The Hippo is a much maligned creature. Forever scorned by data scientists and UX professionals.
If a decision has to be made, the boss’ opinion wins. After all, there must be some reason they get paid more!
If we’re deciding between a red button and a blue button for our Call-To-Action, don’t test, just ask Hi.P.P.O! There’s even a SaaS web site dedicated to asking Hi.P.P.Os.
Personally, I don’t mind the Hi.P.P.O. Sometimes I have to bow before the Hi.P.P.O. Sometimes am the Hi.P.P.O.
What I truly fear is the Z.E.B.R.A (More on that below.)
While the Hi.P.P.O is often rightly derided as a terrible decision making process, there are some situations where it’s actually a good thing. So here are 3 reasons why we might want to listen to the Hi.P.P.O.
1) Testing is Expensive
For Google, it makes a huge amount of sense to test 41 shades of blue. A fraction of a fraction of a 1% increase in conversions can means millions of dollars to google. For the typical startup? Not so much.
Google also has millions of page views and can detect statistical significance out of such small changes. Most companies don’t have the luxury of large sample sizes and a decent test might take weeks or months to gather sufficient data.
If the cost to run an A/B or other usability testing methods is equal to or larger than the expected value of a successful test, then it’s certainly not worth it. Just do what the Hippo says and call it a day.Hint: If you don’t know what the cost to run a test is, you’re not doing it right.
2) Choose Your Battles Wisely
Also, remember that there is a personal cost to endlessly challenge everyone’s suggestions and testing every little thing: People might hate us and we might lose the will to fight.
While we should never senselessly bow to authority, sometimes it’s worth it to just add that stupid link to the CEO’s pet project in the footer.
Will it impact the site performance? Probably not.
Will it make the CEO happy? Yes.
Would insisting on testing it just irritate the hell out of the CEO, the engineer who has to implement the A/B test, and the data analyst who has to do the custom SQL query to crunch the data? Yes.
Is all the increased aggravation and declining team morale going to make it harder to get buy in for running the next test? You betcha.
It takes courage to stand up, challenge authority, and insist on what is right.
3) Authority Ends Debate
There is a third cost (Sensing the theme?) to running tests. The cost of debate.The “product” of lean startup is knowledge, so any company activity that doesn’t produce knowledge is wasteful.
Debating does not produce knowledge. It is a wasteful activity.
Analyzing results is not debating. That’s a useful activity. But debating opinions? Pure waste.
This is where the Z.E.B.R.A comes in.
Fear the Z.E.B.R.AZ.E.B.R.A = Zero Evidence But Really Arrogant
(Credit for this brilliant acronym goes to Emily Chiu.)
The Zebra is often someone with “expertise” who “really knows the customer” but doesn’t have any facts to back them up. Sometimes this is the PM or the CEO, but more often it’s the UX person. (Sorry UX folks!)
Based on years of design experience, we just know that the button should be red, the image should be smaller, and the shading on the button should be 4 px, and, god forbid, not 3 px.
This is far more dangerous than the Hi.P.P.O because while the Hi.P.P.O can be swayed by facts, the Z.E.B.R.A. says, “The customer is wrong.”
The Hi.P.P.O is a default decision. In the absence of facts, do what the boss says.
The Z.E.B.R.A rules through the tyranny of expertise.
For all the Hi.P.P.Os out there: Try to live by this mantra, “I’ve made a decision. If you’d like to change the decision, bring me facts. Define an experiment that will prove me wrong, and I’ll back it, but the debate is over.”
For the Z.E.B.R.As: Repeat after me, “All opinions are equal in the absence of data.”
- Know the cost of running a test.
- Choose your battles.
- Debating opinions is wasteful.
- All opinions are equal in the absence of data.
So…what should I post next? Tweet to tell me what to write:Show me how to test product market fit!
orHow can I do lean startup in my friggin' huge company?