Editor's note: John Goodman is president of TARP, an Arlington, Va., research firm. Colleen Bush is a measurement manager and Eden Segal is a customer service training specialist at the firm.

How can an organization go wrong once it has established the basic infrastructure that lays the foundation for a successful voice of the customer (VOC) process? In TARP's experience the pitfalls of the VOC process do not end once the foundation is in place. In many cases, TARP has found that the research process is misguided and not efficiently designed. This often results in frustration internally and a waste of a company's limited financial resources.

This article builds on a previous article1 that outlined some of the many obstacles TARP has identified from customer satisfaction research assessments and from conducting strategic customer retention studies and tracking systems in 600+ companies over the past 20 years.

TARP has found that, in general, organizations are collecting a greater quantity of data than they can adequately process and use. This usually stems from fundamental shortcomings in the design of the VOC process:

1. Data collection efforts are not supported with sufficient planning, analysis, and translation functions to act on the information collected.

2. Focused objectives are not set to accomplish specific information needs in advance of the research effort.

The result can be a hefty budget for a VOC process that adds marginal value to the organization, positioning it as a prime target when budget cutters are looking for areas to reduce or eliminate costs. However, with proper attention and realignment, a VOC process can be a valuable tool for decision-making within an organization and can offer assistance in pinpointing areas to invest in wisely, in retaining customers, and adding value to the financial bottom-line.

Millions for collection and pennies for dissemination of results

In taking a step closer and working with clients interested in "fixing" their VOC process, TARP has found that all too often organizations are collecting much more information than they can process and use. Time and again, we have seen too much emphasis placed on collecting the survey data and not enough on the allocation of resources to other important research functions that are part of the VOC research process. Fig. 1 below outlines five basic functions that should be part of every VOC research process.

Even though data collection is only one of the five basic functions in the VOC process, many organizations don't devote the required amount of time and resources before or after the data is collected. For instance, month after month, survey data is collected with minimal time and effort allocated to understanding exactly what customers are saying and then integrating the findings into actionable information for decision making. In order to make the data useful, time and effort must be spent on making sense of the data, analyzing and reporting it to the prospective users of the data in an effective manner, and lastly, helping users to take that final step -- acting on the data. Unless the VOC process reaches this final stage, the data collected will be interesting at best and even frustrating for the individuals who receive the reports but don't know how to use the information.

Not separating strategic, tactical tracking of progress, and market research

Further, TARP has found that one of the root causes behind some of the overspending on data collection is that lengthy questionnaires are being employed on a routine basis to collect a range of information to meet countless organizational objectives, which may be better served using a separate initiative. Fig. 2 outlines three types of survey research, each of which has distinct objectives and serves different needs within an organization.

The first box in Fig. 2, "Strategic Baseline" research, should be used by an organization to identify the basic issues and problems that customers are experiencing, the product and service attributes that are most important to customers (key drivers) and to quantify and prioritize what needs to be improved in order to maximize customer satisfaction, customer retention, and foster positive word-of-mouth communication in the marketplace. The survey instrument for strategic baseline research is by nature relatively lengthy because there are many variables that must be looked at before the key issues can be isolated from the not-so-important issues.

TARP recommends that strategic baseline research be conducted once every one to three years. The frequency of repeating the survey should be based on the dynamics of the marketplace that the organization operates in, its ability to mobilize and make changes using the research outcomes, and the customers' perceptions of the corrective actions that have been taken as a result of their feedback. One mistake that many organizations make is using a lengthy questionnaire to collect the same information month after month. Consumer needs and motivators are not likely to change on a monthly basis. It would be more efficient to collect this detailed information less frequently, analyze the data and report the findings and their impacts thoroughly, and then help research users develop action plans to improve the key product and service attributes most in need of improvement.

The second box in Fig. 2 is "Tracking." The task at hand now is to track and measure the success of the action plans that have been put into place as a result of the strategic baseline research already conducted. TARP recommends that tracking be conducted using an abbreviated survey instrument which measures only the key drivers identified in the strategic baseline research and several overall satisfaction metrics (e.g., overall satisfaction, as well as likelihood to continue using the product/service, and likelihood to recommend to others, both measures of customer loyalty). This short survey, ideally one page and no more than two, should be conducted on an ongoing basis (monthly or quarterly) and should be used to monitor the success or failure of the action plans that have been deployed. The resulting data can also be used as a leading indicator to identify potential problem areas that may be brewing.

A major waste of resources is doing tracking without prior identification of key drivers and points of pain. One hotel chain used a four-page tacking survey measuring almost 100 dimensions including satisfaction with the decor of the breakfast room. The color of the wallpaper at 7 a.m. is hardly a key driver of loyalty. The rush to track, and lack of discipline in limiting dimensions to track, inevitably result in an overly lengthy survey, lower response rate, and wasted processing costs.

The third box in Fig. 2 is "Ad Hoc Market Research." The main objective of ad hoc market research is to get new customers. VOC research is designed to help organizations keep a pulse on the satisfaction levels of their current customers in order to increase customer retention and customer loyalty. A common mistake many organizations make is combining VOC research objectives with ad hoc research objectives and trying to collect the data to fulfill both research objectives with one survey instrument.

Organizations are often tempted to use the same survey instrument to gather all of the information that they are interested in obtaining using a single questionnaire. While this may be done under the guise of saving money by collecting it all at one time, it actually leads to inefficiencies. This practice results in respondents having to complete long questionnaires which are usually perceived as neither very focused nor user-friendly. Moreover, an unfocused, lengthy questionnaire generally results in lower response rates which in turn lead to higher data collection expenses because more questionnaires must be fielded to yield a completed return. With these facts in mind, it is clear why ad hoc market research objectives should be coordinated and conducted separately from the VOC process as a separate data collection effort at targeted times on an ad hoc (as needed) basis.

Surveying when other data sources would be better

Another cause of overspending in organizations stems from the practice of using survey data when, in fact, other internal data sources might provide more timely and accurate information but are often erroneously viewed as being academically and analytically unacceptable. Such data sources may include complaint data or other internal metrics that organizations collect and, unfortunately, never capitalize on. For example, if a car is not repaired correctly, there is a high probability that a customer will file a complaint. And why would an airline conduct a survey to ask customers if the flight was on time when they can access the same information with more accuracy using the internal arrival data?

Complaint data must be calibrated occasionally using survey data (e.g., 40 percent of incorrect repairs result in a complaint and 10 percent of dropped cell phone calls result in a complaint). Calibration helps an organization extrapolate the number of complaints received to a rough estimate of the incidence of repairs or dropped calls in their customer base. The advantage to using this alternative data is that it has already been collected and is often much more timely than survey results. For a phone company or home service company, (e.g., pest control) complaint data can actually be used as a leading indicator of dissatisfaction.

If the VOC manager thinks creatively, costs can be reduced by identifying alternative data sources for examining many types of customer issues.

Reallocating a VOC budget

A share of the resources saved from collecting less information on an ongoing basis and utilizing secondary data sources should be reinvested in the other four functions of the VOC research process described earlier and illustrated in Fig. 1. Customers can tell you what their needs are and what needs fixing, but they can't tell you how to fix it or how to re-engineer your processes, products, and services. A good deal of thought and energy must go into analyzing and interpreting the data once it is collected. And instrumental to the whole process is adequate planning and design to make sure that the research goals and objectives are achieved.

Research users should be regularly trained and updated by effective communicators to help them understand the implications of the data, how to conduct root cause analysis, and how to translate the information into actionable plans. Users and translators should meet continuously to discuss and act on the data.

In summary, although data collection is but one stage in the VOC survey process, it is often the one that gets a disproportionate share of available resources. Realigning these resources across all five research functions should result in organizations spending their limited dollars more wisely and on programs that, in the final analysis, facilitate higher levels of customer satisfaction, retention and loyalty.

Notes

1 John Goodman and Colleen Bush, "'Voice of the Customer' Disconnects Still Exist in Most Companies", Quirk's Marketing Research Review, January 1999, pp. 28-31.