David Rodeback's Blog

Local Politics and Culture, National Politics,
Life Among the Mormons, and Other Stuff

Normal Version

Friday, August 14, 2009
The Real Costs of Poor Customer Service

The regional airline which recently kept a planeload of passengers on their plane overnight at a Minnesota airport has me thinking about customer service generally. And something this week has raised both my allergies and my verbosity to almost-irrepressible levels.

The current flap about that Continental ExpressJet flight which kept its passengers imprisoned on the tarmac overnight after a weather-related diversion got me thinking about customer service. It also reminded me of family vacations when the children were younger and our only car was a Chevy Cavalier, but let's not go there.

My own experience in the airline industry has never been as bad as the Rochester story; sometimes it has been very good indeed. But I'm thinking more generally. Here's a long anecdote, then a very short anecdote, then a reasonably short anecdote, and finally a list of customer service principles, most of which, I fancy, the anecdotes more or less illustrate.

I Look Pretty Good in This Tale, but That's Officially Not the Point

Once upon a time, I worked for a small computer software company -- we'll call it Company X -- which built database systems for the real estate and defense industries. For a while, my job on the real estate side was to oversee conversions from old systems to our systems and to be the principal liaison between Company X and the customer during the transition. Typically, the customer was a local or regional real estate multiple listing service (MLS) with hundreds or thousands of members. My barely possible job was to keep them happy until their new system was live; after that it was someone else's job.

After a few years I was laid off, because Company X was struggling and had decided to stop marketing to the real estate industry. A few years passed. Then, just as I was finishing a contract position at American Express and looking for my next gig, the president of Company X called. Was I available?

They had kept and tried to support their existing real estate customers over the intervening years. Gradually, those customers had dwindled. (Industry-wide, the average time between new MLS systems was reportedly an unbelievably short three years, give or take.) Lately, the attrition had markedly accelerated. Former customers were now telling him that his small customer support department had been seriously out to lunch for some time. Phone calls and e-mails -- even calls from executive decision-makers at the customer organizations -- went unanswered for weeks at a time and finally became infrequent. When customers could get through, support personnel often told them that a needed thing could not be done, even if it could be done relatively easily, just to avoid the trouble of doing it. Unsurprisingly, most or all of the remaining customers were making unhappy noises, too.

The president of Company X showed the support department the door and made me an offer. It wasn't a programming job, principally, but it met my needs and suited my circumstances, and he's a friend and one of the real good guys I have known, so I signed on again. I became a one-man support and systems administration department for the remaining seven customers, with periodic trips for onsite user training and occasional brief bouts of programming. That's too many hats for one guy to wear, and it's hard to do meaningful technical work when the phone keeps ringing, but one does what one can, especially when it's for a friend and a salary.

I started answering phone calls and e-mail messages. When people called about one thing, I often asked them if there were other issues getting in the way of their work with our software. I trained them, one call at a time. I modified databases and software configurations to resolve issues and add functionality. I revised and expanded the support-related content on the company Web site. I tested and documented bugs for the programmers, then tested and deployed their fixes. Sometimes, if the phone number or name in a voice mail message was garbled, I even did a bit of detective work to track the the person down and address his or her issue.

Some interesting things happened very soon. First, I received some very happy calls and e-mail messages from people who remembered me. Second, the volume of support calls soared, as word got around that you could actually get help by picking up the phone and calling Company X. Third, the list of unresolved issues shrank to a manageable size. Fourth, the customers found themselves thinking a lot less about finding a new provider. There are only so many hours in a day, and I have to sleep at least part of most nights, so I couldn't do everything for everyone all at once, but I did my best, and the customers seemed grateful.

A few years passed, during which only two of the remaining seven customers left us; we had stopped the hemorrhage. One of the departures was a special case. The other was seriously dysfunctional and had never allowed us to train its users. All of Company X's customer organizations had hundreds or thousands of users, but neither of these departures had a devastating effect on the company's bottom line.

Sea Change: Company Y

Then Company Y bought the real estate side of Company X, and I went with it. The remaining five customers were more than nervous, but stuck around. Later some of them would say that the only reason they didn't go shopping for a new provider immediately was that I was still there support the remaining thousands of users and administering the systems.

Unfortunately, my best attempt at summarizing Company Y's view of customer support is this: Customers should send money and otherwise leave us alone. (That's not quite an exact quotation of management; note the absence of quotation marks.) This attitude had its effects. The quality of customer support at Company Y when I arrived was such that it generally made customers want to leave. And management thought it could pull me away from support and adminstration at will, to program or to design new systems (I enjoy both tasks), and not replace me in the support and administration role, and all without consequence.

Evidence to the contrary notwithstanding, management believed that there "wasn't much of a support load" anyway, and it would be cheaper funnel customers to the Web site, where they could help themselves. In fact, we already had documentation, tutorials, FAQs, and so forth on the Web, but in the real estate industry, for some reason, most users won't use online or printed documentation. They want to talk to a person. I put in as many extra hours as I could, and I found some ways to improve support response times and such. I tried to make some headway against the prevailing attitude at Company Y that customer support was unimportant and an imposition, but without effect.

Finally, Company Y's casual attitude towards payroll, too, made it impossible for me to stay (even had I wanted to). When I gave the boss my notice, he was untroubled. There wasn't much support or administration to be done, anyway, he thought. And, as he told others (strangely, not most of the customers), he had a contract programmer who could "do it in his sleep" in between his other duties. Besides that, he could quickly put a comprehensive list of all issues and their resolutions on the Web, and the support calls to Company Y would dwindle nearly to a very acceptable nothing.

As news spread to the customers that I was about to leave Company Y, they were troubled. I began to receive panicked phone calls from executive officers and others at customer organizations. They had met the boss by this time, and they didn't trust Company Y without me there. They wanted to know if I thought they needed to go shopping for a new provider right away.

Still feeling some loyalty to my employer and even more to the customers, I tried to reassure them by telling them of the things I was doing to smooth the transition for both customers and Company Y, by telling them quite truthfully that my replacement is intelligent and able and will do his best for them, and by saying little or nothing of the troubles behind the scenes at Company Y. In the end, the only thing I could do to talk them down was agree to keep doing some of the administration and even some of the support, as my time permits, as a consultant.

The five customers are about to be three. Things have been so bad since I left that one of those remaining three customer organizations is itself, reportedly, in jeopardy of ceasing to exist. Customers say that Company Y's commitment to quality customer support seems to have dwindled rather badly. They reason, not without evidence, that what they had enjoyed previously must have been my commitment to customer service, not Company Y's.

The story actually gets more interesting, not to mention uglier, from here, and I don't know precisely how or when it will end, but I've told you enough to illustrate my points.

The Very Short and Short Anecdotes, in That Order

Here are two quick, additional anecdotes:

Very Short: A family member's renal failure and subsequent kidney transplant have had us working for more than a year with both our private insurer, IHC, and with Medicare. To make a long, complex saga simple and short, the people at IHC have been consummate professionals, even going out of their way sometimes to help us figure out Medicare. Administratively, Medicare has been a nightmare.

Short: A few years ago I was working at home, using a DSL connection to the Internet which was provided by a major international telecommunications company, which we'll call Company Z. One day it slowed roughly to the speed of smoke signals. I tested everything at my end, and things were fine. Then I started testing what I could see upstream. I was able to identify a consistent packet loss of at least 30 to 40 percent between two IP addresses at the provider; this easily accounts for the near-paralysis of my DSL. I got on the phone with Company Z. I was there a long time. First-level support's eyes glazed over -- I could hear it over the phone -- when I explained the packet loss in simple terms that even I could understand. Second-level support couldn't grasp it, either. Neither level would take my word that I had already done each diagnostic thing through which they insisted on walking me, even though I described the results precisely, so we went through their basic diagnostics repeatedly. Both levels assured me that everything was working fine, according to their diagnostics. I quantified the problem for them over and over again; everything clearly was not working fine. After four hours on hold, in all, and over an hour and a half talking to various level-one and level-two support personnel, I was finally promised a call from third-level support within four hours and given a different phone number to call if needed. That was Friday -- almost all of Friday, in fact.

On Monday morning, more than 48 hours later, still uncalled, I dialed that other number. I was promised a return call within two hours. Late Monday afternoon, still uncalled, I turned to one of the providers on American Fork's municipal broadband system to see what they could do for me. I learned that for only slightly more than I had been paying for DSL, I could get much faster service, and for $10 more they'd throw in that static IP address I wanted. It was after 5:00 p.m.; I asked them when they could hook me up. I was astonished at the answer: 11:00 a.m. the next morning.

They were early. By 11 a.m. I was hooked up, fully operational, and back to work.

I cancelled my account with company Z, and they billed me $200 for early termination of my contract, even though (a) I was well past the early termination period, and (b) they had assured me that they wouldn't have billed me for early termination under the circumstances, even if it had been early. About a month and three phone calls later, that charge was reversed. I no longer do business with Company Z.

Some Principles of Customer Service

Based on these and other experiences on both sides -- customer and service -- I suggest the following principles:

  • Good customer service costs the company which provides it less money than bad customer service, in the long run. The immediate costs of bad customer service may be lower, but in the end it costs the business its customers.
  • Customers frequently make decisions to do business, or to keep doing business, with a company based upon the quality, real or perceived, of that company's commitment to customer service. To a degree, customer service matters more than features, and even more than price.
  • Good customer service requires someone on the ground who not only knows the rules, but also has the judgment and authority to apply the right rules to a given situation. (In the Rochester aircraft matter, the villains keep citing rules and regulations, some of which may or may not have applied, but none of which keeps them from looking inept and inattentive.)
  • One occasionally sees it, but don't expect great customer service from government at any level. The incentives are wrong in most cases. Then again, private industry struggles with these matters, too.
  • When customers stop calling customer support, it's a bad sign, not a good one. It doesn't mean your product is flawless and perfectly intuitive. It means they're about to stop being customers.
  • An extensive support Web site is not always an adequate substitute for a phone call.
  • A phone call is not always a good substitute for a face-to-face meeting. A lot of customers still need to see a face and shake a hand occasionally, in order to feel well tended, no matter how spiffy your online or phone presence might be. (The word "spiffy" is brought to you as a public service by one of my teenagers.)
  • When you do visit, slick face-to-face handling, a firm handshake, and lots of hype are poor substitutes for candor, competence, and keeping your word.
  • When I call customer support, I don't really care where in the world you're sitting. I don't even care that your English is heavily accented, as long as you slow down enough that I (a native speaker accustomed to foreign accents and foreign languages) can understand you despite the accent. If you don't slow down, I won't call back, and sooner or later you and your colleagues will be out of a job.

In the unlikely event that someone from Company Y happens to read this, I close with this note: The advice and analysis are free, but I still want the back pay.

Normal Version