twenty-two seconds to live

I briefly talked with an MBA the other day who was writing his doctoral thesis – or whatever you call it – about his idea of “Service Recovery.”  He posits that business entities that screw up and then do well fixing things are ultimately more successful that those that never screw up in the first place.  It’s all about expectation management.

Imagine you’re at a restaurant you’ve never been to before.  You order a steak, medium.  The waiter brings you a medium steak.  You eat it, pay the bill, and leave.  Unless that steak was phenomenal, you just ate a steak at a restaurant and that’s the end of that.

Imagine, now, you’re at another restaurant.  You order a steak, medium.  The waiter brings you a well-done steak.  You tell the waiter your steak is overcooked, and to please replace it.  The waiter goes into damage-control mode and rushes the offending meat back to the kitchen.  Within seconds, the house manager is at your table apologizing.  Maybe he gives you a bottle of wine on the house, maybe he comps your steak, something.  A new steak is brought to you.  This time it is cooked as you ordered it, and the chef is delivering it personally.  He apologizes as well.  He’s got an off-menu amuse-bouche plate for you to try, too.  It’s all to assuage you, to make you feel special.   Unless you are made of stone, it probably works.  You eat, pay the bill, and leave.

What happens next, however, is different.  You have a story.  It has conflict and drama, rising and falling action.  You tell your friends the saga of this restaurant and how they so effortlessly made things right, and how pleasant it all ended up being.  You would definitely eat there again, and encourage your acquaintances to check it out.

Does it matter if the steak was identical in quality to the one at the place that got it right the first time?

Working in network security, I can definitely say there is truth to this.  My most loyal customers are ones who have been the victim of some sort of malware infection, data breach, intrusion, or other nightmare.  In those scenarios, my products and services have – in a technical sense – failed.  Where I succeed, however, is in the fact that the way my team handles these incidents is off-the-charts five-star excellence.  A company paying for security that never has an incident isn’t thinking about their security.  All anti-virus vendors are equal if you never encounter anything bad.  When you slip – and everyone slips eventually – the eyes are on you.  If you wake up a bunch of engineers at three in the morning, do forensics and analysis, and hand-walk malware samples through the virus lab so you can get an engine update pushed out in less than an hour, people remember that.  They know that even though their security provider slipped – everything was taken care of beyond satisfaction.  They know that they don’t need to fret about the next time, because they know how it will go down.  Even feeling the sting of the incident, switching to a new vendor seems like madness.  They’ll be an unknown, and when they slip you’ll judge their response against what just happened.  It’s all too risky, and network security is all about risk – avoidance, reduction, and management.

Montgomery Scott said to always multiply your repair time estimates by 2.  Say it’s going to take two hours, then do it in one.

Underpromise, Overdeliver.

There are lessons, and then there are unintended consequences.

Let’s go back to the restaurant.

Your close friend goes in and orders a steak.  It comes out exactly as ordered.  Near the end of the meal, your friend calls the waiter over and is visibly agitated.

“I heard you give out free appetizers here, and wine.  I’ve been told the service here really goes to extreme lengths to make sure everyone is happy – and while the meal is certainly fine, I really don’t feel like anyone’s gone out of their way to take care of me.”

What should the waiter say?  At this point, he’s trapped.  What can he do?  Encourage and exacerbate the situation by folding to this customer, perhaps.  He could get offended, he could push back some.  He’s in an untenable situation.  The manager and chef are complicit, too; calling them out will be no panacea.  Everyone showed up and did their job the best that they could, and someone is disappointed.

How do you recover from that?

Let’s move this narrative along.

Your friend tells you about how wrong you were, about how poorly he was treated at this restaurant you so highly recommended.  Neither of you are likely to go back, because one unmitigated failure – however perceptual – overcomes any number of soaring successes.  The story of the cock-sucking bridge builder comes to mind.  What went wrong?

You went in with moderate expectations of the restaurant, and those expectations were likely quite similar to what the restaurant staff expected of themselves.  They actually failed, but then in recovery the raised the bar.  Your anecdote gave your friend a very high bar, a level of expectation that the restaurant was not prepared for.  Everyone was doomed to disappointment from the beginning.

Objectively, the restaurant has made three steaks.  Two of them were correct.  66.7% accuracy.  That doesn’t sound too great, but it’s all about perception.  In your eyes, you don’t see the first one.  They’re 1 for 1, 100%.  For your friend, 0 for 1 – 0%.  The big killer here is the person who seems them as having failed isn’t the person that got served the bad steak at all!

How can an organization – or even an individual – manage expectations at any kind of scale?

Thankfully, and damningly, scale is irrelevant.  At the smallest quantum of interaction – the personal relationship – expectations are constantly recalculated based on a decaying average algorithm that is inscrutably unique to every person.

How many of you have had a teacher or parent tell you, with a look of sadness, that they had expected much more of you?  It’s the “you” in that statement that burns the most.  Other kids may have performed even worse than you – but this is about expectations.

When you disappoint someone important to you, you make a note of it.  If they’re important enough, you intend to give extra effort to avoid that particular failure in the future.  When a relationship is an incalculable series of triumphs and failures, you cannot know where the bar is at any point in time.  When confronted with the awkwardness of a misstep, if communication fails, it will either be a race to the top or a race to the bottom.  If both people are simultaneously disappointed in each other, which do you think it’s going to be?

It’s the answer to that question that I believe is the measure of us, and have yet to be happy with the answer.