Kantar Millward Brown

Point of View

Tracking at the Crossroads

No one wants to hear that the car that has always felt safe and comfortable now needs a major overhaul. When the ride has always been smooth, it's hard to believe that the engine will soon be straining to get the car up hills.

But for those of us behind the wheel of continuous tracking, that's how it feels at the moment. We are still moving forward well enough, but signs we pass along the road are warning of tough conditions ahead. Are the wheels really going to fall off around the next bend?

The drive so far

Continuous tracking was invented by Maurice Millward and Gordon Brown in the 1970s to address specific client questions. Clients commissioned our early studies because they needed insight and actionable advice about marketplace events, such as the launch of new competitors or the start of new advertising campaigns, and continuous tracking enabled them to make informed marketing decisions that helped grow their brands.

In time this longitudinal data also proved its value by revealing the underlying dynamics of how marketing worked. It became clear, for instance, that the majority of ads did not wear out in the way marketers anticipated, and that the most important attributes in a category are often the most difficult to change. Learning built up around the measures and how they could be used to predict the sales impact of marketing activity. Over time, another distinct use for tracking emerged: to monitor Key Performance Indicators (KPIs) on an ongoing basis. Because tracking was proving to be so valuable, the demand for tracking studies increased, and soon they became an essential part of the marketing and research landscape.

Challenges past and present

Tracking continued to be an invaluable tool in the decades that followed. But as marketers relied on it more and more, some problems became apparent. For example, the dual purposes of tracking cited above (tracking advertising as well as monitoring KPIs) created tensions. As KPI output from tracking became part of dashboards and was reported to senior management, the focus on the immediate actionability of tracking data was often relegated to the backseat. This in turn led to questions about the need for such large studies to produce top-line metrics.

Tracking's versatility, when exploited, actually became a drawback. Tracking studies seemed to be convenient vehicles for carrying any and all questions related to marketing, but as the studies were loaded up, they became unwieldy. At the same time, pressure on costs led to a reduction of sample sizes, limiting fast, reliable feedback. Forced to carry more weight with a less powerful engine, the tracking study struggled to perform as it once had.

Tracking also encountered new challenges as the media and research environments changed. For example, a major need has emerged in the last few years: to evaluate the performance of media channels. Existing multipurpose tracking studies can provide a high-level read on two or three major media, but sample sizes and questionnaire space limit the depth of the analysis. Similarly, the broad definition of a tracking sample and the limited sample size make it difficult to give detailed guidance on many digital campaigns.

One of the most recent challenges to tracking is the ready availability of data scraped from the Web. Online data—from social media in particular—provides cheap continuous feedback about brands and their marketing. Thus some advertisers have new reasons to question the value of large-scale tracking surveys.

The view at the crossroads

The signs are clear: The world is changing and research needs to change with it. Respondents are harder to reach, especially those in the most desirable demographic groups, and they don't want to engage with long, repetitious surveys. We need to adjust to that reality.

But not every blinking light warns of a real hazard. For example, unstructured data from online sources is not going to nullify the need for structured survey data. Social media data is crucially important for certain types of brands, such as those that conduct business online, and service brands that have customer or community relationships to manage. But for most brands, coverage on social media is typically at a low level, is often generated by a vocal minority, and frequently relates to events (marketing or otherwise) rather than to the brand itself. While this information has value, if it is evaluated in isolation it may present a distorted and partial view. Businesses need to know what is changing, and a self-appointed online group will not usually provide the consistent frame of reference that is needed to discern if real change is occurring.

The way ahead

To compete in today's fast-moving, competitive, and complex markets, brand stewards need regular, timely, and reliable feedback. Now more than ever, they need to monitor the underlying long-term trajectory of their brands as well as the short-term effects of in-market activity. The question is how to capture this information most efficiently.

Many improvements and modifications have already been made to tracking over the years. In web-based markets, the look and feel of tracking studies have changed enormously. Questions are designed to make interviews more engaging and enjoyable for respondents, and questionnaires have been shortened. Further remodeling is already under way, as interviews on mobile phones need to be shorter still.

But old-style tracking has never been able to cover every aspect of marketing activity, nor was it best placed to do so. As the pressure on questionnaires to become shorter has increased, it has become obvious that there are better ways of tackling some of the important marketing questions.

Reengineering for high performance

We need to think about moving from "tracking studies" to "brand performance programs." A single study can no longer answer all marketing questions, but a brand performance program can employ the individual tools that are best suited to address each issue. To understand how a new ad campaign has broken through, a program can include a short study, executed over two or three days, with a robust sample. To quantify the contribution of individual channels to short- and long-term sales, a program can have a CrossMedia study running over the duration of the campaign with enough questionnaire space to ask the relevant media questions.

A program, however, cannot be a series of disconnected ad hoc projects; the components of the program must provide a platform for integrated storytelling. They should be glued together by the brand, not just conceptually but by consistent brand measures selected by the brand and research teams to address the central questions, such as how marketing activities are expected to influence the brand and what attitudes or ideas about the brand need to be changed. The components of a research program will vary according to brand, category, and circumstances, but an effective program should include the following elements:

A detailed understanding of brand equity

Underpinning the entire program and dictating its components should be brand equity insights that identify the process through which associations build brand equity and how that equity manifests itself in the financial performance of the brand.

This understanding will make it clear what marketing actions need to be taken and what KPIs need to be captured in the ongoing monitoring of brand performance.

A continuous monitor

The second essential piece will be a sleek continuous monitor of the KPIs that signal changes in the health of the brand. For web-enabled studies (via computers or mobile devices), the results will flow automatically to a web-delivered dashboard. Pen-and-paper markets will need more manual intervention but will still be able to input the data to a dashboard via an automated analysis engine. This monitor will cost less than a traditional tracking study, thus freeing up funds to be deployed against other elements of the program.

A single study can no longer answer all marketing questions, but a program can apply the individual tools that are best suited to address each issue.

Insights into channel effectiveness and creative power

CrossMedia studies and digital deep dives can identify the effectiveness of channels. Feedback on executions and campaigns can be provided either continuously or on an intermittent, fast turnaround basis immediately after the start of the campaign. Either method will allow timely adjustments to be made if necessary. A complete picture of the brand activities will need to harness social media data as well as survey data.

Ready for the road ahead

Cars today serve the same purpose as cars 40 years ago. But today's cars look and feel different; they go faster, they're more efficient, and the components and technology that power them have radically changed.

Research solutions must be adapted for the complexity of our current era.

And just as cars have evolved to meet today's driving conditions, research solutions must be adapted for the complexity of our current era. Brand performance programs are, in spirit, totally in tune with the idea that gave birth to tracking 40 years ago. Action oriented and designed to give timely advice on important investment decisions, brand performance programs will provide a set of linked solutions, each solution chosen because it is the best one to answer a specific question. They will harness the latest available technology to be cost-efficient and timely. And because the most crucial factors for the category will be identified early on through detailed brand equity work, the questionnaires that make up the rest of the program can be short and tightly focused.

When designed and implemented effectively, brand performance programs will help brands negotiate the complex interchanges faced at every point of decision-making. Moving smoothly down the highway, through a landscape of challenging and changing conditions, they will carry brands safely and efficiently to profitable outcomes.

Gordon Pincott
Chairman, Global Solutions
Millward Brown