Why We Retired Attrition as a HR Metric
Illustrations by the creative DY LEE
Attrition last proved to be a problematic measure for us when we sat down for a strategic meeting and we began questioning why we were measuring our turnover rate. As the numbers were presented, the quick protests began to show how ineffective it was as a measure for our organization.
“This person wouldn’t have grown with us, she needed a bigger pond to swim in. That’s why we didn’t even give him a counteroffer.”
“What if their career has over time gone out of alignment with the company’s goals, yet they’re still performing? We wouldn’t terminate on that basis so if they leave, it’s a win-win, right?”
“Hasn’t turnover mostly proven to be healthy for us? It’s given us opportunities to reinvent that we wouldn’t have had otherwise.”
Attrition was turning out to be the easy way out. In our pursuit of a data-driven HR method, we had neglected to see that attrition was doing our organization and our people no favours with its one size fits all numbers.
Quick detour: Context
For context, we first need to put up what kind of organization we are, how we hire and what compensation is to us.
Over the last 11 years, we have found that the people who made the most impact on our organization have been people who have taken the road less travelled. Some of them are law graduates who didn’t end up doing law. Others studied philosophy and became project manager extraordinaire. As a result, we’ve moulded our hiring engine to accommodate these exceptions. In a nutshell, I’d say our approach to hiring would be one of hiring for potential. As long as they show promise in terms of skills we value such as critical or strategic thinking, influence and craftsmanship, depending on the role, we make the hire. We cast a vision together with the hire as to what he/she wants to achieve in their career (most times, in totality) in alignment with their life goals over the next couple of years, and then we strive to achieve that together.
While hiring these team members, whom we’ll call Originals, is easy, keeping them in the organization is not. If you ask any of them, they will have stories of how other organizations couldn’t compensate them fairly. Some of them don’t have degrees but can think critically and strategically. Others have accomplished in five years what we thought they would in seven. Most notably, the significant group in our workforce that started off as interns and are expected to perform like a full-time hire from Day 1 — we know compensation can be tricky especially with hires that don’t fit in the prescribed grades or banding. Yet as an organization we chose to create enough flexibility (and headaches for the HR team) within our organizational structure to not only allow them to flourish, but to make sure we get out of their way and not stunt their growth — our own way of taking care of these people, too.
What are you optimizing for?
If you are optimizing not for how long people stay with you, but their performance while they’re with you, then attrition is not fit for your purpose.
The fault in attrition
We notice that in our practice, attrition was commonly used “for show”, to illustrate that HR is effective in reducing costs. What we’re aiming for was, however, not cost reduction but maximization of spend. Attrition fails to capture the efficacy of the initiatives we had put in place to maximize every cent we’ve invested into our greatest asset as a consulting company.
Attrition happens to be one of the best metrics for HR if you reward based on tenure — that happens to be one of the most ineffective measures for us. We haven’t kept track of our company’s average age for a while now, but it’s still not gone above 30 yet (edit: It’s currently 29.5). When you have a company that young, years of experience matters next to nothing, and quality of experience takes a bigger role. This team delivers the high quality consulting on technology, business and design that has gotten us where we are, and they do it with flair and consistency. Tenure with the company matters less when you have the right processes for retention of top talent, continuity and knowledge transfer.
Tenure is also ineffective in measuring the impact of individuals who have grown and exceeded their peers. In say, a 5 year tenure with the company, they might be taking on roles that would have otherwise required 10 years of experience. This is something we value and as a result, tenure is close to irrelevant for us (other than being able to tell stories of old offices and colleagues!).
The only area in which we use tenure as a measure is in rewarding loyalty. If individuals are still growing and see value in staying with the company, we want to be able to reward that, so tenure comes into play in the compensation/bonus framework, not in measuring retention.
Note that we find that attrition taken on its own with no specifications is a problem, but used in a more specific and granular manner, can be very useful — more on that later.
Where we are with HR metrics now
Over time, we’ve replaced attrition with other HR metrics that are more suited to purpose — specifically, employee satisfaction and top talent turnover.
With tools like Glassdoor and net promoter score, we use employee satisfaction to measure the health of our culture. Employee satisfaction only works as a measure for us because we assume that the hires we make are, in majority, good culture fits (or we would have to take a long, hard look at our hiring practices!) Therefore, they are the best evaluators of our culture during their stint with us based on their growth, their experience of the team dynamics and how strongly we hold to the values we’ve defined. As most of our hires rank culture/growth/recognition to be within the top 3 criteria for joining us, this is one area in which our qualitative interactions have given us much needed feedback to change and grow as a company, and we have found that this is a reliable predictor for the company’s performance as well.
Another metric we use is top talent turnover. We use this because when we hire for potential, that often means we make hires that have taken the path less travelled. As a result, our metrics also have to factor in that their career journeys with the company tend to look different than the run of the mill. For us, top talent turnover is measured more granularly than attrition, where we segment turnover for specific buckets of talent ie. within the chatbot engineers or within the platform leads, and supplement it with qualifications as to why these numbers have presented as such. This gives us more meaningful statistics as to problems that could be affecting specific job groups or specific seniority segments. The qualifications also give us perspective when it comes to making decisions — though data-driven decisions are what we strive for, ironically, we often find that without these qualifications, the numbers alone don’t present enough data for us to make strategic changes to the status quo.
The time and place for HR metrics
At the end of the day, HR metrics have to be as accurate and fit for purpose. Most critically, they must be meaningful to the audience that they’re presented to. In this journey, we’ve gone to both ends of the extreme, with fully qualitative HR — which, to no one’s surprise, we found we couldn’t scale past a team of 20 — and fully quantitative HR. As for our approach today, for what it’s worth, we try to err on the side of being more qualitative than quantitative. If you are still measuring the success of your HR practices by attrition, share with us how it has worked or not worked for you!
tl;dr
Philosophy: We believe that compensation is, beyond a method of compensation for work done, is a form of bestowing honour upon an individual for their craft.
Attrition is a relic of obscure practices, where tenure is rewarded instead of performance.
We use employee satisfaction and top talent turnover as a measure of HR success.
At the end of the day, HR metrics must be fit for purpose, accurate and meaningful to the audience they’re presented to.
Net Promoter, Net Promoter System, Net Promoter Score, NPS and the NPS-related emoticons are registered trademarks of Bain & Company, Inc., Fred Reichheld and Satmetrix Systems, Inc.
