talent competitions, American Idol, the X-Factor, and the Voice.
From these results, we see that the X-Factor has the most of Coke‘s engaged users, but American Idol is brighter indicating possible influence of the cross promotion Coke does with that show. This is again why baselines are so important. Without it, one may assume that Coke‘s engaged users lean towards the X-Factor, when it may appear that way just because the X-Factor has more engagements in general.
Other data can be combined to give a more complete picture such as Nielsen ratings for the shows as well as actual pricing data. Means of identifying the intensity of the viewers interest in the show as well as alternative engagement data using other Whit.li data points. But we’ll save this for another post!
Now You Know
We require each brand we profile to have at least 5,000 followers. This is because we need a significant amount of engaged users in order to accurately profile your brand. If we profiled with too few it would not yield accurate results. However, we are working on ways to overcome this and should have a way to profile smaller brands sometime soon.