Predictive routing comparison test

AI-generated summary

Genesys Cloud allows you to directly compare predictive routing performance against your current routing method using comparison tests. Genesys Cloud enables you to run comparison tests to evaluate the impact of predictive routing on key performance indicators (KPIs) like average handle time or customer satisfaction. During a comparison test, Genesys Cloud routes your interactions in an hour-on, hour-off pattern, using predictive routing 50% of the time and an alternative routing method the other 50%. You can start a comparison test from the Queue details page by selecting Predictive Routing, setting the KPI and timeout value, choosing the routing method to compare, and configuring the test settings. Or, you can start from the Queue Benefit Assessment pane after a benefit assessment by clicking Start Routing Comparison Test. Once started, the test runs with Predictive [Processing...] shown in the Routing Method column. When complete, the column shows Predictive Test Mode. The Comparison Test Results pane opens to show the test progress. This pane also allows you to end a test in progress or turn on predictive routing for all interactions on the queue. You can view test performance in the queue performance report. Genesys recommends running tests for at least two weeks to produce statistically significant results. For high volume queues, robust results may come sooner. If a queue shares agents with others, test them together to reflect overall benefit and minimize impact. When starting a test, predictive routing takes time to build the model before becoming available. Changing the tested KPI enters a new processing phase.

Using comparison tests, you can directly compare predictive routing performance against your current routing method. During a comparison test, Genesys Cloud routes your interactions in an hour-on, hour-off pattern, using predictive routing 50 percent of the time and an alternative routing method the other 50 percent.

To start comparison testing, click Queues from the Admin console. The Queue list window opens.

The following video shows you the steps to run a comparison test. The written procedure follows the video.

You have two choices for where to start your comparison test. The two offer the same configuration alternatives but follow different routes through the workflow, so a separate section explains each alternative.

Start from the Queue details pageClick to expand

Click a queue name. The Queue details window opens.
From the Routing tab, select Predictive Routing as the routing method. By default, the window defaults to comparison test mode and the Predictive Routing Comparison Test pane opens on the right side of the window.
Set the key performance indicator (KPI) and timeout value. The benefit assessment for this queue indicates which KPI shows the best potential for optimization.
Choose which routing method to compare with predictive routing. You can choose standard or bullseye routing.
Configure your comparison routing method as explained in the “Specify routing method” section of Create and configure queues.
To use workload balancing with predictive routing, set Workload Balancing to On.
To use skill matching with predictive routing, set Skill Matching to On.
Set the evaluation method for the fallback routing method. The fallback is always standard routing.
Click Save and Continue to make more changes or Save if you are done. After you click Save, the Queue list page reopens, showing Predictive [Processing…] in the Routing Method column. For more information, see The KPI processing phase.

Start from the Queue Benefit Assessment paneClick to expand

After running a benefit assessment, click the Optimization Potential link in the Predictive Routing Evaluation column for the queue you want to test to open the Queue Benefit Assessment pane.
Click Start Routing Comparison Test. The Predictive Routing Comparison Test pane opens. If you already ran a comparison test on the queue, the button reads Run Another Test.
Choose the KPI to optimize. Predictive routing defaults to the KPI with the best potential for optimization.
To change the timeout value from the default, 30 seconds, to a different value, click Advanced test settings.
(Optional) Configure the routing method you are comparing with predictive routing. By default, predictive routing is compared with standard routing. To make changes, do the following:
1. Click More settings. If you do not see this link initially, click Advanced settings to open the additional configuration fields.
  The Queue details window opens.
2. Choose which routing method to compare with predictive routing. You can choose standard or bullseye routing.
3. Configure your comparison routing method as explained in the “Specify routing method” section of Create and configure queues.
4. To use workload balancing with predictive routing, set Workload Balancing to On.
5. To use skill matching with predictive routing, set Skill Matching to On.
6. Set the evaluation method for the routing method you are comparing with predictive routing.
To start the comparison testing process, click Start Test. The Queue list page reopens, showing Predictive [Processing…] in the Routing Method column. For more information, see The KPI processing phase.

When the processing is completed, the Routing Method column shows Predictive Test Mode and the Predictive Routing Evaluation column shows Test Mode.

Next Step

Click Test Mode from the Predictive Routing Evaluation column in the Queues window. The Comparison Test Results pane opens. The results pane also enables you to end an in-progress comparison test or turn on predictive routing for all interactions on the queue. Alternatively, you can also view the performance of the comparison test from the queue performance report. For more information, see View performance of queues with predictive routing.

Notes:

Genesys recommends that you run the comparison test for at least two weeks. This period typically provides a large enough number of interactions to produce a statistically significant result. For some high volume queues, you might get robust results sooner.
If a queue you plan to test shares a group of agents with one or more other queues, Genesys recommends that you test predictive routing on all of those queues at the same time. This simultaneous testing reflects the benefit across all queues and minimizes any potential impact on other queues as a result of activating predictive routing.
When you start the comparison test, predictive routing requires processing time to build your model for testing. During this processing phase, predictive routing is not yet available. See The KPI processing phase for how Genesys Cloud handles routing during this period.
If you have selected a KPI and saved that choice, and then change the KPI and save again, predictive routing enters a new processing phase. This is true even if you later revert to your original KPI.

[NEXT] Was this article helpful?

Get user feedback about articles.

Predictive routing comparison test

AI-generated summary

To start comparison testing, click Queues from the Admin console. The Queue list window opens.

The following video shows you the steps to run a comparison test. The written procedure follows the video.

Start from the Queue details pageClick to expand

Click a queue name. The Queue details window opens.
From the Routing tab, select Predictive Routing as the routing method. By default, the window defaults to comparison test mode and the Predictive Routing Comparison Test pane opens on the right side of the window.
Set the key performance indicator (KPI) and timeout value. The benefit assessment for this queue indicates which KPI shows the best potential for optimization.
Choose which routing method to compare with predictive routing. You can choose standard or bullseye routing.
Configure your comparison routing method as explained in the “Specify routing method” section of Create and configure queues.
To use workload balancing with predictive routing, set Workload Balancing to On.
To use skill matching with predictive routing, set Skill Matching to On.
Set the evaluation method for the fallback routing method. The fallback is always standard routing.
Click Save and Continue to make more changes or Save if you are done. After you click Save, the Queue list page reopens, showing Predictive [Processing…] in the Routing Method column. For more information, see The KPI processing phase.

Start from the Queue Benefit Assessment paneClick to expand

After running a benefit assessment, click the Optimization Potential link in the Predictive Routing Evaluation column for the queue you want to test to open the Queue Benefit Assessment pane.
Click Start Routing Comparison Test. The Predictive Routing Comparison Test pane opens. If you already ran a comparison test on the queue, the button reads Run Another Test.
Choose the KPI to optimize. Predictive routing defaults to the KPI with the best potential for optimization.
To change the timeout value from the default, 30 seconds, to a different value, click Advanced test settings.
(Optional) Configure the routing method you are comparing with predictive routing. By default, predictive routing is compared with standard routing. To make changes, do the following:
1. Click More settings. If you do not see this link initially, click Advanced settings to open the additional configuration fields.
  The Queue details window opens.
2. Choose which routing method to compare with predictive routing. You can choose standard or bullseye routing.
3. Configure your comparison routing method as explained in the “Specify routing method” section of Create and configure queues.
4. To use workload balancing with predictive routing, set Workload Balancing to On.
5. To use skill matching with predictive routing, set Skill Matching to On.
6. Set the evaluation method for the routing method you are comparing with predictive routing.
To start the comparison testing process, click Start Test. The Queue list page reopens, showing Predictive [Processing…] in the Routing Method column. For more information, see The KPI processing phase.

When the processing is completed, the Routing Method column shows Predictive Test Mode and the Predictive Routing Evaluation column shows Test Mode.

Next Step

Click Test Mode from the Predictive Routing Evaluation column in the Queues window. The Comparison Test Results pane opens. The results pane also enables you to end an in-progress comparison test or turn on predictive routing for all interactions on the queue. Alternatively, you can also view the performance of the comparison test from the queue performance report. For more information, see View performance of queues with predictive routing.

Notes:

Genesys recommends that you run the comparison test for at least two weeks. This period typically provides a large enough number of interactions to produce a statistically significant result. For some high volume queues, you might get robust results sooner.
If a queue you plan to test shares a group of agents with one or more other queues, Genesys recommends that you test predictive routing on all of those queues at the same time. This simultaneous testing reflects the benefit across all queues and minimizes any potential impact on other queues as a result of activating predictive routing.
When you start the comparison test, predictive routing requires processing time to build your model for testing. During this processing phase, predictive routing is not yet available. See The KPI processing phase for how Genesys Cloud handles routing during this period.
If you have selected a KPI and saved that choice, and then change the KPI and save again, predictive routing enters a new processing phase. This is true even if you later revert to your original KPI.

[NEXT] Was this article helpful?

Get user feedback about articles.

Predictive routing comparison test

Next Step

[NEXT] Was this article helpful?

Loading...

Predictive routing comparison test

Next Step

[NEXT] Was this article helpful?