Are Set Top Boxes the New Nielsen?
I’ve read a lot of marketing research lately on Set Top Boxes and how they could possibly be a replacement for Nielsen ratings. Each time my VP of Marketing or my company president contacts me to ask my opinion, I have to deliver a solid “Unsure” answer.
One of the main companies providing this service is Rentrak. The idea is that Rentrak will partner with a Satellite cable provider such as Dish Network and a cable provider (currently they are working with AT&T, I believe), and capture reams of data on what the digital cable boxes that connect to our TV sets are tuned into. The theory is that there are thousands of set top boxes in any given DMA, and in the best of circumstances, competitor Nielsen only has hundreds of boxes.
Here is a situation where more is not necessarily better. For example, although Nielsen only has 610 homes in its Tampa sample at any one time, extensive statistical analysis has been run on our market to prove that number will provide statistically relevant results with low standard of error. Beyond 610 homes, it doesn’t matter if you have 1200 homes or 12,000 homes, you’re not going to significantly improve the standard of error on any particular rating. In addition, I know to a day, the demographic composition of each of those 610 homes, and I know that sample is very closely matched to represent the market demographics, in terms of age, cable/satellite usage, children in home, race and geography.
Nielsen’s Pat Ligouri made an excellent point in a Media Post article this week:
In a DMA where a telco’s STB HHs represent only 8% of the market’s TV HHs, you will not get a complete picture of tuning activity, especially if those STB HHs are clustered geographically or if the telco’s subscribers have specific age, income, or other characteristic skews that don’t mirror those of the DMA.
Secondly, I have concerns about limiting TV measurement to only homes with these digital cable boxes. Again, in the Tampa market, there are over 115,000 homes that still receive cable only over the air (about 6% of the market). Even if we were to suppose that those 6% of homes won’t drastically affect the ratings one way or the other, Tampa is still a very fractioned media market. (See my earlier post about cable advertising in Tampa Bay). Currently, Tampa residents are 56% Bright House cable, 15% VerizonFIOS cable, 15% Satellite only (no cable), 8% other cable providers (such as Comcast) and 6% over the air. In order for me to comfortably accept ratings from a service such as Rentrack, I’d want to see them form partnerships with all of the cable and satellite providers that cover our market, otherwise I would fear highly skewed and inaccurate ratings.
And even IF that were to occur, I’d still want to investigate viewing patterns by TV set. For example, in my own home, I currently have only one set-top box, yet I have four televisions. The single set-top box is connected to our “main” TV, but the other three are connect in the old fashioned cable-straight-to-the-wall mode. One thing we learned during last summer’s digital transition is that viewing patterns by television differ greatly. The kinds of programming, times of viewing day and number of hours I use my main TV are very different from how I view my television in my bedroom, or how my husband uses the TV in his “mancave”. During the digital transition, many people would buy the digital converter box for their main televisions, but neglect the bedroom TV. Correspondingly, viewing levels for the TV dayparts most highly viewed on that bedroom TV (namely early morning news and the late night talk shows like Leno) dropped pretty significantly.
So, while I watch the growth of set top box ratings services like Rentrak with interest, it may be quite some time before I make a recommendation for our company to seriously consider this as a viable ratings option. Wait and see.