Custom KPI for predictive routing
Custom KPI enables your organisation to optimise KPIs for strategic outcomes such as sales conversion, churn, and customer retention. This type of optimisation is based on following types of tracking mechanisms:
- Using wrap-up codes such as ‘Sale successful’ and ‘Subscription renewed’ that track the achievement of a desired outcome. Genesys Cloud uses information from an internal data source for tracking and measurement of wrap-up code based custom KPI.
- Using outcomes that track the achievement of a pre-determined goals that Genesys Cloud receives from an external system. Genesys Cloud uses information from an external data source for tracking and measurement of outcome-based custom KPI.
When you define your custom KPI, you will decide which tracking mechanism to use and the type of KPI. Once defined, the custom KPIs can be used in queues that use predictive routing.
How wrap-up code-based custom KPI works:
- You require a custom KPI using the tracking mechanism ‘wrap-up code’ of type ‘Sales conversion’.
- Create a new custom KPI with the name ‘Home Insurance sale’ that tracks the number of interactions that have the wrap up code ‘New home insurance sale’.
- At the time of queue activation, select the ‘Home Insurance sale’ KPI from the list of available KPIs.
- When a new interaction arrives, predictive routing will try to route the interaction to the agent predicted to have the best chance of achieving the ‘New home insurance sale’ outcome for that particular customer.
How outcome-based custom KPI works:
- You require a custom KPI using the tracking mechanism ‘Outcome’ of type ‘Customer churn’.
- Create a new custom KPI with the name ‘Credit card customer exit’.
- From the Custom KPI page, click ‘Credit card customer exit’ custom KPI and save the auto-generated outcome ID dedicated for this KPI outcome.
- Using the outcome Attributions API, send information of all interactions in which the agents prevented customers from quitting the service along with the outcome ID that you saved in step 3.
- At the time of queue activation, select the ‘Credit card customer exit’ KPI from the list of available KPIs.
- When a new interaction that could be a potential customer churn arrives in the queue, predictive routing will try to route the interaction to the agent predicted to have the best chance of achieving the ‘Customer churn’ outcome.
- Genesys Cloud leverages wrap-up codes and outcomes as conditions to track the desired results to score and select agents for an incoming interaction.
- Wrap-up code-based custom KPI: Genesys Cloud uses interactions that have associated wrap-up codes that summarize the interaction outcome to train the machine learning models. Genesys recommends that you use specific wrap-up codes which capture the actual outcome of the action. For example, to track sales conversion, use the wrap-up code that agents add when a sale is complete and not the wrap-up code that agents use when a sales pitch is made.
- Outcome-based custom KPI: Genesys Cloud uses pre-configured outcomes (known as types) to track the result of interactions and relies on data from external sources. When you create an outcome-based custom KPI for a pre-configured outcome, a dedicated outcome ID is created by Genesys Cloud to map the outcome with KPI. Once the queue with the KPI is activated, Genesys Cloud uses the outcome ID to track the achievement of the outcome and the touchpoints that led to the outcome. For example, when an agent achieves a sale, the sale outcome is tracked with an outcome ID that is dedicated for the pre-configured outcome ‘Sales conversion’ KPI. This information helps score the agent involved and use the score for future predictions. Use the Outcome Attributions API to send the interaction outcomes from the external data source to Genesys Cloud. For more information, see Outcome Attributions API.
For better data model training and prediction results, outcome-based custom KPIs are best suited for queues that service a specific outcome rather than a support queue that caters to all types of requests.
- Currently, Genesys Cloud offers the following types of KPIs for each condition :
- Sales conversion KPI type – Helps you maximize the sales conversion of an organization.
- Customer retention KPI type – Helps you maximize the number of customers who continue business with the organization.
- Customer churn KPI type – Churn indicates the number of existing customers who intend to discontinue business with the organization. Helps you minimise customer churn.
- Sales value KPI type – Available only for custom KPIs tracked with outcomes. Helps you maximize the sales value of an organization.
- Genesys recommends a periodic upload of outcome data to ensure better model training and prediction accuracy. The recommended frequency is once a day, or, at the minimum, once a week.
- Readiness of custom KPI for use:
- Wrap-up code-based custom KPI requires the wrap-up codes to be in use and with a significant number of interactions associated to them. If you are creating new wrap-up codes, Genesys Cloud waits until the volume of data reaches the required threshold value. See Sources of data for predictive routing decisions.
- Outcome-based custom requires the outcome data to be available through the Outcome Attributions API. The new custom KPI is available for use after 24 to 48 hours of the upload of outcome data. However, prediction is more accurate when the number of outcomes tracked with the outcome ID of the KPI is high.
- Custom KPIs are at the organization level. All divisions and queues in the organization can use the KPIs defined.
- You can use the custom KPIs to perform a benefit assessment on your queue, activate predictive routing, run a comparison test, and evaluate the ongoing value.
- You can view the benefit of the predictive routing with custom KPIs from the comparison test results pane. Depending on the KPI set, the Benefit pane lists the actual number of sales conversion, number of customers retained, or number of customers that were prevented from leaving the service.