Our Approach

Why Did You Leave?

Virtually all organizations conduct exit interviews to determine why their employees are leaving. However, the traditional method, with its inherent flaws, fails to produce reliable, actionable data. We offer a more reliable and comprehensive approach. But first, let’s delve into the four reasons why the traditional method falls short.

One of the main drawbacks of exit interviews is their timing. Often, employees are ‘interviewed’ just as they are about to leave. In such a situation, the employee has little incentive to be completely honest. They are unsure about their future success and are reluctant to risk a negative reference from their current employer. This leads to a lack of transparency and potentially skewed data, a significant concern in the realm of employee departure data. Further, changing jobs is a relatively high stress situation and any feedback that is given may be emotionally charged and not necessarily reflective of the employees’ experience.

The exit interview is often conducted by one person who has to ask questions, assess the answer, determine the response, and make notes. A wealth of evidence shows that this level of multitasking inevitably leads to errors in documenting what was actually said.

Another significant issue with exit interviews is the potential for bias. These interviews are typically conducted by a single person, who brings their own experiences, biases, and perspectives to the table. This can lead to interpretations that may not be accurate and, in some cases, the interviewer may record what aligns with their expectations, rather than what the employee actually says. This phenomenon, known as ‘confirmation bias’ where the interviewer seeks to confirm their preconceived notions, can significantly distort the data collected.

Another issue with traditional exit interviews is that the data is usually written down but rarely shared beyond the immediate HR team and functional manager group. This lack of data sharing limits the potential for a comprehensive understanding of employee departures and hinders the development of effective retention strategies.

Our approach addresses each of these problems.

We typically recommend collecting data approximately three months after the employee has left. This delay allows the former employee to settle into their new role and lessen their anxiety about the risks associated with honesty. It also allows emotions to settle and experiences to be placed in an appropriate context.

We guarantee anonymity to all respondents unless they choose to disclose their identity. Our customer is responsible for providing our details to their departing employees, and we are responsible for collecting their information. We NEVER disclose raw data back to the customer. We use our AI tools to summarize the information we receive, aggregated amongst at least ten respondents to ensure that information is not connected to any specific person.

Our data collection process involves an interactive AI tool that asks consistent questions, probing deeper where appropriate and skipping to the next question where not. The ex-employee records their responses directly into the tool, vastly reducing the risk of recording errors and eliminating the risk of third-party bias.

Our reports summarize and show trends within the company, for example, over time and between teams. We also provide some color on external experience—is it the “great resignation,” or is something else happening?

We distill everything down into a simple two-dimensional graph, which records on a scale of 0 – 100 whether the employee left because they were attracted to a new role (”Pull” with a score of 100) or felt that they had to leave (”Push” with a score of 0) and on a scale of 0 – 100 whether the employee exhibits anger/resentment in their response (”Detractor” with a score of 0) or whether they speak highly of the company (”Promoter” with a score of 100).

Ready to Understand the WHY?

Gain deep insights into employee behavior and drive better decisions with data you can trust.