Truth and Consequences
It’s an odd place in time for me.
On the one hand, Healthcare.gov just went belly up. I’m not sure what else to call it. According to Reuters, The site has 50,000 sign-ups in October, about ten percent of the five hundred thousand predicted by the Administration. The current rhetoric is to get the bugs fixed by the end of November, and then things will be better.
Still, what happened? Robert C. Martin offers his thoughts and eviscerates the entire software industry. Martin claims it was we, the technology people, who let the world down. Oh, he blames management and government too, sure, but Uncle Bob doesn’t let us off the hook.
After all, someone on the project knew, right?
I can’t say I disagree too much; I just wrote an asessment for a major outlet, real journalism, that came to the similar conclusions. The public comment was that the site needed “months” of testing that was shortened to “weeks”, but those of us in testing know that “months” was really “one week of testing and weeks of fixing.”
Someone, probably many people, working on that project knew the site wouldn’t go up, but they were overruled. They probably had memos, emails, a paper trial a mile long -- they did their job, dag-nab-it, but some executive at some level decided the risk was nominal.
When you think about the system forces behind a multi-level message, even that makes sense. The tester says “can’t ship - doesn’t work!”, the supervisor, wanting to take responsibility, says “major problems we are working on”, the manager says “we’ve got problems but we are handling them”, on up the chain, until an subordinate tells the Secretary of Health and Human Services that “we’ve got it handled.”
In order to fix it, we need to jump the chain of command, which we call whistle-blowing. Is a website that allows people to sign up for insurance really worth blowing a whistle over, when those same people could google for “health insurance (state_name)” and call an 800-number instead?
But there’s a bigger problem
At one point in his upcoming book, Joy Inc., Richard Sheridan tells the story of speaking at a Project Management Conference. With every head bowed and every eye closed, he asked that conference to raise their hands if they had ever had to fake test data on a project.
Every hand went up.
Every. Single. One.
Notice the language I used above “had to”, as if the project manager who lies about his project status is the victim of some sort of intense social pressure.
Most of the time the ‘social pressure’ is the boss looking at you funny, a bad email, perhaps a mediocre annual review. Is lying about test data really worth avoiding these things?
I mean, say you get the bad review which, by the way might not happen. Say it is really bad, a zero percent raise. You’ve just learned a whole lot of valuable information about the company -- the company values being certain and wrong over being uncertain and right. You are then free to decide if you can live on small raises, or to consider selling out, or to look for another job. If you run a test and have the information, you get to make that decision of your free will, not as a result of following the path of least resistance.
But it is possible the bad review does not happen.
Every time I hear someone say “hey, I have to feed my family”, I start asking what they have done to test the theory that without lying, they would be unable to feed the family.
Invariably it is nothing. The person is acting out of fear, uncertainty, and doubt.
In The Final Analysis
There are a thousand articles (including my own) about all the things that have gone wrong at Healthcare.gov and how to fix it. Today I wanted to hit on one -- our willingness, as an industry, to lie and blame it on social pressure.
You see, five years ago, I was a project manager at a health insurance company. And I was awkward, and imperfect, and failed in a thousand ways ...
... but if I had been at that conference, and Rich asked his question, I would not have had to raise my hand.
I didn’t need to lie to feed my family and I don’t think you do either.
Now, what do we do to help the rest of the world figure it out?