News Stay informed about the latest enterprise technology news and product updates.

To index or not to index, that is the question

To index or not to index, that is the question

After navigating all the elements of your customer loyalty program - the qualitative and quantitative research, the planning, the design, and the launch - your expectation is that targeted customers, i.e. the ones you truly want to retain, will modify their purchasing behavior by increasing their loyalty to your company.

The type of loyalty scheme selected may be simple or sophisticated, such as multifunction 'smart card' approaches, free bounce-backs, buy one-get one vouchers, coupons, or gifts, clubs, and the like. It may also include upgraded levels of service, relationship, communication, or enhanced operational execution.

How will you know if your program really works? Will you look for higher revenue and lower customer churn in the short-term, and conclude that you have succeeded? Or will you seek to strategically understand and monitor the impact of your program on customers and their perception of value? And if you take this latter approach, how will you choose to do the monitoring?

Most customer satisfaction research techniques, especially those with weighted indices, will struggle to reveal the relationships between loyalty programs and real customer loyalty. In the United States, J. D. Power and Associates, most notably, has creatively parlayed our national obsession for 'scorecardism', publicly showing superiority over competitors, by creating a series of satisfaction index studies, and then announcing the winners (and often allowing those winners to incorporate their good results into advertising programs and copy points). While Power is most known for automobile industry satisfaction indices, they have also developed them for rental cars, home builders (so far only in the Denver and Phoenix markets), credit cardholders, cable satellite TV, and local telephone service, among others. Just so you won't think we're totally unique with our need to index, Power has also created a satisfaction index for pay TV customers in the United Kingdom.

While Power's Sales Satisfaction Index has wide acceptance by auto dealerships, my automotive clients have, for years, found its value as an owner loyalty barometer questionable at best. As a senior executive at an automotive market research firm noted: "Current measurements such as the Consumer Satisfaction Index do not sufficiently explain loyalty and defection. The top five manufacturers have about the same CSI from year to year."

Our hunger for satisfaction indexing doesn't stop at private industry goods and services. The various branches of our armed services each has one, and Cabinet departments and agencies either have them in place or are developing them. The Air Force web site on this subject shows that its CSI scores were up in 1998 over 1997, and these results allow me to sleep much more soundly each night.

Even the United States Postal Service has one. Several years ago, I had the Postal Service as a client. The fact that no one on their senior analytical staff could demonstrate to me a relationship between changes in their satisfaction index, overall performance, and continued service usage did not seem to deter them at all from endeavoring to apply its results around the country.

The penultimate satisfaction index in the United States, however, is the American Customer Satisfaction Index (ACSI), a purported national economic indicator of evaluations of quality of goods and services from companies and government agencies. Originated by the National Quality Research Center of the University of Michigan in 1994, and co-sponsored by the American Society for Quality, it covers 40-45 percent of the Gross Domestic Product. The ACSI develops a national index, indices for seven sectors of the economy, 40 industries, and 200 individual companies and agencies. Results are published annually, among other places, in FORTUNE Magazine. This is truly a Bernstein symphony among customer loyalty indices. I find its data and conclusions interesting, even absorbing, but not very illuminating.

So, with the abundance of satisfaction measurements and indices around - and, very prevalent in on-line research tools as well - and the many approaches to customer satisfaction indexing available, to what extent will they help us understand how the performance of customer loyalty schemes leverages customer loyalty? Not very much, I would submit.

In previous columns, I've discussed the difference between customer satisfaction and customer loyalty. Satisfaction will measure attitudes, a predominantly transitory, passive, and benign state for a customer, reflecting a relatively low threshhold of their involvement with a company. Achieving true advocacy as a result of a customer loyalty program, or other customer-focused efforts and initiatives, means that the customers believe that one or more elements of the scheme have created desired value. Customers can tell you about performance, value and loyalty - if you ask about it - and then establish a relationship between them when you evaluate their responses. The objective here is to link beliefs, inherent in performance, and the actions and behaviors associated with probable future purchase and recommendation.

We begin by measuring performance and delivery of the loyalty program, which is far more rigorous in nature than satisfaction. Since we have found that any rating less than the highest may indicate a level of attrition, or loyalty breakdown (so that we also rarely use mean data in our analyses), we also seek the verbatim, or anecdotal, reasons behind those ratings as well as any areas of customer complaint. To effectively prioritize program, and other operational and performance, factors driving customer loyalty, we believe research should contain the following elements:

- Attribute performance
- Attribute importance
- Competitive attribute performance
- Best performed attribute/attribute most needing improvement
- Overall performance, and performance change experienced period to period
- Likelihood to repurchase or continue purchasing, as well as anticipated level of purchases and reasons for low intent level
- Likelihood to recommend, and reasons for low intent level
- Complaints, expressed and unexpressed

Other useful information, such as customer demographics, new or modified loyalty program reaction, and response to new concepts and processes, can also be obtained.

Specifically with regard to attributes, we feel it's extremely valuable to have elements of performance which are both tangible (time, precision, accuracy, completeness, cost-related, etc.) and intangible (empathy, trust, image, assurance, etc.), as defined by customers. Attribute definition is done through prior qualitative research.

Finally, our research invariably includes one or more cells of client staff. We ask them to respond to overall performance, elements of attribute performance and importance, complaints, loyalty, and recommendation the way they think customers would. From this information, we can make determinations about the cohesiveness of company culture in support of customers, and identify areas of staff training or process modification Plus, the mere act of participation, and having information shared with them after the research is completed, helps to build staff morale and a sense of involvement.

Our research produces a set of action-oriented predictive, easily interpreted visual models which can be used to identify which attributes leverage customer loyalty, which most need to be improved, which are expected, and which require more focused communication to be considered important by customers.

From time to time, we have also created indices for our clients; however, they are loyalty indices rather than satisfaction indices. They most frequently carry balanced weights of overall performance, likelihood to repurchase, and likelihood to recommend. These three metrics, we feel, are at the heart of any successful loyalty program; and, in addition to the leading-edge loyalty modeling we provide, offer a great deal of guidance with regard to strategy development and refinement. Notice, as well, that - similar to all other areas of customer research we conduct - we do not measure satisfaction at all; and we certainly do not include any satisfaction measures in our indexing.

One American marketing research firm has developed what they term a 'secure' customer index, but it includes an overall satisfaction measure as one of its foundation elements. Because the correlation between customer satisfaction and customer loyalty has proven to be so low and so inconsistent, we believe that the worth of such an index is compromised and diminished as a result.

We have utilized such loyalty indices to provide assistance and support in many areas. Indeed, if a client is about to establish a new customer loyalty scheme or upgrade the elements of its service proposition (quality levels, guarantees, time specifics, contact and communication, etc.) our research will be very targeted and focused on the overall performance, loyalty and recommendation impacts of those initiatives.

Establishing a customer loyalty index, or CLI, before those changes are made can serve as a benchmark, rather than a scorecard, and a barometer for the effectiveness of programs and process modifications.

In one instance, for example, we conducted strategic customer loyalty research for a home products retailing chain, and then created a customer loyalty index for them around overall service, future purchase intent, and likelihood to recommend scores. This was done prior to their introduction of a well-defined customer loyalty scheme.

The scheme involved the chain's best customers, identified by the variety of products and services purchased and the frequency of store visits. They received special vouchers, a priority customer care service line, private sales events, and special financing offers. Training of sales assistants and department managers in providing these customers with exemplary customer service, plus active direct response communication with these customers were also key features.

Research was conducted again following the launch and roll-out of the scheme, and we were able to help determine, on a segment by segment basis, where the program was most effective and where it needed to be refined. Because we also obtained verbatim customer responses on an attribute-by-attribute and overall program performance basis, the research provided added depth and dimensionality to the decision-making.

Our research concluded that, while other incentives were attractive, extra attention and value created in the stores, particularly with regard to time and staff effectiveness, were the principal loyalty drivers. With this insight, we helped develop new training targeted at empowering staff in helping customers. This led to the creation of new sales assistant categories, in which selected staff were given extensive training in attitude and interpersonal skills with customers. These staff members are also eligible for special reward and recognition for outstanding service. Also, we participated in store layout modification so that products were grouped more logically, and we worked with the company's IT group to streamline the check out process.

The result has been significant improvement in the Customer Loyalty Index scores, sales, and amounts spent per visit. In addition, the chain has vaulted ahead of competitors on every performance category; and loyalty-impacting complaints have dramatically declined.

Needless to say, such approaches have been successfully applied across a broad range of consumer and business-to-business product and service situations. Francis Bacon said that knowledge is power. True enough. We also know, however, that customer insight, intelligently applied, can create customer loyalty. This is real power for any company.

Dig Deeper on Customer loyalty and retention

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.