Unique Surfboard metrics:
Other metrics:
Utilisation
"Utilisation" expresses the percentage of time that an agent was scheduled on service activities* in the selected period.
This metric provides a view of how well utilised agents were on service, and can be used to add context to other metrics. For example, an agent with low utilisation is likely to work on much less tickets than an agent with high utilisation
Here is the formula used to calculate "Utilisation":
View example
View example
Total scheduled time, i.e. shift length: 8 hours
Time spent on service activities, e.g. Phone: 4 hours
So, the utilisation in this example is 50%. This means that 50% of the total scheduled time was actually used for service activities.
*Service activities are activities that are linked to a ticket group. Read more here
Total closed
"Total closed" counts the total number of tickets with a 'closed' status.
For Zendesk integrations "Total closed" = number of "Closed" and "Solved" tickets.
This metric provides a measure of total output.
This data will change slightly depending on if you're viewing the metric per agent or per activity:
Agent: That were assigned to the agent at the time the status changed to 'closed'.
Activity: That were set to 'closed' by agents who were scheduled on this activity at the time.
Closed per hour
"Closed per hour" represents the average number of tickets closed for each hour that was scheduled for work.
This metric provides a insight into the output per agent relative to the invested hours. This can be helpful when measuring BPOs or analysing cost-to-serve.
Here is the formula used to calculate "Closed per hour":
View example
View example
Example calculation:
Total tickets closed: 20 tickets
Total scheduled hours (i.e. shift length): 8 hours
So, the "Closed per hour" in this example is 2.5. This means that, on average, 2.5 tickets were closed for every hour that was scheduled.
Closed per service hour
"Closed per service hour" represents the average number of tickets closed for each hour that was specifically scheduled for service activities.
This metric provides a fairer, unbiased measure of productivity by removing any time that was not scheduled on service. This is helpful when agent's utilisation can vary each shift.
Here is the formula used to calculate "Closed per service hour":
View example
View example
Total tickets closed: 20 tickets
Total scheduled time, i.e. shift length: 8 hours
Total hours scheduled on service activities, e.g. Phones: 4 hours
So, the "Closed per service hour" in this example is 5. This means that, on average, 5 tickets were closed for every hour scheduled specifically for service activities.
✨ Unique to Surfboard ✨
Time-on-task
"Time-on-task" represents the percentage of total scheduled hours that were spent on-task.
This metric is an alternative to the industry standard of "adherence", and indicates how well the agent actively followed their schedule across the entire shift.
Unlike other tools, Surfboard verifies if the agent is actively working by using tracked activity from linked customer service platforms, rather than relying on shallow data such as status
Here is the formula used to calculate "time-on-task":
View example
View example
Total Scheduled Hours: 8 hours
Time Spent Off-Task: 1 hour 30 minutes (1 hour offline and 30 minutes working on the wrong task)
First, convert the total off-task time into hours:
Next, calculate the hours spent on-task:
Then, calculate "time-on-task" as a percentage:
So, the "time-on-task" in this example is 81.25%. This means that 81.25% of the total scheduled time was spent on-task, correctly following the planned schedule.
What does on-task mean?
When agent's are scheduled on service activities, their activity is tracked within the linked customer service platform (e.g. Zendesk). If no activity is tracked, then an agent will be marked as 'off-task', and this time will be detracted from their score.
When agent's are scheduled on non-service activities, breaks, or meetings, it is expected that they won't be working in the customer service platform (e.g. Zendesk), so no activity is classed as 'on-task'.
This definition applies to both time-on-task and occupancy.
Occupancy
"Occupancy" is the percentage of the total hours scheduled for service activities that were actually spent on-task.
This metric helps to focus in on time spent on-task while agents were scheduled on service, removing any time lost at other times, e.g. a meeting overran, or the agent went to lunch late.
Here is the formula used to calculate "time-on-task":
View example
View example
Total Scheduled Hours: 8 hours
Hours Scheduled on Service Activities: 4 hours
Time Spent Off-Task: 1 hour
First, calculate the hours spent on-task for service activities:
Then, calculate "occupancy" as a percentage:
So, the "occupancy" in this example is 75%. This means that 75% of the total hours scheduled for service activities were actually spent on-task, correctly following the schedule.
✨ Unique to Surfboard ✨
Average interaction time (AIT)
"Average interaction time", abbreviated as "AIT", measures the average time it took agents to complete each interaction. An interaction is an event related to a ticket, i.e. a comment was added/sent, or the status of the ticket changed (e.g. open -> closed).
This metric helps in understanding how much time, on average, is dedicated to each interaction within a given period.
Here is the formula used to calculate "time-on-task":
View example
View example
Total Time Scheduled on Chat: 2 hours
Number of Interactions: 50
First, convert the total scheduled time into minutes:
Then, calculate the "average interaction time":
So, the "average interaction time" in this example is 2.4 minutes. This means that, on average, each interaction takes 2.4 minutes.
✨ Unique to Surfboard ✨
Average conversation time (ACT)
"Average conversation time", abbreviated as "ACT", measures the average total active working time spent to close a ticket, removing waiting or idle time.
This metric is an alternative to the industry standard of "average handle time", abbreviated as "AHT".
Many ticketing systems and WFM tools rely on measuring time from open to close as their average handle time.
This is flawed in that, for asychronous channels like chat or email, the time is inflated greatly by time spent waiting for a response. In the majority of these cases, agents are able to move onto other work, i.e. they are not actually sat waiting, which means by using this AHT you are going to inflate your staffing requirements.
Here is the logic used to calculate "average conversation time":
When we build the activity timeline, we group together interactions that happen on the same ticket within a reasonable time* of each other into ‘ticket sessions’.
Once the ticket is closed, we add together all of the ticket sessions to produce the total conversation time
The average conversation time is made by getting the average of all closed tickets on that activity for the period viewed in the performance page, e.g. day, week.
Here's a visual representation:
View example
View example
Ben exchanges 3 messages with a customer on live chat. The customer then stops responding. These 3 messages form 1 ticket session that lasted 300s.
The customer responds again after work and the 2 more messages are sent, then the conversation is closed. These 2 messages form another ticket session that lasted 240s.
There are now 2 ticket sessions on this ticket, from open to close. These are added together to produce a conversation time of 540s.
300s + 240s = 540 seconds