I started my career at Bain & Company, a global management consulting firm charged with helping some of the world's largest organizations solve their most critical business challenges.
A central part of the job was showing up at a client's headquarters with often just a few weeks to not just diagnose an (often very complex) problem, but come up with a solution as well.
This was driven by the fact that not only were the problems being solved often urgent, but Bain's expensive monthly rates meant their solving was even more urgent, at least for those controlling the checkbook.
There were two basic but recurring insights I picked up on doing this again and again for clients over several years:
- First, complex problems are rarely (if ever) solvable without data. Whenever we found a solution, the process almost always involved extracting insights from data.
- Second, data never tells the full story at the aggregate level. Whenever we had aggregate numbers, the first thing to do was almost always to cut the data into its relative slices and then diagnose.
For example, if we were trying to help a client with declining sales, aggregate sales numbers never told you what was happening so we'd cut the data by region and by product. This led to a few different possible scenarios:
- If sales were declining across all regions and products, the cause might be due to a large macro event (a market recession, a change in consumer preferences, etc.)
- If sales were declining in a particular region or product, the cause was likely to be something more nuanced (a disruptive move by a key competitor, a product pricing/messaging mistake, or even personnel issues, such as one region or product having a particularly ineffective team).
The point? It wasn't till we looked at the data and cut that data in the appropriate way that we had a playbook for how to go about diagnosing and solving the problem in question.
SO WHAT DOES THIS HAVE TO DO WITH TURNOVER?
When we talk about the problem of reducing turnover, the same rules I learned at Bain apply:
- First, turnover problems are rarely (if ever) solvable without data.
- Second, your turnover data will never tell the full story at the aggregate level. You have to cut the data into its relevant slices to diagnose.
Most people get this at the high level. After all, you can't even know you have a turnover problem without data since it takes data to calculate what your turnover is to begin with.
Similarly, companies rarely talk about turnover at the aggregate (company-wide) level. You have to break turnover out by position to even know where the problem is serious (#2). Unfortunately, this is where companies far too often stop.
So where should you go next? Cut the data before you diagnose.
My first recommendation is to look at how turnover varies across each employee:
- If all employees turn over roughly after the same period of time (e.g., at the 1-year mark), then you likely a problem related to career-advancement at the company. For example, no viable path to further promotions, not enough investment in learning & development, an incorrect compensation structure, etc.). There's an analogy for this in sales and marketing, where if customers are consistently dropping from the funnel at a certain stage of the funnel, you likely have a problem with that specific stage.
- If employees turn over after highly variable periods of time (e.g., some in less than 30 days, others in 30-90 days, others in 90-180 days, and so on), then you likely have a hiring problem. There's an analogy for this in sales and marketing, where if many customers make it, but even more do not, you likely have a problem related to how you're targeting customers. Target more customers like those that make it. Learn to better screen up front for those that do not.
Of course, you can have multiple sub-problems going on at once, but the point is data can help you identify where to focus most in order to have the largest impact in your turnover reduction efforts.
YOUR MOST EXPENSIVE TURNOVER IS ALMOST ALWAYS DUE TO HIRING MISTAKES
It's important to note that not all employee turnover is created equal. What I mean by that is that for a fixed position (e.g., customer support rep, etc.), it's the employees that turnover in the first few months that are costing you the most.
Why? Because you basically just spent an enormous amount of money to find a candidate, screen that candidate, onboard/train that candidate, only for him/her to leave before they generate any meaningful ROI for the company. And as one last gift, you now have to go spend more money to find a new candidate (only for the process to repeat itself).
IMPROVING HIRING IS CRUCIAL WHEN REDUCING TURNOVER
It's true, people leave jobs for a variety of reasons, but if you have employees leaving jobs just a few months (if not weeks) in, there was probably something you missed when hiring them in the first place.
For example, either you missed a key trait(s) they have or didn't have or you didn't correctly explain what the job was going to be like in the first place. Both could represent potential opportunities for improvement in your hiring process.
Because of that, if you want to get rid of your most expensive turnover first, take a look at how you're screening candidates. That's usually the best place to start.
Enjoy this article? This is a small portion of our hiring guide "The Ultimate Hiring Guide for Reducing Turnover."