Getting Talent Retention Right: What Data to Track
Track data to see what is changing, then investigate further.
"The most exciting phrase to hear in science, the one that heralds new discoveries, is not Eureka! (I found it!) but rather, 'hmm... that's funny...'" - Isaac Asimov
The same is true for retention data. Data rarely shouts: "Danger, danger!" Mostly, something odd shows up as a clue that there's more to uncover.
"Data can often warn us of issues before we notice them," explains Ahu Yildirmaz, vice president and co-head of the ADP Research Institute. "Understanding your data and having a regular schedule for reporting and analysis helps organizations see what's happening. Tracking data over time allows you to spot potential problems earlier, before they become bigger problems."
For employee retention, it's not enough to just monitor turnover rates. When you only focus on who is leaving, you ignore the most important part of retention: who is staying. As Yildirmaz suggests, "Get as complete a picture as you can about what is happening in your organization. Benchmark your data internally over time and against data for your organization's size and industry. Become familiar with what is your 'normal,' so you can understand when things are changing."
To see your organization's retention data picture, here is the data to start with.
Turnover rate tells you who is leaving. Benchmark your overall rates against similar organizations in your area and track your internal rates monthly to spot changes and trends. For retention, focus on voluntary turnover rates.
Tenure rate tells you who is staying. Know the average tenure rates for your industry, and levels of hierarchy. You can also check specific departments, teams or individual jobs. Where you have long tenure, talk to those who have been around a while to get their insights on what is changing, what is working and what can be improved. And when you have departments with long lengths of service, plan ahead for retirements.
Flight risk predictions use data correlations for many aspects of employment to determine a likelihood that employees will leave soon. This can give you an overall sense of anticipated turnover to understand your recruiting needs and do workforce planning. It can also help you identify where you need to intervene. Like all predictive analytics, a prediction is an indication of likelihood not a pronouncement. So, if you are concerned about individuals, don't just rely on the prediction. Find out what is actually happening.
Terminations/reasons tell you why people left. This data is often kept as part of separation records. Also consider doing exit interviews when people leave voluntarily to learn more about why. Some organizations like to wait a few months after people leave to do a survey or phone interview because emotions are not as high and the former employee is often more comfortable offering insights.
Engagement data tell you what people think and how people are feeling about their work and the organization. This gives you clues into all culture issues including retention. If you send out surveys about particular issues, be prepared to take action or explain why you can't make a change right now. You're asking for people's time, thoughts and feelings. If you do nothing, it can backfire because people may not feel heard or cared about.
Survey response rates tell you who is willing to respond. There are both good and not so good reasons for low survey response rates, such as people are too busy, you're sending too many surveys (leading to survey fatigue), or people are really unhappy and don't want to say anything. Urging people to respond can signal that you want them to say what you want to hear. This will give you inaccurate data. So, if your response rates are low, take it as a suggestion to talk to people and find out why.
Absence rates tell you whether people want to and can show up at work. If absence rates are trending up, look at specific divisions, locations and other factors in your data to see where problems are — then investigate. There may be environmental issues that need to be addressed, like a difficult manager or a team that has worked really hard and just needs some time off. Check this monthly so you can see seasonal trends and account for normal fluctuations during summer and the holidays.
Average earnings tell you about your compensation within your organization and can be benchmarked against similar organizations. Look at employee demographics to see whether there are pay equity issues, which can be indicators of larger issues. Tracking compensation is essential to both retention and staying competitive in recruiting new employees.
Time to promotion tells you how long it takes the average worker to advance. This will depend on how your organization is structured. So, when you are benchmarking against other organizations, make sure you look for similar structures, as well as size and industry. ADP's data shows retention usually increases as employees become managers, with the biggest shift happening when they first get reports.
If you see something that makes you say "hmmm . . . that's funny," then focus in on where the data shows changes and narrow the focus by location, division, department or employee group. Try a pulse survey. Talk to your managers to find out what they are learning during check-ins. Then go back to your data to see if what you're hearing is consistent with what the data shows. If not, investigate further.
Having accurate data and a comprehensive picture of what is happening in your organization and the market is the first step in creating an effective retention strategy.
Go deeper
For more on how to effectively use data in your organization, read the The Workforce Analytics Workbook.
Download the report: Getting Your Retention Strategy Right.
Read our Getting Talent Retention Right series:
Do You Have a Retention Problem?
Common Causes of Voluntary Turnover
