Improving the worst experience

At this moment I’m in the worst seat on the plane.

In 32B, jet engines thunder mere feet from my eardrums, a fact confirmed by a window whose proximity to the engine affords a vista only of three sheets of metal and 37 rivets. The seat-back is jacked into an even-more-upright-than-usual position due to a trash can welded to a storage compartment lodged behind, which I’m constantly reminded of by the clacking of aluminum doors — the only sound sharp enough to pierce the roar of turbines. Seat 31B has no such restriction and so reclines into my lap. The odor of twin toilets waft from behind, competing with pre-event smells emanating from perpetual line which forms at my seat, asses at face-level, overstuffing the aisle with impatient two-way traffic.

I can’t help but think about our equivalent “worst seat” at my company WP Engine — the customer experience that’s so unconscionably bad that they’re justifiably outraged that a company could inflict such injustice on a paying customer, even after assurances that everything would be fixed, after speaking to several people on the phone who gave empty promises of reform and reparations, being led by someone (me) who claims his reputation is on the line and whose rhetoric implies such heinous things could never happen under his watch. A thing so atrocious it’s actually worth even more time and effort to publicize their outrage to everyone they can find.

I know it happens. I’ve seen it at WP Engine. Mostly you get only peeks into how bad it really was, like a post-mortum private email to me personally, riddled with valid complaints and specific injustices that had been inflicted upon the hapless customer, exacerbated by our indifferent silence broken only by callous human contact.

We founders unwittingly focus on how to make the best customer experiences better, rather than making the worst experiences less worse, and sometimes this is a mistake.

When we focus on our perfect customers and make sure they’re completely blown away from start to finish, it’s true that it generates grateful customers who love you through thick and thin, and cheerleaders bearing testimonials. But what happens when you improve the experience on the other end of the spectrum?

For example, instead of letting those ill-fitting customers hang around long enough to have bad experiences, you could figure out what they look like and reject them early and humanely, while also helping them find a better, happier solution.

This happened recently with a WordPress consultant we work with (Bill Erickson). We had a new high-profile customer who required specific expertise, so I connected them to Bill. Bill impressed them and they were ready to begin, but Bill decided this was too far outside his experience and so told them, while it would be interesting and fun for him, and he was confident in his abilities, he isn’t comfortable accepting this job, because he wants no chance that they’ll have a bad experience.

Of course this only won the customer over still more. Bill won’t do this particular gig, but I guarantee that when something else comes up in six months, he’ll automatically be offered the job. As for me, I’m going to continue connecting customers with Bill because there is no seat 32B with Bill.

It can’t always go so swimmingly, and that’s OK. It just has to be “not 32B.”

For example, at WP Engine we had another customer with lots of custom needs. After a few weeks it became apparent that the “technical” person on the customer’s side also didn’t fully understand the scope and implications of the customizations. Still, it was early in our life and we wanted every customer we could get, especially since they were willing to pay a good rate for hosting. So we kept on.

You can guess how this story ends — it was a rollercoaster of problems, some due to the customer’s ignorance, some due to our own mismanagement or misconfiguration, and still more that’s difficult to pin down. But it doesn’t matter who’s responsible for what — the real problem is that we were a bad match and we both wasted time, money, and reputation.

We did learn from the experience — we did a post-mortem and decided what evidence should have stopped us from accepting the account, and now we build that into our sales cycle. When we see those warning signs we can “pull a Bill” and explain to the customer why it’s not a good fit, thereby earning a positive reputation which we’ve already seen spill into new customers because — oddly enough — the folks we send away actually recommend us to others, even though they technically have no operational experience with us. Why? For the same reason that I keep recommending Bill.

Said another way, this combination of honesty and introspection is so rare and appreciated, it’s rewarded by repeat patronage.

Another thing we can do is literally measure customer “happiness” and intercede before it goes from bad to 32B. Hubspot has this down to a science, more so than almost any other company. In this presentation at last year’s Business of Software conference, founder Dharmesh Shah explains his metric CHI (Customer Happiness Index), how it’s measured, how they use it to “save” customers who they predict are about cancel, maintain a deep comprehension of their cancellation rate, and ensure that even the worst Hubspot experience is not a 32B.

Perhaps measuring “happiness” continuously and along multiple dimensions like Hubspot is the ideal, but we all can take simple measures, today, to fix the worst of the worst.

So do it, not just to avoid the occasional disaster, but because bringing up the rear will surely improve the experience for many others, possibly saving some from cancellation, and definitely forcing you to run your businesses with more introspection and even compassion for your dear customers who are not, in fact, just an addend in your revenue column or an entry in your cancellation log.

Even the one you turn away could become a fan.

Let’s continue giving ideas and stories in the comments about saving (or not saving) problematic customers, how you identify them, and when this advice is applicable.