Good data visualization is a balancing act. Communicating facts and statistics in a way that's both pleasing to the eye and conveys meaning intuitively requires skills that are not always easy to find.
It's not surprising when charts and graphs sometimes misfire, especially when the designer tries too hard to be clever and just winds up being confusing.
Just as shipwrecks are sometimes useful ways to spot where the rocks are, really bad data visualizations can help us avoid the mistakes of others.
WTF Visualizations is like a roadmap of how not to communicate data. I highly recommend you spend some time browsing their examples of really terrible data visualization.
First you will laugh. Then you will think. And then, I hope, you will resolve never to venture into those same waters.
I was going through some of our old client data and discovered something very interesting. In every single case where a client has adopted the broad outlines of the kind of survey process we advocate in Agile Customer Feedback, the client has seen substantial long-term gains. For example:
These improvements are changes in full-year survey scores from the first year we started working with each client up to the most recent year we have comparable survey data.
I didn't include all our clients on this list, but we have seen statistically significant (and often remarkable) improvements in survey scores at every single client where we:
It's not often in the business world where you can honestly say you have a solution that works every time. But in our case it's true. Every single time we've implemented an Agile Customer Feedback process for a client, it's delivered significant and sustained improvements over the long term.
In the news today, we learned that Wells Fargo had a small issue with employees defrauding customers by creating phony accounts in order to run up fees and hit sales targets. And by "small" I mean "mind-bogglingly massive."
Over the past several years, 5,300 Wells Fargo employees have been terminated because of these fake accounts. That's roughly 2% of the company's workforce, one in 50 employees.
When you consider that not all Wells Fargo employees have the ability or incentive to set up fake customer accounts in this way, and that probably not everyone who did it got caught, it would appear that this behavior was extremely widespread within Wells Fargo.
It's not hard to speculate on why this probably happened. Like many organizations, Wells Fargo probably set aggressive performance targets for employees with serious consequences for missing a goal. Over time, employees learned that they could bend the rules a little to hit their targets, and then bend them a little more. And over time, the goals probably ratcheted up and got harder and harder to meet without cheating. After a while, committing a little light fraud to meet your numbers seems perfectly normal. After all, everyone else is doing it.
Wells Fargo leadership will probably insist that they did not condone any of this, but widespread employee misconduct does not happen in a vacuum. If the consequence for missing your sales targets is the same as the consequence for creating a fake customer account (you lose your job in both cases), the incentive is clearly there to commit fraud and hope you get away with it.
All of this is a little far afield from my usual topic of writing about customer surveys. But a lot of companies have the same sort of high-stakes incentive systems around their customer surveys as they do for hitting sales targets.
Most people have a fairly strong moral compass, but they can be gradually led astray if they're in an environment where misconduct seems normal and carries few consequences. Cheating on a survey is certainly not as bad as creating fake bank accounts to generate fees from customers.
Your goal in any customer feedback program should be to improve your performance, not just generate high survey scores. But if you treat your survey as a high-stakes performance metric for employees, you are almost guaranteed to get cheating. When that happens, the survey is no longer giving you meaningful information.
It used to be that when you opened up your wallet, you knew whether you were buying a product or a service. New pair of shoes? Product. Haircut? Service. Trip to Disney World? Service. Mickey Mouse T-Shirt? Product.
Knowing what you're getting is important for setting your expectations as a customer. If you're buying a product you don't expect the seller to do much other than deliver it and fix it if it breaks, and you get to continue enjoying the product until it's either used up or worn out.
When you buy a service you expect the seller to do something for you. This could be one time or an ongoing basis, but you generally expect to keep paying as long as the seller is providing the service. Once you stop buying the service (or the seller has fulfilled their obligation) you aren't entitled to keep getting the service.
One of the joys of living in the 21st century is that we now have consumer products that are both products and services, thanks to the Internet and smart devices. That lightbulb you just bought from Home Depot might actually be a "smart" lightbulb which requires ongoing services provided by the manufacturer in order to deliver all the features listed on the box. And if the manufacturer decides to stop providing the services, the lightbulb stops working.
That may sound like a made-up example, but it literally happened this summer. "Connected by TCP" brand lightbulbs, sold by Home Depot among other places, needed to connect to an online service provided by the manufacturer in order to do all the cool things the box said they would do. TCP decided it was no longer worth providing the service, so they shut it down on June 1st, and all those expensive "smart" lightbulbs (as of today, still for sale for $137/pair) instantly became a whole lot dumber.
This is a customer experience nightmare. The company has done a terrible job of setting expectations about what exactly customers are buying. It sits on the shelf at Home Depot just like any other lightbulb, looking and acting like a consumer product, but what customers are actually getting is a product plus an embedded service, and neither the product nor the service is fully functional without the other.
This is also not the first time a "smart" product was crippled or completely disabled when the manufacturer decided to stop providing the hidden service that made it work. Nor will it be the last. As these incidents keep happening, I expect the negative customer experiences to taint the whole market for adding smarts to ordinary consumer products. Most consumers will not be willing to buy "smart" lightbulbs, thermostats, refrigerators, and other devices when they know the manufacturer might not be willing to keep supporting the product for its entire expected lifespan.
To get past this, manufacturers of "smart" products need to make sure that customers are guaranteed a good customer experience even if the company isn't around to provide it. That probably means a combination of setting customer expectations ("Comes with five years of service included!" makes it clear that the service is part of the product), and making better choices about how to deliver the embedded services.
Because the only thing more disappointing than buying a product that breaks, is buying a product that breaks because the manufacturer broke it.
Customer experience failures have many causes. Poor employee training and morale, rigid adherence to policy, broken processes, understaffing, bad design, a culture of indifference, and occasionally--very occasionally--lack of some critical piece of IT infrastructure.
But even when lack of technology isn't the problem, often the first solution a company will reach for is technology.
I think this is just human nature. Internal problems are hard for organizations to solve. Root causes can be buried deep under years of corporate politics and history that nobody wants to unearth. It's much easier to leave the skeletons in the ground and look for the technological quick-fix.
And so the company reaches for the latest state-of-the-art buzzwords and implements Big Data, EFM, Analytics, or (in an earlier era) CRM, ERP, WFM, or some other technology to solve what is fundamentally a problem with execution.
There's no doubt that these technologies bring value and have an important role in any company's infrastructure. But the technology can't solve a problem where the root cause is people and process. Technology is just a tool, and like any tool can be used well or poorly.
For example, if you are delivering poor customer experience because your employees are not empowered to solve customers' problems, implementing Enterprise Feedback Management will not solve that problem. At best, it might make it more obvious that there's an issue with corporate policies but you're still going to have to drain that swamp to fix things.
On the other hand, EFM can be a valuable tool when your organization is ready and able to make better use of customer feedback from top to bottom.
The mistake is in thinking that the tool will, by itself, drive the needed organizational changes. Instead, implementing the technology is something companies often do because it's easier than addressing the real problems.
Technology vendors are happy to encourage this thinking: it's easier to sell a product that the customer thinks will solve all their problems. But the net result is disappointment when the big technology projects fail to deliver the hoped-for results.
We saw this a generation ago, when CRM was the hot new thing. Failed CRM implementations were so common that some pundits went so far as to predict that CRM itself would prove to be just a fad. Of course CRM wasn't a fad: eventually we figured out what CRM is useful for (keeping track of customers), and what CRM couldn't do all by itself (increase sales, make your customers happier and more loyal). Today nobody questions the value of CRM, but we also have much more realistic expectations and nobody begins a CRM project thinking it will fix deep-seated organizational problems.
In the Customer Experience world, we need to keep in mind that our problems are often not solvable by technology. Technology can help, but the root causes are usually leadership, culture, people, and processes.
The good news is that it may be difficult and slow, but the problems are solvable with the right commitment.
And you might need to dig up a few skeletons along the way.
Harry Brignull has spent some time collecting and contemplating "Dark Patterns" in online commerce, the unethical, manipulative, and sometimes outright deceptive things companies do to try to manipulate customers into things they probably didn't want to do.
For example: the form that has a "sign me up for email marketing" checkbox in tiny type at the bottom you need to find and uncheck if you don't actually want to sign up for spam. Brignull has examples that take this to an appalling (and hilarious) extreme.
The full 30-minute video shows the evolution of Dark Patterns and offers some insights into why companies persist in these customer-unfriendly (and sometimes illegal) tactics. Brignull maintains a website with a catalog of Dark Patterns he's collected over the years, and it's well worth browsing.
Coming from a Customer Experience perspective, I think it's especially useful to think about how Dark Patterns come to be, and how they illustrate some of the challenges in trying to design and implement outstanding CX. Dark Patterns almost always involve either misleading the customer or making it hard for the customer to do something (like cancel a subscription). The end result is likely to be an upset customer (or former customer) and very poor CX.
By the time the customer gets upset--when she discovers she's been fooled or trapped--the company already has what it wants. It's too easy for the company to think that CX doesn't matter, because the short-term metrics (sales, newsletter subscriptions, etc.) don't adequately capture the damage that's being done to the customer relationship.
So while you may be able to boost the percentage of customers who buy travel insurance over the short term, those customers aren't going to be happy when they discover the extra charge. They probably won't be fooled a second time, and may take their business elsewhere or warn their friends. Some may call and demand refunds.
In extreme cases, government regulators may get involved.
In most cases, CX professionals who are doing their jobs properly will be working against Dark Patterns. The challenge, as it so often is in the Customer Experience world, is to help the rest of the organization understand that while manipulation, deception, and intentional barriers may sometimes improve short-term metrics, they rarely pay off in the long run.
We just published the 99th issue of our newsletter, Quality Times. This month I discuss what the Margin of Error in a survey means (and more importantly, what it doesn't mean). You can also read about what happened when one company's survey left a very poor brand impression for some of its customers.
This newsletter is one of the ways we get to know prospective clients. So if you find this useful and informative, please help us out by forwarding this to other people who might also enjoy it, and encourage them to subscribe. You can subscribe to this newsletter on our website.
I can think of a couple companies where they probably really do strive for 10% satisfaction.
Most companies can handle the ordinary customer service issues just fine. Often, the difference between a company with terrible service and great service is what happens when things go a little wrong. Do customers get their problems resolved quickly and painlessly, or do they fall through the cracks and get ground up by the machine?
Last week I had the experience of being ground up by the Verizon machine.
I have a prepaid Verizon SIM for my iPad, and I was on vacation in a rustic camp on an island in Northern Minnesota. The iPad was my only reliable Internet service, and when I used up my data allotment I needed to refill my account.
At first I tried to use the "Manage my Verizon Account" feature on the iPad, but that gave me an error message and instructed me to call a toll-free customer service number. I called the number and explained my problem to the customer service rep, who told me that I needed to log on to my Verizon account through a web browser to refill it.
This presented a problem, since the iPad's data allotment was used up and there was no other reliable Internet connection available. I explained this to the Verizon rep, who gave me a verbal shrug and told me there was no other way.
So I sat at the end of the dock, where I could just barely get an Internet connection on my T-Mobile phone, to slowly and painfully navigate the Verizon website. After probably a half hour of this, I was told that I could not reactivate my account online, and I was given a different customer service number to call.
I called that number, explained my problem to the rep, who forwarded me to a different department and another rep, who forwarded me to another rep in another department, who forwarded me to another department. Of course I had to explain my problem all over again with each rep and read off the same IMEI and ICCID numbers each time.
For those playing along at home, the IMEI is 15 digits and the ICCID is 19 digits. You can imagine how much fun it was reading 34 digits to each of four different Verizon reps, all while sitting at the end of a dock with the wind blowing across the microphone of my phone.
The final rep, who was apparently in the same department as the very first person I spoke to, chastised me for calling back after I'd already been told I needed to go online (seriously!). When I explained to her that I had managed to go online and that hadn't worked either, she offered to transfer me into an IVR where I could refill my account through the phone.
After punching in the same 34 digits into the IVR, the prerecorded voice pleasantly informed me that I could not refill my account through the IVR and I would need to call customer service.
It was at that point (after two hours, five reps, and three different self-service channels has yielded nothing) that I gave up on trying to give Verizon any of my money. There is a happy-ish ending, though, in that I managed to find a particular spot on the island where I could leave my T-Mobile phone, activate the hotspot, and get a solid Internet connection in the cabin. T-Mobile was a lot cheaper, too.
I'm not sure what crack I fell through at Verizon that day, and I don't think I want to find out. Whatever the situation, it was clear that none of the normal service channels were able or willing to help me.
Verizon likes to talk a lot about how they have the best coverage, and there's no question that Verizon's coverage on the island was much better than T-Mobile's. But in the end, Verizon's coverage didn't matter because T-Mobile made it so much easier to be their customer.
The lesson from that day is that it doesn't matter how great your core product or service is if your overall customer experience is bad enough.
Let us put our expertise in customer feedback to work for you.