January 25, 2023
TiVo Steps Up Metrics, Increases Household Sample
- by David Goetzl @dgoetzl, April 1, 2009
While TiVo may
have been the first to provide advertisers with access to second-by-second ratings that track DVR viewing, challengers are now emerging. Looking to maintain a leg up, the company said Wednesday that
it will increase the number of set-top boxes from which it derives its data by 200,000 -- greatly widening its sample size. The TiVo ratings service, known as StopWatch, has provided advertisers with granular ratings data -- both for live and time-shifted viewing -- on an overnight basis based on 100,000 TiVo boxes. The data is collected anonymously.
This fall, that figure rises to 300,000 as TiVo aims for further insight into viewing patterns for advertisers and networks.
TiVo also said the bump will allow for "stable ratings data for dozens" of cable networks that "have gone unmeasured," due to low distribution levels. Tracking those networks is also a trouble spot with the industry standard Nielsen ratings.
advertisement
advertisement
TiVo said that as it readies to bolster its sample size, StopWatch will soon begin tracking Fox Business Network, History International and the MLB Network. The company, which is best-known for marketing DVRs but is looking to bolster revenues by selling ratings data, said all networks with at least 10,000 viewers from its pool will be tracked. Currently, it measures 93 networks.
A TiVo competitor in data reporting -- cable operator Charter -- provides second-by-second data collected from a pool of set-top boxes in 300,000 Los Angeles-area homes. However, Charter does not track time-shifted, or DVR-enabled, viewing as TiVo does.
Still, cable industry initiative Canoe Ventures is likely to do so from millions of set-top boxes in the not-distant future, while others -- such as DirecTV, Dish Network and even some MSOs -- see potential value in doing so as an added revenue stream. Set-top box data, some may argue, provides firmer information than the industry-standard Nielsen sample.
TiVo's coming 300,000 sample is a subset derived from its 4 million TiVo devices in circulation. StopWatch was introduced in early 2007. The service does not provide viewership data with demo breakdowns.
That is offered through a second TiVo service known as PowerWatch. PowerWatch's reliability may come into question, since data is based on what happens in 20,000 households, where residents have volunteered to participate.
Omnicom Media Group and IPG are among the agencies that subscribe to both services. CBS is the only network TiVo notes publicly as a customer.
Please be aware that this data is not ratings data "as we know it". The data from the STB is "tuning data". TV ratings is typically "viewing data". STB data most closely correlates to HH ratings.
The STB knows that someone in the household has chosen to turn on the TV at that time, switch to that channel at that time and so on. What it doesn't know is how many people in the household were watching. It also doesn't know when a channel change occurs whether this was the reason for a viewing audience change (does Mum leave when Dad switches the football on?) Of course in single person homes this tuning data is a very good surrogate for viewing data - apart from when there are visitors, which is a pretty common occurrence in younger single-person homes. There is also on other situation that can crop up - people leaving the STB on, but turning the sreen off when they go to bed. This however is pretty easy to spot (long continuous tuning) and can be edited out.
Given that STB data is tuning with unknown audiences attached it is actually a pretty blunt instrument albeit based on huge sample bases and at the second-by-second level. This is in contrast to the TV ratings that are a pretty sharp instrument based on small sample sizes but only at the minute-by-minute level.