Peakon’s predictive analytics helps to predict costly turnover before it happens and delivers insights into the possible reasons why. The Attrition Prediction model estimates the attrition risk for your employee populations in real-time, which is recalculated every time an employee submits feedback. The aggregated, segment-level view keeps the accuracy of your predictions high while protecting individual employee identity. The ability to predict departures in your organization is extremely valuable, as it may assist you in optimizing your engagement strategy and prevent regrettable churn. Check out our Heartbeat report: The 9-Month Warning: Identifying Quitters before It's Too Late.
This article covers:
- How the attrition risk is calculated
- Enabling Attrition Prediction
- Improving the prediction accuracy
If you are looking to report on attrition risk, see Reviewing your attrition risk forecasts.
How the attrition risk is calculated
The Attrition Prediction model is trained on hundreds of thousands of Peakon employee data points. This model takes into account key factors for each employee and general trends to indicate a high risk of departure, including:
(References to survey question scores and behavior are based on our analysis of hundreds of thousands of aggregated survey responses across the Peakon customer base.)
Engagement |
Response to the eNPS question: “How likely are you to recommend [your organization] as a place to work?” We observed that employees who provide a 0-6 score to our eNPS question are more than 3 times more likely to resign than those who score the question 9-10. On average, employee engagement declines by as much as one point during the first two years of employment, regardless of whether the employee has any intention of leaving. This is expected and the decline is likely to level-out after two years of tenure. It’s natural that employees are most engaged when they first start a new job. Ideally, employee engagement wouldn’t decline at all but the fact that it does should not be alarming. Employees who have many years of tenure in a company would also have gone through this process. |
Loyalty |
Response to the eNPS loyalty question: “If you were offered the same job at another organization, how likely is it you would stay at this organization]?”. |
Growth |
Response to the Growth driver question: “I feel that I’m growing professionally.”. This question relates to an employee's perceived opportunities to improve their personal and career growth. Enabling this question will contribute towards a higher quality prediction. |
Responsiveness |
Engagement survey response rate. |
Tenure |
Time spent at the organization. Employees in the 3-12 month tenure bracket are most likely to leave an organization. This is also when the biggest drop in engagement occurs - after 3 months of tenure. This presents a paradox, as on the surface you will have an employee with higher engagement, relative to other tenure segments, but who is also more likely to leave your organisation. It’s therefore crucial to factor length of tenure into the attrition risk calculation as relying solely on engagement is not an accurate indication of attrition risk. |
The Attrition Prediction heatmap also displays a column with past resignations. The column is there for your reference, and the attrition model doesn't use this data in its calculations. The Resigned column is directly linked to the Resigned segment within the Separation Reason attribute.
Attrition risk levels
The attrition prediction model presents a segment's attrition risk relative to the rest of your organization. The risks are displayed within your segment heatmap and employee cycle reports, as six distinct risk brackets, designed to help your organization identify areas that may require further examination and attention:
Risk |
Meaning |
Severe |
The risk of attrition in this segment is in the top 10% of your organization. |
High |
The risk of attrition in this segment is in the top 25% of your organization. |
Elevated |
The risk of attrition in this segment is in the top 50% of your organization. |
Reduced |
The risk of attrition in this segment is in the bottom 50% of your organization. |
Low |
The risk of attrition in this segment is in the bottom 25% of your organization. Engagement in this segment is high. |
Minimal |
The risk of attrition in this segment is in the bottom 10% of your organization. Engagement in this segment is excellent. |
Note: You should consult with your legal counsel to determine whether your configuration of segments, and thus the attrition risk by segment, satisfies your organization’s compliance requirements. Customers can configure and provide instructions to their workforce on the use of this feature to ensure it does not directly or indirectly cause discrimination or discriminatory results, whether intentional or not. Customers are responsible for understanding and complying with any legal obligations arising from their use of the Attrition Prediction model and attrition risks, including any assessment, testing, or documentation that may be required under anti-discrimination laws.
Enabling Attrition Prediction
Attrition prediction is enabled by default for all administrators (the Administrator access control group). To enable this for other users within Peakon, administrators need to follow these steps:
- Click on Administration.
- Select Access control.
- Select the Access control group you’d like to enable it for.
- Under the Access statistics permissions toggle on the Attrition prediction feature.
Improving the prediction accuracy
Peakon’s Attrition Risk model works most accurately when the following is implemented:
- Increasing active survey participation whilst also actioning feedback.
- Moving to a higher survey frequency to identify trends and risks earlier - the more recent the answers, the more accurate the prediction of current risk.
- Enabling Peakon’s standard ‘Loyalty’ question.
- Collecting leaver data using Peakon’s Separation date and Separation reason attributes.
The above points are not required for the Attrition Risk feature to work. However, enabling them will improve your score accuracy and contribute to the overall aggregated data set used in the model.
Comments
0 comments
Article is closed for comments.