The Forecasted vs. actual report provides interval-level breakdowns that help you understand the performance of your staffing plan. In particular, the report helps you understand when you missed SLA across each forecast, and which planning factors you might adjust in the future to improve performance. It's structured to compare what was forecasted to happen (volume & productivity-wise) with what actually happened.
Because the report anchors on assessing one forecast at a time, you should filter the report to match the level that you do staffing. In other words, you must select a Channel, and if you forecast & staff at a queue level, you should filter the report to that queue.
The report is comprised of a chart at the top, a table beneath it, and exports of "actuals."
Chart for forecasted vs. actual metrics over time
Table for interval-level SLA breakdown
Chart for forecasted vs. actual metrics over time
After indicating which Channel, Queue, Date range, and Interval you'd like to report on, select from the dropdown of metrics on the top left of the chart to indicate which you'd like to view. The X axis on this graph represents the time period, while the Y axis corresponds with metric values. This graph will automatically first open to Cases
Opened or Contacts Received depending on the filtered channel (see metric definitions here).
For any of these metrics, the red text indicates what was Forecasted to happen at each particular time interval, while the purple records what actually happened at that time interval. While some variance between Forecasted and Actuals is expected, optimal planners aim for minimal discrepancy. Check your forecast settings on the Forecast management page and and add outliers and adjustments on the Forecast configuration page.
Table for interval-level SLA breakdown
The table is designed to be read left to right, row by row as follows:
- Did I hit SLA during this time interval (interval indicated in the first column)?
- How did the plan that I created based on my forecast hold up in actuality? Specifically, did I get more or less volume than I expected? Were my agents more or less efficient than expected? Did I schedule the recommended number of headcount?
Comparing each metric's actual values to its forecasted values will help you iterate on your planning process to optimize for SLA most efficiently.
You can customize which values you'd like to include in the report by clicking the "Select metrics" button in the top right of the report and checking / unchecking options. These changes will only apply at the user level (other users at your company will have the ability to show or hide different metrics).
Forecasted values will be on the left, with actuals on the right in each metric column. At the bottom of the table, there are average, minimum, maximum, and sum values as relevant for each metric in aggregate over the filtered time period.
Clicking the column headers in the table with the metric names will change which metric is shown in the chart above.
Case breakdown
Clicking certain, hyperlinked metrics (the "Actuals" in each column), will allow you to view the case-by-case breakdown of a metric, showing you which cases were included in each interval's calculation of a particular metric. The case breakdown will show information given the most up-to-date data at the time you click (as opposed to the report itself, which is pre-computed on regular intervals to optimize for loading speed). Expanding "Show calculation details," where available, will give you information about relevant behind-the-scenes configurations that you can request to change and integration-specific explanations that can help you understand how a metric was calculated.
Case activity timeline [Alpha]
Clicking "View timeline" will open the Case activity timeline, which is currently in Alpha, and only available to customers included in the pilot build.
The Case activity timeline allows you to better understand how the metrics in the Forecasted vs. actuals report are being calculated (only applicable for case-based performance metrics like average handle time, cases solved, and messages sent).
Exports
Whole report
Users can export the entire table into a downloaded CSV file, or to a Google Sheet.
Interval-level metrics
After clicking the hyperlinked numbers to get the Case breakdown, the side panel will open and show an export button in the top right corner. This button will export the case-by-case breakdown of a metric, for the selected time interval and applicable filters, into a CSV file.
Metrics Glossary
Service level: Service level is the percentage of contacts that meet the target response time you customize in your configurations for each channel, excluding any contacts that do not have an SLA.
Service level missed: Service level missed is the number of contacts that were responded to beyond your target response time.
Service level met: Service level met is the number of contacts that were responded to within your target response time.
Occupancy: Percent of logged-in time that an agent spent actively working on customer issues. For example, if Sam is available on phones for 8 hours, and on calls with customers directly or in wrap-up time for 6 hours, her occupancy rate would be 75%.
Scheduled staffing: The amount of people currently scheduled to that channel by day and time. This is based on agent schedules in the Staffing Timeline for the selected interval.
Required staffing: The amount of people that we predict you'll need to have staffed during the time interval to meet your SLA goals.
Net staffing: The amount of people you have staffed during the time interval for the channel (scheduled staffing) minus the amount of people we predict you'll need staffed (required staffing).
Channel specific metrics:
Contacts received (phones/chats): Contacts received is the total amount of calls that came in during the time interval.
Avg. time to answer (phones): Avg. time to answer is the average time a customer was waiting to talk to an agent.
Handle time (phones/chats): Handle time is the average time an agent spends on a call. This metric does not include time the customer is waiting to be connected or any wrap up time related activities in the average, only time when the agent is directly talking with customers.
First response (chat/email): Similar to "avg. time to answer" for phones, this metric is the average time a chat/email contact waits for a first response.
Cases opened (email): All email tickets initially created during the specified time interval.
Cases reopened (email): Email tickets that switched from a "responded" state (done, closed, solved) to an "unresponded" state (new, open).
Cases solved (email): Email tickets with a status change from {new, open} → {pending, solved, closed} during the time period. Each instance of a solve will be counted (not just the last solved).
Comments
0 comments
Please sign in to leave a comment.