Planning feasibility for tracker studies is more complex than it is for ad hoc studies because you need to anticipate the target population of the future and how finding and communicating with them might change over time.
It is essential to fully understand the feasibility of the tracker throughout its lifetime, especially in small markets or with narrow population targets. Even in large markets there may be limited numbers of participants in your key demographics. If feasibility is tight, think about the population segments you will be analyzing. Is it necessary to have an equal number of men and women? Rich and poor? Young and old? Depending on the topic being studied it may not be.
Will the population being targeted become scarcer in the future? This could happen if the target is users of a product which may become less popular over time – a study of home PC purchasers for example.
If feasibility is tight you might consider shortening the lock-out period. Research done by SSI and others has demonstrated how difficult it is to condition people. It may be possible to have some of the same people take the survey again after a few weeks or months have passed.
Will technology change the way you might collect data in the future? For example a questionnaire which is not mobile-friendly may struggle to continue successfully over the long term as more and more people choose to switch to taking surveys on mobile devices. SSI sees over 25% of our panelists choosing to take a survey on a mobile device. Without these people, both feasibility and representivity could be at risk. This is particularly true in international research, where in some countries data collection methods are rapidly evolving from face to face directly to smartphone.
It is important to understand not just whether the sample is large enough to sustain your project, but what the sample consists of: proprietary panels, “river” or web intercept sample, (sometimes called “just in time” sample) or a mixture of these. If multiple sources are used, how are they combined? At SSI, to maximize consistency we blend sample based on broad personality and behavioral traits of individual people, not on individual sources – because sources can and do change over time. (For example the profile of Facebook users a decade ago was mostly US college students – very different from today’s Facebook user population.)
Consistency is the Golden Rule for trackers, but keeping every aspect of the sample identical (specific sample source, mode of questionnaire completion), as consumer behavior and technology changes around it may result in a sample which has only an illusion of consistency. The right sample plan anticipates potential changes and is designed to maximize consistency while managing change.
For more information on successful trackers, see: