How to improve tracking studies and deliver real-world marketing decisions

Editor’s note: Melissa A. Kevorkian is vice president of research and innovation at Directions Research, Cincinnati, Ohio.

In today’s complex and fast-paced marketing environment, tracking is an essential component of managing your brand and maintaining a competitive advantage. Consistency and comparability over time remain critical. Yet far too many trackers unnecessarily sacrifice sensitivity and relevance to the perceived need to be absolutely longitudinally consistent.

For tracking studies to realize their full potential and deliver value in the form of real-world marketing decisions rather than merely reporting results, researchers must use creativity in study design and leverage modern data collection technology and analytical techniques.

Tracking studies are relatively resource-intensive, making them highly visible to management and prime targets for cutbacks if they fail to deliver perceived value. However, their value is easily undermined if, as trackers often do, they deliver period after period of no change, or inexplicable variations in attitudes or usage. Further, the noise inherent in tracking data can lead to critical delays in recognizing actual market disruptions. Finally, when claimed usage data does not mesh with sales, the credibility of the entire study may be tainted.

That’s the bad news. The good news is there are several ways to improve tracking studies to minimize the risk of these problems occurring with your tracker.

Validity: Accurately measuring brand usage

Linking claimed usage data from tracking studies to actual product sales is a critical step in delivering value and credibility from your tracking. Key steps in providing this linkage include:

1. Providing a relevant, category-appropriate time frame for usage responses. Behavior is most accurately measured within a relevant time frame. Fast purchase cycle products and services – such as beverages, snacks and quick-service restaurants – require almost in-the-moment interviewing for accurate behavioral tracking. For products and services with longer purchase cycles, where decisions are more memorable given typically higher costs, behavior can be tracked using longer time spans from purchase. The key is to establish a relevant time basis that allows respondents to accurately describe their behavior.

For example, asking respondents to recall all the restaurants they’ve visited in the past three months places an undue burden on them as many are dining out multiple times every week. A better approach is to tighten the time frame to one that can be more easily and accurately recalled such as the past day or two. Of course this isn’t a one-size-fits-all solution – asking for mobile phone purchases made in the past two days, while producing a highly accurate picture, will result in excessive costs as you comb through respondents trying to find buyers of a product that is typically bought every two years.

2. Grounding the usage responses in a real-world competitive context. Brand decisions are made in a comparative environment, thus a preference structure for brand attribute evaluation is more realistic than a straight brand-by-brand rating structure. To establish a link between attitudes and behavior (brand preference and usage) customers should only be asked about those brands for which they have an opinion. Individual relevant brand sets can be established using a funnel approach beginning with awareness and ending with the respondent’s brand-used-most-often (BUMO). Within this range it is important to capture each respondent’s most- and least-preferred brands, in addition to their BUMO. It is an important distinction as a respondent’s most-preferred brand is not necessarily his or her BUMO.

For example, Starbucks may be my most-preferred coffee, the one I like the best and rate most highly. But I may buy McDonald's most often because of convenience and cost considerations. In this example, convenience and value are deciding factors and if we don't make the distinction between most preferred and BUMO, those behavioral drivers are lost.

3. Capturing brand usage across all outlets. If your category has a meaningful amount of volume in distribution channels that are not well-defined or well-measured, it is critical to frame your purchase and usage questions in a way that ensures all volume is captured. Explicitly asking respondents to consider purchases in convenience stores, other non-scanner outlets and even bars and restaurants will provide better usage information and allow you to create a stronger link to sales data.

A good example is alcohol: A usage measurement that does not include items purchased/consumed at convenience stores, bars, restaurants and even friends’ homes misrepresents the actual usage of that respondent and is less likely to link well to sales.

The best tracking systems have experience validating their measurements to sales across multiple categories. Starting with a validated system is the best way to ensure credibility for your tracking study.

Sensitivity: Better detection of and insight from attitudinal shifts

Historically, a tracking study’s ability to assess attitudes toward a variety of brands on a large number of attributes was constrained by cumbersome data collection techniques and the very real risk of respondent fatigue. Asking a respondent to rate five brands on 20 attributes results in 100 very repetitive questions, as any researcher knows. The advent of online data collection provided some welcome relief compared to the tedium of a phone interview. However, true innovation only arrived with next-generation online data collection technology and the gamification of market research.

Studies that gather ratings and rankings information simultaneously expand the number of brands and attributes that can be included, while also simultaneously improving the accuracy, differentiation and sensitivity of the data.

Figure 1 shows a simplistic example of the information garnered from assessing both ratings and rankings. Your brand ranks lowest on both Attribute 4 and Attribute 5, yet suffers most on the more polarizing Attribute 4.

The sensitivity of these measurements allows marketers to link usage data to brand perceptions, identifying the drivers of usage with greater accuracy. Once you know which attributes drive usage, you can create more finely-tuned product offerings and communication focusing on those key attributes – all thanks to a tracking study.

Cohesiveness: Incorporating other data streams for a broader view

Your brand does not exist in a vacuum and neither should your tracking study. Every marketer has access to a variety of disparate sources of data that together provide a cohesive view of their brand’s health. Company data about sales, promotions, distribution and media are critical to understanding and enhancing your tracking results. Specific distribution plans and ad flighting schedules can be overlaid with the corresponding schedule and geography of tracking results to enrich and explain the findings.

Beyond that, any other available sources of customer feedback should be included in your brand position evaluation. Segmentation studies, ad hoc research studies and even employee satisfaction and engagement studies can all be seen as a part of the larger picture.

Finally, social media has given marketers a rich new source of information about their brands and marketing efforts, if it can be harnessed systematically and comprehensively.

To fuse these external data sources with your tracking studies in a meaningful way:

• Conduct a data audit to determine what data and which metrics are available. The audit also lays the groundwork for identifying any knowledge gaps that need to be captured in the tracking system as well as creating the data integration map. 
• Identify processes for formatting and incorporating these sources into your tracking. Time spent up front will more than pay for itself on the back end when stakeholders are eager to see the results.

For many brands and categories (though not all), millions of consumers are giving you hints about your brand health and competitive position every day through social media – all for free. Your customers and your competitors’ customers are tweeting, blogging and posting about your products, giving you positive feedback about a new brand introduction or negative publicity about a product or service issue.

Many marketers – large and small – are now establishing internal social listening staff or outsourcing this important function. While many of them use the results independently of market research, the savviest marketers are finding ways to marry their social media analysis with tracking work through careful planning and ongoing monitoring.

For socially active brands, incorporating a listening tool into your brand health instrument provides an early alert system, the proverbial canary in a coal mine. This quick identification of marketplace shifts allows you to be poised for rapid turnaround of an in-depth study when social media uncovers a potential issue.

Social media listening may not be representative in the traditional market research sense but it still can provide invaluable insight if monitored appropriately.

In Figure 2, social media chatter is tracked over time, resulting in a control chart. This age-old approach gives new insight when peaks and valleys in chatter can be aligned to marketplace events and survey results. In turn, social media listening can lead to survey questions that further explain what you’re hearing.

Quick response: Take a modular approach

We recommend implementing an ongoing tracking study that provides an established structure to which marketers can add modules to easily access a target audience and address specific ad hoc information needs. The benefits of a modular approach include:

• Speed: If your tracker is continuous, the lead time for getting answers to timely questions is substantially reduced in comparison to a commissioning a custom study.
• Cost efficiency: As opposed to a custom study, the modular approach is significantly more cost-effective.
• Context: Adding a module to your tracking study provides answers to ad hoc questions within the context of established equity results. No need to wonder how the results would mesh with your tracker.

Whether it is to evaluate potential tracker additions, dig into unexpected marketplace occurrences or better understand the motivators behind social chatter, a modular approach positions your tracker as the go-to brand health resource and keeps it fresh and relevant for your end users.