Data and Deniability

A few months ago, I was getting ready for a talk on legal issues in HR technology for a group of employment lawyers when my co-presenter mentioned something about ‘maintaining the cloak of deniability’.

The cloak of deniability is the idea that sometimes it’s better not to know. For a long time, employment lawyers have told organizations not to look for trouble. If you can avoid knowing someone is in a protected class, then you can’t intentionally discriminate. If you have no idea that nobody over 40 has been hired in the San Francisco office in two years, you also don’t have to deal with it. Well, until the age discrimination claims show up.

I was getting ready for a talk on legal issues in HR technology for a group of employment lawyers when my co-presenter mentioned something about ‘maintaining the cloak of deniability’.The problem is that in most employment law claims, not knowing isn’t really a defense. The legal standard is usually what the employer knew or should have known. Employers are liable for the things they reasonably could have discovered even when they did not actually know. They are also generally considered to ‘know’ anything their managers and executives knew at the time, even if those people didn’t tell anyone else.

With exponential growth in data available about employees and the workplace, and increasingly sophisticated technology and people analytics, we can’t pretend not to know things anymore. The cloak of deniability is no longer cover.

Instead, organizations need to understand what data they have and what it means. We need to watch for changes and wonder when things are not changing. We need to look at different factors in combination and understand the stories they tell and the questions they raise.

Then, we need to investigate to find out what is really happening. Even if, especially if, there may be trouble. You can think of it as a wellness tracker for organizations. And just like other health issues, they are much easier to treat before they become severe.

If you would like to know more, come hear my Tech Talk on Trouble Metrics, Wednesday at 1:20 at the Expo Stage. I’m also talking about Bias in AI with my friend and colleague Kate Bischoff on Friday at 9:30 in Delfino 4102.