We don't often pull back the curtain to disclose the sausage-making process that goes into producing these columns. In most cases, the process of writing columns is just not that interesting. But when it entails a major car company, a cranky local dealer, a deflecting attempt to blame IT to hide incompetence, and valuable lessons, pull back the curtain we will.
The story began last week, when the back window lifting mechanism on my wife's car suddenly failed, preventing her from closing one of the windows. It needed to be fixed or replaced. She took it to our local dealer, and I met her there to talk to the service agent and bring her home, since the car would have to be in the shop overnight.
The service manager seemed very clear on this: Storing records "takes space" and if the records are what we in IT call "warm" it's too challenging or costly or whatever to keep them around.
Since 2005, I've bought three new cars from this dealer. We live at the bottom of nowhere, so we don't have a lot of dealer choices. We're not thrilled with this dealer, but the car has actually been surprisingly reliable for one that's six years old, so we soldier on.
Before she brought the car in, my wife called the service department, as she has once a year or so for the past six years for the various routine repairs and maintenance her car has needed. This time, though, they couldn't find any records of her, me, or the car. She spelled her name to the service manager three times, more slowly and more carefully each time. She spelled my name to him as well. She then read him the VIN number.
It turns out they had a record of the vehicle's characteristics from the VIN number, but they had none of the car's previous service records. The last time we brought the car in for service was November 2014 (a little over a year ago), and according to the service manager, since the car wasn't active in their system, they purged the records to make room for active customers.
The premise, and the service manager seemed very clear on this, was that storing records "takes space" and if the records are what we in IT would call "warm" (because 14-month old records are definitely not yet at "cold"), it's too challenging/costly/resource-intensive/whatever to keep them around.
An IT perspective
So let's examine this limiting design statement from an IT perspective. First, there's the cost of storage. Then there's the time to do a record retrieval. Finally, there's that old IT mistake of making the indexing value too small to scale, so when you have too many items, the index breaks.
Let's take each of these items in reverse order. There is no longer any excuse for programmers to build non-scalable indexes. Scaling has been a problem that we've been aware of for decades now, so making sure your system can handle growth is a Design 101 sort of skill.
Consequently, there's no reason to limit the number of customer records because of the customer ID index or some other sort of code number.
What about retrieving records? After all, if you have to paw sequentially through a hundred records, that should take less time than searching through a thousand records, right?
Uh, no. Only the rankest of systems design amateurs do sequential searches. Searching is no longer a problem. Look at how fast Google and Facebook can query billions of records. A car dealer's service management system should be able to use basic (and we're talking very basic) database indexing to make searches fast.
So there's no reason to limit the number of customer records because of search. That brings us to the elephant in the room: the cost of data storage. Let's do some simple math.
Let's do a quick Amazon search on "internal hard drive." At the top of the results list is a Seagate 3TB drive, which can be bought for all of $85.78. That's $25.59 per terabyte, or 2.8 cents per gigabyte.
So how much data do car repair records take? Let's be wildly generous and say there are one hundred populated fields, each containing 256 characters. No, let's get crazy. Let's say each of those fields can contain 32K, like the old PalmPilot memo field's character limit. And let's assume each repair record is fully populated, so all those one hundred fields are full with 32,000-plus bytes of character data.
This would obviously not be true, since most service records are sparsely populated data sets. But let's give it the worst-case scenario. Each record, therefore would be a whopping 3.276 megabytes. Clearly, that wouldn't be the case, but work with me. So, now, let's assume my wife had been in for service twice a year for the past six years. Records of her repairs would -- in our completely over-the-top worst case scenario -- take 39.321 megabytes.
Going back to our eight-five buck 3TB hard drive, how much does 39.321 megabytes cost? Ready for this? One tenth of one penny. Yep, at the very most outlandish of data storage estimates, my local dealer could store ten customers' data, for all those years, for one cent.
In thinking about the situation as it unfolded, another possibility came to mind. Perhaps a legacy computer system broke in some way or was "updated", and records were lost. Things happen, after all. Maybe it would have been truly cost prohibitive to bring the old data into the new system, and it was a strategic business decision to give up on migrating the data.
But if that's the case, the service associates should be instructed to tell the customers that they updated the computer system to a better one in order to improve service, rather than imply that it's the customer's fault for not bringing in the car often enough.
No matter the reason: The service manager couldn't find our data, and the fact that he implied it was our fault for not bringing the car in more recently was probably the biggest problem, especially since it was reflective of his attitude towards his customer.
Think about that concept for a moment. The car didn't need to be brought in for repair for a year, which meant it was working properly. So essentially, we were told we were being punished because their product wasn't as problematic as they might have expected in order to keep the records current.
The customer confidence issue
Let's do some serious math. The lease on my car will be up in a few months, so I'm shopping for its replacement. My wife's car is paid for. My wife has been planning on keeping her car, and the plan is that when my lease ends in about eight weeks, I'll turn in my leased car for a newer model. So that means I'm in the market for a new muscle car.
But I could see it in my wife's eyes. She lost confidence in our dealer's ability to maintain and repair our cars. Since the dealer just lost a lot of cred in her eyes (mine, too), it would be an impossible case to make that buying another car from this dealer would be a good idea. First conclusion: I'm not trading in my car for another one from this dealer.
As it turns out, I did not have all the facts. This was not a storage story at all.
Remember, I've bought three cars from them already. But this 0.01 of a penny data frugality just might cost them the chance of a $40K car sale.
But there's more to the story and this is where the sausage-making happens
The rest of the story
Whenever a columnist is about to write something strongly negative about a company, we try to reach out to the company to give them an opportunity to clarify or comment on the situation. This helps establish a dialog, prevent enraged phone calls from surprised executives, and makes sure we actually have all the facts.
As it turns out, I did not have all the facts. This was not a storage story at all.
I reached out to the car company's communications team for a comment. In particular, I wanted to know if there were any regulatory, compliance, or security reasons why repair records might have been expunged.
As it turns out, the data was not deleted at all. After getting a call from corporate, the local dealer was able to provide the corporate team with actual screenshots showing the repair history of my wife's car.
The problem was that while that data apparently wasn't expunged, that's what we were told by the service rep, who tried to locate the information on our repair history using several search methods. He failed each time. Given that, when pressured, the dealer was able to produce our repair records for their PR folks, I'm left wondering how to reconcile the existence of the data with the service manager's absolute statements claiming the opposite.
I can only chalk this up to human error, either due to insufficient training on the part of the dealer for its employees, or simply a bad day on the part of the service manager himself. Either way, the result was the same -- a loss of customer confidence. After all, not every customer has the opportunity to have their missing data researched by the head office.
Our initial impression of weird data storage miserliness actually turned out to be the result of some sort of human error seasoned with impatience and a bad attitude on the part of the service manager.
The fact that we now know it wasn't IT stuck in the 1990s doesn't change our impression. In fact, it's worse, because rather than an employee trying to be helpful, here's a case of a service manager blaming IT for his inability to use the system. That's either the result of poor training on the part of the dealer in basic system usage or poor training on the part of the dealer in customer service methods - or just an employee with a bad attitude.
Either way, this mistake has proven costly. First, of course, they've just lost the sale of my next car, which is just a few months away. Second, especially after this little chain-yanking-by-corporate episode, my wife says she dreads bringing the car back for service. We can only hope that, as apparently has been the case in the last few years, the next time she needs repair, a whole new crop of service people will be working in the bays.
More likely, though, she'll take her out-of-warranty repair money to some other local repair shops. We may not have a lot of new car dealers in the area, but we have some amazingly cool local mechanics. They'll get her business, not this dealer.
Lessons to learn
There's more than one lesson our business readers can learn from this story. First: Squandering a customer's confidence through the appearance of shortsighted customer service decisions, insufficient training and oversight, combined with a generally cranky attitude can lose you the potential for large future sales.
Another take-away is that blaming IT is not a good strategy. If your customers get the impression you are incompetent in how you manage your systems, you will lose their confidence. It's even more criminal if your company is actually investing in IT - this dealer is due to get iPads in the service bays in a few months - but you're using a made-up limitation as your foil for lack of competence.
Let's be clear here. This particular dealer has a huge institutional advantage in getting our business because (a) we've bought from them before, and (b) they're pretty much the only game in town. So, if they were good, the decision to stay a customer in one way or another would be a no-brainer.
But they're losing our business because they were poorly run when we bought our last car (that's another long story) and they're clearly poorly run now. That's not a good impression to give when your customer is making another car buying decision.
It is disturbing and demoralizing how easy it seems to be for businesses to lose customers' trust over really tiny and stupid things. Don't be like our local dealer. Don't let your employees lose the trust and confidence of your customers, make monumentally stupid mistakes, and then blame it on IT. Think about the big picture and remember that now, more than ever, your customer is able to communicate with your other customers and prospects about his or her level of dissatisfaction.
Making the sausage
You may have noticed that I haven't named the car company or the dealer. We originally were planning to use the car company's name. But the dealer isn't actually owned by the car company. It's an independent franchise.
While the dealer's behavior reflects negatively on the car manufacturer, the manufacturer really has only very limited control over how their product is represented in the field. At least Apple owns its stores, so if you have a bad experience, you know that somewhere, somehow, it's actually the corporate office's responsibility.
When I contacted the car company itself, they rapidly and actively investigated the situation, they were polite and helpful, and frankly didn't do enough wrong to be the subject of a hit piece. Sure, they probably could use more mystery shoppers to keep dealers in line, but after a while, unless you have complete and total control over your dealer network, you have to rely on the kinds of workers you can find in the middle-of-nowhere area in Florida where I live.
Our dealer is another story. There is no excuse for unhelpful and poor attitudes - not to mention the whole "blame IT" thing when a manager couldn't use his own tools. But since the dealer's name shares the name of the parent company, and, frankly, the job market isn't all that great in central Florida, there's no reason (other than our own crankiness) we need to rip these guys a new one, just to share a learning opportunity.
How we present companies to you is a challenging issue, one that most readers don't think about. We are dogged about giving you the most honest, best experience, with the most essential information. We also heartily beat on companies when we're looking at their overall strategy, both to help them find the right path and help you understand the issues. If a company is being predatory and we think you need to be protected from them, we'll disembowel them for your protection right in the first few paragraphs.
But when it's not really the fault of the name-brand company, the dealer is local and you're unlikely to shop there unless you live here in town and have few other choices, and while we want better customer service, we'd rather our neighbors don't lose their jobs, we sometimes decide that it's not necessary to take them down, just for the sake of it. We're here to inform and teach and help, not practice gotcha journalism.
At least not this time. I'm sure there's somebody I'll get to mercilessly mock in the next few weeks. Hey, if nothing else, there's a whole presidential campaign that's fertile ground for an unprecedented level of mocking.
Before I wrap this up, let me share one last thought. I've been noticing a disturbing trend lately in customer service. It seems to me that more and more reps are just saying anything they can think of to move the customer along. It doesn't seem like they're getting proper training, and we, the customers, are subjected to whatever randomness that lack of training or ready access to institutional knowledge results in. Are you noticing this, too? Comment below.