Leavers communicate with the band - Why people leave 2/3

On Thursday 23 May 2024, we published the results of our Why People Leave survey. In our previous blog, we reported that the biggest differences between the responses of those who left and those who stayed were found in questions related to learning and development practices. This theme has increasingly emerged in a variety of studies around the world. This was the first time that it has emerged on this scale in Finland. However, it is not worth looking at the responses alone. The response rates are equally telling.

How was the research carried out?

The survey was carried out by analysing 1 000 000 responses from 85 500 people to various staff surveys over the last 6 years. The analysis looked at how the response of those who left differed from those who stayed. The data is truly massive by Finnish standards.

The nature of the data is such that the result does not reveal causalities (I left because) but correlations (having left, at the same time having responded like this). With common sense we can conclude that there is of course a link between these things, but strictly speaking the result cannot tell us the causes, it tells us about the differences between the response rates and responses of those who left and those who left.

What did you find?

The biggest differences in responses were therefore found in the questions on learning and development, job satisfaction and local management. However, it is not worth looking at the results alone. The response rate of leavers fell systematically and by roughly the same amount for all questions and themes. Across the board, the response rate of leavers was about 20 percentage points lower than that of stayers. The response rate was about 10 percentage points more critical than for those who stayed. Sometimes the difference was significantly smaller - as little as 3 percentage points!

"Across the board, the response rate of those who left was about 20 percentage points lower than those who stayed. Responses were about 10 percentage points more critical than for those who stayed. "

The survey first analysed the data according to the main themes of our questionnaire. Of these, it was the themes of learning and development, local management and job satisfaction that showed the greatest differences between those who responded and those who did not. However, the differences between the main themes are not very large. The difference is more in the response rate.

• Main theme of learning and development:
  6 percentage points lower results
  22 percentage points lower response rate
  1.6x more likely to leave if not answering
  
  Result is the difference between the proportion of positive responses (stayed - left)
  Sample: 31 000 people
  Responses: 196 000. 38 000 comments.
  Probability of leaving calculated using Bayes' theorem.

• Main theme of job satisfaction:
  3 percentage points lower results
  27 percentage points lower response rate
  1.9x more likely to leave if person does not answer the question
  

  The result is the difference between the proportion of positive responses (those who stayed - those who left)
  Sample: 47 000 people
  Responses: 206 000. 42 000 comments.
  Probability of leaving calculated using Bayes' theorem.

• Main theme of local management:
  3 percentage points lower results
  24 percentage points lower response rate
  1.7x more likely to leave if person does not answer the question
  

  The result is the difference between the proportion of positive responses (those who stayed - those who left)
  Sample: 47 000 people
  Responses: 571 000. 132 000 comments.
  Probability of leaving calculated using Bayes' theorem.

  
  
  In the subcategories of the main themes, a greater difference was found between leaving and remaining responses.

• Leadership questions:
  10 percentage points lower results 18 percentage points lower response rate
  1.5 x more likely to leave if no response
  
  Result is the difference between the proportion of positive responses (stayed - left)
  Sample: 28 000 people
  Responses: 75 000. 33 000 comments.
  Probability of leaving calculated using Bayes' theorem.

• Subcategory: questions related to recommendations:
  11 percentage points lower results

  22 percentage points lower response rate

  1.7 x more likely to leave if the person does not answer the question
  
  The recommendation subcategory can be found in our question bank under the main topic of job satisfaction.
  The result is the difference between the proportion of positive responses (stayed - left)
  Sample: 15,000 people
  Responses: 47,000. 7 000 comments.
  Probability of leaving calculated using Bayes' theorem.
  

• Subcategory: questions related to learning and development practices:
  11 percentage points lower results

  20 percentage points lower response rate

  1.6 x higher probability of leaving if the person does not answer the question
  
  The subcategory Learning and development practices can be found in our question bank under the main theme of learning and development.
  The result is the difference between the proportion of positive responses (stayed - left)
  Sample: 27,000 people
  Responses: 39,000. 6 000 comments.
  Probability of leaving calculated using Bayes' theorem.
  

What did we learn?

The criticality of the answers does not necessarily reflect the person's starting intentions. Rather, we should look together at the response results and the response rate. Across the board, the response rate of those who left was about 20 percentage points lower than those who stayed. The responses were about 10 percentage points more critical than for those who stayed. By following this rule of thumb, it is possible to reach a large number of people who are about to leave and, if desired, to try to establish a new working relationship between employer and employee.

In some cases, people prefer not to answer rather than express their dissatisfaction. This seems to be the case, for example, with local management. Those who left were only 3 percentage points more critical than those who stayed. So it may be that the management is not working, but there is still an emotional bond with the immediate manager and he or she is, for example, seen as a "good guy", so that one does not want to give crushing feedback, even if one secretly thinks so. It is easier not to respond. In this category of questions, it is therefore particularly important for the line manager to monitor the response rate and, as it falls, to try to stimulate an open dialogue to give feedback and discuss how the work is going.

Does this result also apply to you?

Probably yes, but if you want to be sure, we can do a similar analysis of your staff survey data. Then you'll know exactly which factors are most important to your retention rate, in which organisational line, for which occupational group and, for example, for which age group. This will enable you to strengthen your retention by "pinpointing" rather than "spot-firing".

For more information on the study, please contact Mikko Ruokojoje, mikko.ruokojoki@vibemetrics.com.