The Million-Dollar Cobra in Your Comp Plan

“It is difficult to get a man to understand something, when his salary depends upon his not understanding it!” – Upton Sinclair

In imperial India, the British Raj had a problem.

I mean, presumably they had a great many problems. But among those problems, one stood out:

Cobras.

Lots, and lots, of Cobras.

This is not a recipe for a good time. It’s a recipe for a lot of dead Imperialists. Which depending on how you feel about colonialism may not be a bad thing but stick with me here. Regardless of how you feel about it, you can guess how the Raj felt. No bueno.

So they did what any enterprising, thoughtful leadership group would do: They paid the locals for dead cobras.

Makes perfect sense, right? Incentivize the locals to kill cobras, they make some bank, you stop dying because there are no more cobras. Easy peasy lemon squeezy, right?

Of course it’s never easy peasy lemon squeezy. It’s usually difficult difficult lemon difficult. (Or maybe kumquat difficult, I dunno.)

The ruse really did work…very well. Too well. In fact, our creative Indian population in this story realized that there’s something that’s way way easier than catching a bunch of cobras: breeding them in a shed. Before long, the region of Delhi was littered with actual cobra farms, which the Indian population would then kill and bring to the ruling Raj for a clean bounty.

Yeah that’s right all your cool “growth hacker” ideas? They aren’t original. The Indians beat you to it. I tend to imagine in some local newspaper some guy dressed in a suit jacket with question marks all over it with the headline “Get rich quick, I’ll tell you how!!!” as he holds up a brace of dead cobras.

This, of course, is not what the Raj *actually* wanted, so they killed the bounty program. You can guess what happened next.

…Yep. The locals released all their cobras into the wild. The Raj ended up with more cobras on their hand then they had before the program even started.

The Cobra Effect is all over your businesses incentives. So many things seem like a good idea at the time, to use the famous last words…but as with many things, the devil is in the details with incentives, and as with many things involving behavioral psychology – which an incentive, with its goal of behavioral modification, falls cleanly inside – the results are often counter-intuitive until you really get down to how people truly think. Here’s a few common ones.

Local Optimization (or: This Band Sucks.)

Imagine you pulled together a jazz combo. Or rock band, or sad girl band, whatever your little heart desires. (As long as it’s not bro country. Seriously.) Anyway, imagine you had great players…really great players. And you expected them to make great music. But in order to do so, you incentivized each of them individually. You tell the drummer he’ll get paid a set amount every time he strikes a drum. The bass player will make money every time he plays an E-flat because you really like the sound of E-flat for some reason. (I’m more of an A-flat guy myself but y’know.) The guitarist you give a big paycheck but say he’ll lose money for every string he breaks. And the pianist you give money for every 10 seconds of silence to “let the music breathe”…what kind of music will this create? A drummer who’s about to pass out he’s playing so fast, a bass stuck on one note, and a guitarist and pianist who are sitting on their hands. It’s awful.

And yet, this is exactly how we incentivize most employees.

It’s not because leaders are stupid. It can be very difficult to incentivize units. The local metrics for each department or discipline are easy to measure and reward. And we tend to default to what is easier to measure.

Let’s say you tell marketing that you need more customers. Feed the pipeline, find leads, gimme dem sweet sweet MQLs. They’ll get a bonus if they can fill the pipeline. What happens? Marketing creates a facebook giving away a free iPad if people fill out a form. Two weeks later your VP of sales is in your office with steam shooting out her ears because her team is calling all the new leads and finding they’re a bunch of 19 year olds who wanted some free swag.

Marketing hit their number. Got the bonus. And somewhere across the building, your sales team spent the next two weeks manually disqualifying 400 leads of people who don’t know what your company does and would like their free iPad now please. Marketing did exactly what you paid them to do. Sales did exactly what the situation required. And the business got measurably worse because of it.

Everyone played their instrument perfectly. Nobody made music.

Goodhart’s Law (Or: You Broke Your Own Compass.)

In 1970s Soviet Russia, the state decided to measure factory productivity by the number of nails produced. Simple. Countable. Very good. More nails equals more productivity equals more of that sweet, sweet communist triumph. Trotsky would be proud.

Worked great, right? Well yes, they got exactly what they incentivized.

The factories proceeded to make nails the size of paperclips. Millions of them. Absolutely useless. Couldn’t hang a picture with them, let alone build a nation that would crush the capitalist pigs under your feet. But the number? The number was incredible. Who doesn’t love presenting a graph that goes up and to the right?

But obviously not really what they *wanted*. So the party changed the metric. Now they’d measure by weight.

….You already know where this is going.

The factories immediately pivoted to nails roughly the size of fence posts. Maybe useful if you’re one of Gulliver’s Brobdingnagians, but otherwise completely worthless.

This is Goodhart’s Law: the moment you make a number a target, it stops telling you what it used to tell you. The metric stays “healthy”. The thing underneath it doesn’t. You’ve essentially handed your team a map and told them the map is the destination. They will optimize the map. (They will optimize the map so hard.)

I once took over management of a sales team. That sales team was commissioned on top-line and was given the ability to grant up to a 20% discount to close a deal without approval.

Our margins on some products were as low as 25%.

Do the math…yep. They were giving away 80% of the profit and sleeping like babies afterward.

You’ve seen this in your own building. The sales team whose CRM hygiene is spotless and whose last four deals closed at 35% off. The marketing report showing record engagement while revenue is flatter than Topeka, Kansas. All perfectly healthy nails. Useless for building anything.

One of the most common ones: ticket time for customer success. Your C/S team is closing Zendesk tickets at record rates…and you have a horde of angry customers with torches and pitchforks camping outside your office because they’re all even more pissed than before they talked to your team.

Make a metric a goal, and you’ll get useless nails every time.

The Closer Problem (or: I Thought We Were Friends…)

Think about the last time you bought a car.

…no, it’s okay, I’ll give you a few minutes to work out your anger. May I suggest punching a pillow? I know…I know. (I suggest Aunt Maybel’s green tartan throw pillows, they absorb a punch really well.)

Okay, you’re back. Good. Remember that sales guy? No, I mean the *good* one. Every lot has one.

From the moment you walked onto the lot, the experience was borderline magical. The salesman remembered your kids’ names. He knew your needs. He let you drive *really* fast on the test drive. He found you the exact color you wanted and somehow got the price to a place that felt almost reasonable and made you feel like you were getting away with something. You drove off feeling like you’d made a friend for life.

You know when you heard from him again?

Never.

(Okay, maybe a birthday card the first year if you’re lucky and he’s unusually organized.)

That’s not because he’s a bad person. It’s because his compensation ended the moment your taillights left the lot. His incentive was to get you to sign on the dotted line. Not to make sure the car was right for you, not to make sure you’ve figured out how to use CarPlay, not to see if your kid has already ground animal crackers into the backseat. Everything after was someone else’s problem.

Now ask yourself how many people on your payroll have this exact arrangement.

The sales rep who gets 100% at contract signing, not renewal. The account manager who makes the same whether the client is thriving or churning. The business development team whose metric ends at “deal closed” and whose relationship with the client ends approximately eleven minutes later when the commission posts.

You’ve handed them a finish line. They crossed it. Everyone clapped. They got coffee, because coffee is for closers.

And now your customer is wondering why nobody calls anymore and your competitor just bought them lunch.

Nassim Taleb calls this skin in the game. If the person making the recommendation doesn’t share in the consequences — if they get paid regardless of what happens next — you don’t have an incentive structure. You have a liability with a bonus attached to it.

***
Here’s the only question that actually matters after reading all of this: What does your comp plan actually incentivize? What does your weekly dashboard incentivize? What does your staff meeting incentivize? What does your attitude toward your team incentivize?

Not what you designed it to incentivize. Not what you told your team it incentivizes. What it actually, demonstrably, produces in the real world.

That answer is usually where the cobras are.