Let’s get tracking

Editor' note: Elliot Savitzky is senior vice president at TRC Market Research. He can be reached at esavitzky@trcmarketresearch.com.

Brand, advertising and market tracking are vital to the success of any business, whether it be a fast-food franchise or a utilities provider. If you don’t know what’s happening in your particular sector of the market, how can you plan, strategize and future-proof your or any business?

For marketers, those of you entrusted with generating and acting upon market data, it is critical you have a complete understanding of tracking studies.

To help with this, our how-to guide for tracking studies has been divided into three easily digested sections that will provide you with all the information you need. In this issue in Part 1, we will cover the reasons for conducting tracking research and provide detailed descriptions of the principal trackers used. In Parts 2 and 3 next issue, we will cover key metrics, how to control variables, how to analyze data and, of greatest importance, how to effectively report the findings of any analysis.

While methodologies have evolved and adapted – from landline phones in the old days to online panels today – the desire and need to frequently, if not constantly, monitor a market remains necessary to provide relevant input to the executive dashboard and is the standard by which the performance of marketing teams and agencies are evaluated.

So, why conduct a tracking study? Are you concerned about your position in the competitive landscape? Worried about brand equity? Are you about to launch a new ad campaign? Are you planning to introduce a new product or line extension? Have you revised your product formulation? Do you want to test a heavy-spending media plan? Are you going to test the market with a new product?

All of these are valid objectives for conducting some form of market tracking. How you design and conduct a tracking program can differ and these two elements are heavily dependent on the level of commitment, the budget available and the types of decisions to be made.

No substitute

Dependent variables like trial and repeat are successfully obtained from third-party data like panels and scanner data but there is no substitute for asking consumers questions that provide softer data like attitudes, emotions and intentions in order to understand the whys around how people ultimately behave. That is where tracking studies really shine.

Biometrics may one day provide non-conscious feedback on a reliable and scalable basis online but for the foreseeable future we will still be conducting tracking studies in order to find the answers to our important marketing questions. 

By the time you have read through this article, you should have the answers to the following tracking-related questions:

  • How do we go about designing a program that best meets our needs? 
  • When do we start our tracking program? 
  • How frequently should we conduct the research? 
  • When do we stop, if ever? 
  • Who do we target? 
  • How do we find them? 
  • How quickly can we expect to see an ROI on our efforts?

We will be focusing primarily on brand and advertising tracking examples as we continue but the principles apply whether you are conducting tracking that relates to corporate reputation, brand health, longitudinal studies or customer experience (Figure 1).

Two key components

The answers to the above questions largely depend on two key components of your plan: what the objectives of the research are and, for brand and advertising tracking, what the media and spending plans are.

The media plan is dependent on the objectives of the program. If the plan is to conduct heavy up-spending in specific markets, then the media plan needs to reflect increased spending for the brand in those markets, relative to a set of control markets. The objectives, therefore, will dictate when, where and how to conduct your research and how your action standards for success should be created.

The first step is to determine when to conduct your research in an effort to gauge results.

Each type of tracking program listed below can be used to measure the impact of different types of objectives and will dictate the timing for in-market assessment: pre-post; pre-post, test-control; pulse; continuous; longitudinal; digital tracking; social media listening.

Let’s discuss each in some detail.

Pre- and post-wave

In almost every situation, you are going to want to conduct pre-wave interviews in order to establish a benchmark for later performance., such as a heavy-spending program, a new campaign launch, a new product launch or just monitoring new competitive activity. 

One of the most effective ways to measure the impact of your marketing activities is by including control market(s), which will help to diagnose whether any pre-post bounce is truly due to your marketing efforts or if it is simply a result of the market’s normal course of events.

Pre- and post-, test-and-control

Employing test-and-control markets depends on several factors including: 

  • Whether your program is national or regional.
  • If your media plan includes a heavy-up component.
  • Whether you can match other market(s) to the area in which you are tracking. By “matching” markets, you need to consider category development indices (CDI) and brand development indices (BDI) that are similar to the test-and-control geographies. CDI is applicable if you have just entered that market, BDI if you already have an established business. Most often, you are targeting high BDI or CDI markets in order to improve the probability of successfully predicting the response of target groups in those markets.

In the case of any advertising tracking study, the media plan becomes an integral part of the timing of the interviewing. This should go without saying, but many times I have seen advertising testing programs where the ad spending for the brand in question was nowhere near what competitors were spending. There is no real chance of identifying differences in awareness, brand perceptions and, hopefully, behavior if your brand’s message can’t be heard. You don’t have to outspend your competitors for advertising but you do have to be able to break through with a minimal spending threshold or you will never be able to read differences in the key metrics from the pre-wave to the post-wave and/or from test markets to control markets.

Three green boxes showing the types of tracking data.

Pulse vs. continuous tracking 

Whether to pulse the reading or conduct a continuous read of the market will depend again on the media plan. If spending is relatively consistent after the initial launch of a new campaign, you will want to read the market on an ongoing basis, carrying out a minimum number of interviews daily.

There are many advantages to continuous tracking, including the flexibility to read the market before, during and after specific competitive events, PR disasters and/or market crises (e.g., a significant drop in the financial markets or a pandemic). Additionally, continuous tracking can save you from either totally missing out on a one-time event that might occur the week after you collected data, or from including it in your data, which may only have a minor longer-term impact on the market. It also allows you to aggregate the data into rolling periods (rolling four-week, 12-week periods) in order to smooth out the data, which increases the reliability and reduces the risk of obtaining biased results.

Conducting pulse waves is more applicable if your media plan has predictable spikes that render continuous tracking unsuitable. For example, if your plan is to spend heavily each time you launch a new advertising campaign with new executions, or even just a pool-out of an existing campaign for a short duration and then go dark, then there is little benefit to conducting continuous tracking. In this scenario, there will be minor impact on key metrics for the duration of the blackout period prior to new spending. Simply time each wave to be conducted shortly after each heavy spending period and then read the results. 

However, there is no guarantee you will even see an impact, as the timing of the post-wave and the amount of spending can dictate the potential impact on the key metrics that include awareness, familiarity, consideration, usage and brand perceptions. Of course, this assumes your advertising is effective in that it drives awareness and motivates the target to want to purchase the product or service. (Pre-testing is critical before a launch.)

Suffice it to say that you need to spend enough on advertising to break through the clutter and for a sufficient period of time for your audience to notice. There is a ton of evidence to show that there is a lead-lag effect in advertising, so truncating your post-wave tracking can wind up with you totally missing the impact that you were hoping for. Again, timing is critical.

Longitudinal tracking studies

Longitudinal tracking collects data from the same respondents over time. This type of data can be very important in tracking trends and changes by asking the same people questions in multiple waves. It is particularly helpful if you are trying to determine if a targeted group of consumers is either adopting the attitudes and behaviors of the next age cohort they are moving into or if they are taking their attitudes and behaviors with them. Think about how the Gen Z segment might change as they move into the cohort currently inhabited by Millennials.

This is obviously time-consuming and expensive and is more commonly conducted and funded by large government programs but it does have its place in the system of tracking programs.

Longitudinal tracking can also be applied to programs where you are looking to assess if a particular group is adopting new learning or is influenced by messages that are meant to effect change. Think about campaigns that might be designed to positively influence how a group of physicians view new treatments for specific disease states. A pharmaceutical company could conduct a longitudinal assessment of the same group of physicians, one that may be limited in size. They may have been detailed over time and/or received specific articles or other marketing communication that were designed to inform and influence behavior. When the universe is small, it often helps to reach out to this same target group on multiple occasions for the sake of efficiency. When the universe is large, there is little reason to reuse samples as these may often lead to biased results.

And this brings up the important point about the potential dangers of conducting longitudinal tracking programs. It is quite possible that the act of conducting a longitudinal study, among the same sample, can serve to educate that specific population regarding the topic and can bias future waves. Respondents can take it upon themselves to further educate themselves and respond as others might, as opposed to reflecting their personal attitudes and behaviors. 

Over time, participants may stop taking part in a longitudinal study. This is known as attrition. Attrition can result from a range of factors, some of which are unavoidable, while others can be reduced by careful study design or practice. Attrition results in a loss of sample size that in turn impacts sample reliability and the potential to extrapolate results to a larger universe. 

By definition, the value of longitudinal studies builds gradually. However, this means that researchers need to wait for more time to pass before they can answer some key research questions.

Additionally, some of the original questions could appear out of date while the environment and social landscape may have changed. Similarly, the design itself will be a function of what are considered best practices at the time of the initial setup of the program. As referenced earlier, tracking studies have evolved dramatically over the past 20+ years. What makes sense to implement now may seem out of date two years from now. 

Digital tracking studies 

Digital tracking is a technique that has only become prevalent within the past 10 years along with the advancement of digital advertising and digital tracking tools. Digital tracking can either be conducted in isolation or, more effectively, in concert with survey research tracking. 

Access to survey data about what consumers think and their stated intent is essential for marketing, product development and more. However, while these insights are central to understanding consumers, there is a massive opportunity for observed data to complete the picture. It is much easier to understand consumers when you combine what they told you they would do with what they actually did. Advances in digital tracking now enable marketers to accurately gauge consumer behavior and couple those insights with needs and intent to bring the digital consumer profile into focus.

Digital tracking is crucial to understanding whether your target audience has been able to view your messages online and then to track their journey and path to purchase. What the audience is being exposed to (passively or intentionally) is instrumental in determining if your messages are properly targeting the right people. Digital tracking can shed light on the pre-purchase online searches that consumers conduct, establish what media they engage with, what sites they visit and whether they make a purchase. Ultimately, you would like to determine how you can positively affect behavior as consumers make their journeys on the path to purchase.

Work we carried out from 2010 to 2014 clearly indicated that consumers exposed to both digital ads and on-air advertising were much more likely to be influenced than those who were only exposed to one of them. It may seem obvious but we were able to actually identify the level of influence each type of ad had. We accomplished this by conducting simultaneous surveying and digital tracking among the same set of respondents. This is why it is important to consider both approaches when designing a program to track the market. 

This can all be combined with social media listening programs to better understand how consumers are talking about your products/categories, determining the types of segments that exist and key search terms that will effectively reach your target. There are several tools to assist in collecting this data, some free, but most are paid-for services.

Next issue, we will continue our exploration, looking at key metrics, controlling variables, data analysis and the effective reporting of the resulting findings.