Customer Satisfaction

After All You've Done For Your Customers,
Why Are They Still Not Happy?

Customer Satisfaction is Headed Down in the U.S...

A surprising and troubling new study shows. What's going wrong? And what should companies do to make it right?

by THOMAS A. STEWART
REPORTER ASSOCIATE WILTON WOODS

He (or she) is alpha and omega, the beginning and the end, a business's reason for being and the reason it continues to be. Providing for him is--or should be--the reason we go to work in the morning; worrying about her is--or should be--what keeps us up at night.

Yet how little we know about these indispensable strangers, our customers. Are they happy? Restless? Was it good for them too? Will they come back tomorrow? The answers to these questions profoundly affect any business. Indeed, one could say they define a business. They also define the meaning of economic activity, for in the largest sense an economy cannot be described by adding up how many tons of rebar it makes, how many passenger-miles of air travel it logs, how much wood its woodchucks chuck per hour. All these count (and we count them), but in the final analysis what matters is how well an economy satisfies its customers' needs and wants.

Now we can count that. In the autumn of 1994, the University of Michigan's business school and the American Society for Quality Control, a venerable professional organization of 130,000 quality experts based in Milwaukee, released the first American Customer Satisfaction Index (ACSI). The second year's results are now in, which makes year-to-year comparisons possible. The ACSI tracks customer satisfaction in more than two dozen manufacturing and service industries and several public-sector functions--all in all, about 40% of U.S. gross domestic product. This makes the ACSI the first large-scale, systematic attempt to measure the quality of economic output.

The news is--well, there's good news and there's bad news. The good news is that some industries--most notably domestic automobile makers--have made significant gains in customer satisfaction. The bad news: Overall, U.S. companies satisfied their customers less well last year than the year before, with computers a conspicuous decliner. The likely villains: misconceived penny-pinching, a failure to respond to customers' rising expectations, and insufficiently rigorous ways of tracking customer attitudes. (For details of the results and methods of the ACSI, and a look at companies that are champion customer satisfiers, see the following story.)

The American Customer Satisfaction Index is newsworthy not only because of its results but also because, like any good scorecard, it provides insights into the nature of the game itself. A major feature of the ACSI is that it provides a direct way to include customers among the measurable facts of economic life. Says economist Claes Fornell, a professor at the Michigan business school who designed the index: "At the macro level, you should consider this an economic indicator, like indicators of price and productivity."

Other indicators cannot measure quality. One cannot draw any inferences about quality from gross domestic product, for instance; indeed, poor quality can actually increase it. If a mail-order house sends you the wrong pillowcases, the phone calls and back-and-forth UPS shipments required to correct the mistake add to GDP. The Bureau of Labor Statistics tries to take account of quality improvements in calculating the consumer price index. For example, if a carmaker raises a price but the new model includes new features like airbags or has better fuel economy, the BLS will subtract the estimated retail value of the improvements from the price before calculating the CPI. But the BLS makes these quality adjustments for only a few industries and without direct feedback from customers, the final arbiters of quality. The adjustments then disappear into the CPI, leaving no visible footprints.

Productivity statistics come closer, but again not close enough. Efficiency is at best an incomplete and indirect way to measure quality. Some of the cost of poor work (replacing those pillowcases, for example) shows up in lower productivity figures; shoddy work usually means lower prices, which also leaves its traces in measures of productivity. In its landmark 1992 study of service sector productivity, the McKinsey Global Institute wrote: "In other cases quality differences are not picked up [in productivity data], and in such situations it may make sense for a company to reduce its measured productivity and achieve higher quality." A retailer, for example, might please more customers and make more money if it stocked a wide array of merchandise, even if that lowered its efficiency.

It is also notoriously difficult to measure productivity in service industries, education, and government, where output may be hard to quantify or may have no market. This means that there can be no single yardstick for measuring the productivity of the economy's dominant sectors, and therefore also no reliable inference, even indirect, of its quality. Anyone who has lingered for three hours over lunch at, say, Le Table du Comtat, spectacularly perched on a Provencal hilltop, can tell you that for some industries customer satisfaction is at least as significant a measure of economic value as efficiency is.

Clearly, quality is too important to companies and the economy to be measured only (and imperfectly) by inference. It wasn't lower prices, after all, that allowed Japanese automakers to lay waste to Detroit as mercilessly as Sheridan did the Shenandoah Valley; nor have the Big Three won buyers back by productivity improvement alone. There is also evidence that customer satisfaction can be a leading indicator of economic performance. Although the ACSI is too new to be judged on its ability to contribute to forecasts of the economy as a whole, it seems able to foretell corporate performance: A study by David Larcker of the Wharton School, using data from the National Quality Research Center, which Fornell directs, showed that companies that ranked highest in the first year's ACSI significantly outpaced lower-ranked companies in the stock market.

Considered as an economic indicator, the 1995 results flash a warning. Productivity and GDP rose strongly in the 12 months that ended in September 1995 (3.4% and 3.3%, respectively). But, says Fornell, "there is no evidence in the last year that quality is increasing, except in the manufacturing of durable goods. In services it's down. In the other sectors it's steady." Though the overall decline is small, this is not an index that one would expect to shift abruptly, and the fact that it has moved in the wrong direction is not encouraging. Certainly the ACSI score raises questions about whether, as Federal Reserve Board Chairman Alan Greenspan and others have proposed, the consumer price index should be adjusted downward to account for improved quality in U.S. goods and services. Customers, at least, see no such improvement.

But how can this be? After eight years of Baldrige awards, after so much corporate attention to quality that TQM is an acronym that no longer needs to be spelled out for FORTUNE's readers, after so many tens, hundreds, thousands of thousands of customer-satisfaction surveys by hotels, amrlines, banks, even the New York City Parking Violations Bureau--after all this, how can American consumers, the ungrateful wretches, be less satisfied? Since the customer is always right, has business done it wrong?

In important ways, yes. The ACSI results suggest at least three wrong turns business may have made in its drive to improve customer satisfaction: It has viewed customer service as a cost rather than an investment; it has been insufficiently aware of customers' rising expectations; and it has not yet figured out how to define customer satisfaction in a way that links it to financial results.

Corporations--or those that take total quality management seriously--long ago realized that quality pays for itself. Eliminate defects and you eliminate rework, cut back or cut out inspection, reduce inventory and other drains on working capital. It's less clear that companies know that customer service can pay for itself. Perhaps this is because the rewards of eliminating defects show up in reduced costs, while superior customer service, like advertising, is a cost whose payback is on the other side of the ledger, in growth.

Take computers. Surely here is a group of products whose quality has soared and price has plunged: An Apple Powerbook 145 with a 40-megabyte hard drive and a chip speed of 25 megahertz was priced at $2,400 in 1993; a Powerbook 190 with 500 megabytes, more features, and more than twice the speed costs about $1,800 today. Yet the ACSI shows customer satisfaction with computers fell 3.8% last year. One explanation: As the PC market swelled and individual consumers' share of PC purchases jumped (from 20% in 1993 to 27% in 1995, according to BIS Strategic Decisions), hardware and software makers tried to control burgeoning customer-service costs. Customers fumed while being kept on hold for toll calls to help lines. Yet, says Joseph O'Leary, a customer-satisfaction expert at Arthur Andersen, "by definition these people are not early adopters and need more handholding than earlier buyers did."

Strangers to the consumer market--remember Intel's inept handling of the discovery of a bug in its Pentium microprocessor? --the industry forgot that customer service is not just a cost but as much an investment as product excellence is. Keeping customers is a great way to grow. Studies (such as one done by the Technical Assistance Resource Project for the U.S. Office of Consumer Affairs) show that the price of ecquiring new customers is five times greater than the cost of keeping old ones. Says consultant Frederick Reichheld, a director at Bain & Co.: "You can fill a bucket a lot faster if it's not leaking." Few companies in any industry realize the awesome financial leverage of a loyal customer base (see box).

Rising expectations have also increased the customer satisfaction stakes. Says Gun Dukes, group director in the research department of J.D. Power & Associates, which has studied customer satisfaction in the auto industry since 1981 and in computers (since 1991), airlines (since 1992), and now long-distance phone service: "What makes customer satisfaction so difficult to achieve is that you constantly raise the bar and extend the finish line. You never stop. As your customers get better treatment, they demand better treatment." Expectations are fueled not just by how well a business performs vis-a-vis its direct competitors but also by standards set in other industries. Hubert Saint-Onge, vice president of Canadian Imperial Bank of Commerce, points out that a bank might think it has done a great thing if it processes mortgage applications in three days instead of ten days, but still might not impress a customer who can buy a pair of boots from L.L. Bean at 2 a.m. Sunday, then call his broker and place an order to buy 1,000 shares of Intuit.

Finally, businesses should reexamine how they measure customer satisfaction. What Arthur Andersen's O'Leary calls "the Big Bertha customer survey" is of trivial strategic value; customers' loyalty is won in the trenches. Says Wharton's David Larcker, a professor of accounting: "Measurements of this sort are well known to be unreliable. Customer satisfaction is too complicated to measure by means of an unscientific sample giving you a knee-jerk rating on a scale of one to five." (And have you noticed that those surveys seem designed to monitor employee behavior rather than customer satisfaction? When the front-desk clerk told you your room wasn't ready, was he courteous?)

A test: If you can't demonstrate the link between increased customer satisfaction and improved financial results, you're not measuring customer satisfaction correctly. Happy customers should exhibit at least one of three measurable characteristics: loyalty (retention rates), increased business (share of wallet), and insusceptibility to your rivals' blandishments (price tolerance). The efforts a business takes to increase customer satisfaction should be ones it knows--not just by gut feel but from rigorous study--have a provable effect on those characteristics. If you want your customers to stay longer, you should find out what specific factors--faster delivery? electronic billing and payment? better-trained personnel?--make the most difference in your retention rate, and how much that difference is worth. Almost by definition, this means that measuring and managing customer satisfaction cannot be the private preserve of the market research department. It's the job of the whole business. It is the whole business.

What's A Loyal Customer Worth?

Customers are the most important asset a company has. They are the source from which all cash flow flows (unless your derivatives traders had a good year). You know this in your heart; the chairman writes it every year in the annual report. Surely, then, you have ready answers to some simple questions: What is the value of your customer base? How much is a new customer worth? How much should you pay to keep an old one?

The math to evaluate "customer capital" is straightforward. Finding the data requires legwork, but companies like Lexus and credit card giant MBNA find the effort is repaid many times if it helps keep customers. Says Frederick Reichheld of Bain & Co.: "Raising customer retention rates by five percentage points increases the value of an average customer by 25% to 100%." Here, based on work by Reichheld and the University of Michigan's Claes Fornell, is a summary of how to calculate what a customer is worth.

First, decide on a meaningful period of time over which to do the calculations. This will vary depending on your planning cycles and your business: A life insurer should track customers for decades, a disposable-diaper maker for just a few years, for example.

Next, calculate the profit (net cash flow) customers generate each year. Track several samples--some newcomers, some old-timers--to find out how much business they gave you each year, and how much it cost to serve them. If possible, segment them by age, income,`sales channel, and so on. For the first year, be sure to subtract the cost of acquiring the pool of customers, such as advertising, commissions, back-office costs of setting up a new account. Get specific numbers--profit per customer in year one, year two, etc.--not averages for all customers or all years. Long-term customers tend to buy more, pay more (newcomers are often lured by discounts), and create less bad debt.

Then chart customer "life expectancy," using the samples to find out how much your customer base erodes each year. Again, specific figures are better than an average like "10% a year"; old customers are much less likely to leave than freshmen. In retail banking, 26% of account holders defect in the first year; in the ninth year, the rate drops to 9%.

Once you know the profit per customer per year and the customer-retention figures, it's simple to calculate net present value. Pick a discount rate--if you want a 15% annual return on assets, use that. Apply the rate to each year's profit, adjusted for the likelihood that the customer will leave. In year one, the NPV will be profit — 1.15. Next year, NPV = (year-two profit € retention rate) — 1.152. In year n, the last year in your figures, the NPV is the nth year's adjusted profit — 1.15n. The sum of years one through n is how much your customer is worth--the net present value of all the profits you can expect from his tenure.

This is invaluable information. You can use it to find out how much to spend to attract new customers, and which ones. Better still, you can exploit the leverage customer satisfaction offers. Repeat business--the ultimate measure of customer satisfaction--almost certainly merits bigger investments than you make. Take your figures and calculate how much more customers would be worth if you increased retention by 5%. According to Reichheld, for advertising agencies a 5% increase in retention rates translates into a 95% increase in customer NPV. For credit card companies: 75%. Even software makers, hotfooting after new business in a fast-growing industry, would see a 35% increase in customer value if they lost fewer old accounts.