Do you know what your company’s turnover rate is? Should you? Does it really matter?
Over the years we have become slaves to the magical annual turnover figure. Many hours have been spent by HR conducting exit interviews and analysing exit data and trends and reporting it back to a largely uncaring business.
Too high, and turnover is HR’s problem to sort out. Too low, and this is supposedly a problem of a different kind.
In recent years as turnover has crept up across most industries, HR have invented another way of measuring turnover to make it more palatable. We call it regretted attrition. Or put another way, how many have we lost that we really didn’t want to lose i.e. a rock star?
That’s a more valid measure perhaps but is it that relevant?
And how do we feel knowing that most people who leave our organisations are not regretted? Have they made such little impact?
The world of work has changed and continues to change. In New Zealand it is becoming the accepted norm that if you want a big salary increase or significant new challenge you have to move jobs.
The reality for all of us in HR is that our precious employees are more mobile, connected and actively sourced by others than ever before. They are also more discerning about who they work for, how they work, where they work etc.
The other reality is that most organisations over a certain size will have turnover somewhere between 15-25 per cent every year.
Even Google, whose recruitment methods, working conditions and benefits are supposedly second to none, have an incredibly high turnover rate and median staff tenure of around just 12 months.
So perhaps we should just accept that high turnover is something we can’t do much about?
I once used to work for an executive who told me that he didn’t mind people leaving, so long as they left for the right reasons.
And I think that’s probably the best measure. Focus on those who are in the organisation not on the way out. Challenge your managers to know their people, their aspirations and therefore their flight risks.
Make it as easy as possible for people to succeed in their role, make sure they are continually challenged and learning and provide the working conditions that make it enjoyable to come to work every day. And if they are good, reward them well and show them the next step they can take with you.
Now doesn’t that sound more sensible than talking to them when they’ve already opted to take their talents elsewhere?
Up the Down Escalator 
I do wish that some of engineers whose English is so “challenging” to understand in conference calls would chose to leave. However, they tend to stay, especially since we seem to have lost their passports somehow.
Turnover is global and sustainable, and keeps costs low. Mais oui.
http://ramsbottom-lemieux.blogspot.co.il/2013/12/new-retention-tool-gangman-style.html
Agreed Richard. I always wonder why so much emphasis was placed on this statistic when an organisation was achieving its objectives and creating cultures that appealed to most.
At one point I was an ‘Employee Retention Specialist’ (I’ve had some very special job titles…). We had a meeting every single week about leavers and did a ‘cradle to grave’ review of every aspect of their hiring, management and decision to leave. The issue is that capturing what goes wrong uniquely for an individual usually masks a litany of broader and bigger problems.
After a number of years of evolving our approach we eventually reached a more enlightened position – don’t worry about the individuals, worry about creating an environment/culture/place of work that did the right things to help people want to stay. The big challenge are normally the ones in front of you, not those behind you – ‘what can we do to keep this patient alive?’ is a bigger challenge than ‘why did that other guy die?’.
Oh – and whilst I’m waffling – attrition benchmarking is normally guff. If you benchmark higher than you think should then you say that the benchmark isn’t a good fit as your company is unique. If you benchmark and think you are doing well then you lose momentum. Deal with what is in front of you and trust your instincts where they say you could be doing better. Even better trust what the people you are concerned about tell you.
Just stumbled across Michael Carty’s recent article on big data and that we should be predicting how many will be leaving not reporting historical stuff. But even that seems incredibly hard to do!
I missed that one, but we did spend plenty of time discussing potential leavers too. The issue is that you then invest time swinging their marginal decisions rather than focusing on the big picture. It’s a tough one to get right. I think the best solution is to focus on getting the big picture right and then test your assumptions with some live human beings and a smattering of data (if you are of a size that needs it). You struggle to predict the future without at least learning some lessons from the past e.g. it was only digging into our data that showed us that the most likely reason for someone to leave was that they were only using us a stopgap before University (hence a peak every August…)
When I worked in financial services I did look at how many people were leaving for specific reasons (e.g. didn’t get on with manager, leaving to go OS, more money, lifestyle change etc) but what we didn’t do is look at which categories we wanted to reduce or whether if we improved culture/managers that category reduced.
Good stuff Mr Westney!
Really good post, Richard, and a very interesting perspective on the topic of HR data usage in general, and of data relating to turnover in particular.
I’ve added a link to this post in the comments field of my Personnel Today article on HR and big data, which you kindly referenced in your comment above: http://www.personneltoday.com/hr/big-data-hr-needs-stop-reporting-start-predicting/
I know it’s getting a little bit away from the main focus of your post, but I was wondering, do you think there are any other “magical” metrics to which HR might be at risk of becoming enslaved?
Thanks Michael. I do think we measure recruitment stats wrongly in most organizations. Things like time to hire and cost of hire are not really relevant any more. We should be looking at quality of hires rather than quantity which is probably a good rule for most HR metrics.
That’s an extremely interesting response, Richard. Your suggestion that a focus on quality rather than (or perhaps in addition to?) quantity in HR metrics is particularly intriguing.
If it’s not too complex a topic to get into here, what sort of factors do you think could be used to measure quality of hires in a meaningful way? Off the top of my head, some potential factors to incorporate into a quality of hire measure could include rates of attrition for new recruits during their first year in post (though I suppose this would be getting back to the turnover metrics discussed in your original post, above), new recruits’ achievements against objectives, their individual contribution to the success or profitability of their team/business unit/organisation (delete as applicable)… but I’m sure there are many more likely candidates for inclusion here!
Some good ideas Michael. I think the issue with the basic metrics I mentioned previously are that they only measure the efficiency of your process. I confess I have been guilty of that in the past. If you are doing high volume, low engagement recruitment that’s fine. For the rest of us it should be more about the quality of your hires. As you say, longevity, performance, culture fit, value add, long term potential. What about the return on investment with your branding initiatives? Or the quality of your relationship with external recruiters? If we are talking big data, what about where your most successful places/methods are to attract the right candidates, brand perception, all your psychometric assessments, what are the skills and personalities that succeed best in each area of your organisation. At the end of the day, it doesn’t matter if you get five responses to a vacancy or 500. You just need to find the right person.
Great comment there, Richard… were there a “like” button on here, I would definitely “like” it.
I like the big data approach to producing quality of hire metrics there. Do you know of any HR departments or professionals out there who are already actively producing such data?
(Hopefully that’s not one of those “questions for which the answer is Zappos”!).
To be honest Michael, no. I am sure they are out there though!
Interesting read – thank you for the refreshing perspective! While it can be argued some attrition as inevitable (it’s described as “good” and “bad” attrition in this article by Pasha Roberts: (http://www.kdnuggets.com/2014/05/employee-churn-202-good-bad-churn.html). However, all hope isn’t lost. As you said, the focus should be not only on employees you have now but future employees the business needs to have that will perform in a specific role.
To get to this point, you need to predict top performers on the way in and that way you can get to the retention numbers you’d like.
Thanks Mike. I’m with you on that!