Recent searches


No recent searches

What are the main metrics you track to gauge agent success?



Posted Apr 18, 2024

Hi there! I was hoping to get some feedback regarding metrics we could use to track our agent performance without needing a separate (paid) integration. Right now one of our main focuses is on tickets solved per hour but this doesn't seem like the right metric to track for one reason: if a customer writes back in, the ticket opens again. 

So what metrics do you track? We also track CSAT, internal and public comments, ticket volume, response time, time spent in ticket, response/resolution time, first contact resolution, etc… but what is the combination of metrics you refer to when it comes to discuss performance. And is there one you heavily weigh on? I am also doing research on what other companies do but wanted to ask here first. 

 


2

7

7 comments

image avatar

Harper Dane

Zendesk Luminary

Hi there Naomi Greenall ,

I personally don't believe it's possible to obtain a complete view of agent performance using native Zendesk tools only. If you want a full understanding of agent performance (unless you have a very small tight-knit team) you'll probably need some add-ons sooner or later. That's because while Zendesk is excellent at tracking counts, those hard numbers often aren't actionable without deeper insights into root cause — it can be very difficult to determine the “how” and “why” of  these numbers using only native ZD metrics. 
For example: if an agent has a high solve rate, that could mean they're a super-efficient rock star. Or… it could instead mean they're cherry-picking your queue for easy tickets and leaving the heavy lifting to their colleagues. Solve rate also doesn't tell you if the agent is sending the correct answers to your customers on those tickets, or if they are following the correct processes to complete tasks.

 

Consider placing agent performance into 3 buckets: 

 

Ticket Efficiency Statistics (Hard Numbers):

  • Metrics: Tickets solved per day, time spent per ticket, one-touch tickets, reopens, average handle time, SLA adherence, etc. Check out the precanned “Zendesk Support” dashboard for many more examples.
  • If you haven't done so already, consider Zendesk's free Time Tracking app to enrich your time-related data.
  • Cost of Service: You can use data from the Time Tracking app to generate useful Cost Of Service metrics. While this probably isn't a report you want to share with agents, you can use this to identify possible training gaps and/or SOP and systems improvement opportunities.

Customer Satisfaction & Branding Metrics:

  • Metrics: Good:Bad customer ratings, % of cases escalated to a manager, brand/tone adherence, accuracy of answers provided. (You may need to build out a custom field + trigger framework to track escalations.)
  • If you do not have a formal QA process yet, consider tools like Klaus or Maestro for assessments.

Time Management Metrics:

  • Metrics: time + quality data like the above, but for work done outside of Zendesk (for example, in an OMS).
  • A Workforce Management (WFM) tool can provide data on agent activities done outside of Zendesk. This type of data will help you identify training gaps and streamline your SOPs, and provides deeper insights into what agents spend the most time on.

 

I hope this helps!

0


Hi Naomi, just on the when a customer replies on a Solved ticket if it is just a Thank You, we have a macro for agents to use that acts as a quick close - it moves the ticket back to Solved and adds a tag to the ticket. You can then filter them out when reviewing your reopened tickets.

0


image avatar

Brandon Tidd

Zendesk LuminaryUser Group LeaderThe Humblident Award - 2021Community Moderator

Hey Naomi,


I'm always a fan of comparing agents against the average.  If your team average is 3 hours to resolve a ticket and you have one agent that is taking 6 or 1, those are outliers worth investigating.  There may be a valid reason, an opportunity for coaching, or cause for celebration.  Service Level Agreement adhereance (even if you're only benchmarking internally) is also another good leading indicator.  Hope this helps!
 

Brandon

1


Harper Dane Thank you for the feedback! We are using the Zendesk Support dashboard and the time tracking app. We recently started calculating cost per ticket as recommended from another user in Zendesk. In terms of satisfaction we are using CSAT but are looking into other sources for a more in depth analysis. I'v also done some research on WFM software but there seems to be some gaps still in areas we need. 

 

Chris H I haven't heard of anyone else doing this before. When you run a report you just filter out that tag that has been assigned through the macro? Do you filter out any other tags maybe for other comments other than thank you?

 

Brandon Tidd I could see this metric being very useful for our team. Thank you for providing this feedback!

 

 

1


Hi Naomi, I didn't build out our Explore dashboard but I believe we are using the tag to help filter out in our customised dashboard.  We are not excluding any other tags at the moment. You could have a custom reopen field with additional reasons that the agent selects. Whilst you may have some of that data in your CSAT responses, you won't capture for all.

1


image avatar

Grace Maxon

Customer Success

Hi Naomi! 
 
Thank you for your question, and thank you, Christopher and Brandon, for your responses! 
 
There are lots of reports you can pull to measure agent productivity, outside of the ones listed above; here are some of our favorites: 
 
Ticket creation heatmap
Comparing an agent's tickets to the rest of the agent's group
Creating KPIs for tickets by status
Average ticket resolution time without pending or on-hold time
Reporting on agent reply brackets using the Updates history dataset
 
We also have our Explore recipes reference that has a ton of different report recipe cards you can follow to implement in your own instance! We also are expanding our CSAT capabilities to allow for more granular insights. 
 
Hopefully this helps! 
Grace

1


I think there is a significant difference in one's approach to metrics in Zendesk, whether you have or have no operational capacity to maintain complex dashboards and reports (tagging triggers if any, and so forth). One should first answer this question: if I need to have +1, +2, etc team members focusing solely on QA, or data moving forward to have a complete control on what's going on with the rest of the team performance, does this effort pay off? The answer may depend both on the scale of the product / service and consequently the volume of requests, and the size of the team and other factors. I believe there is a threshold, which I honestly find hard to define for the moment, not passing which would rather incentivize you as a support manager to stick to default time metrics, maximum + SLA feature in Zendesk and the default dashboards. 

1


Please sign in to leave a comment.

Didn't find what you're looking for?

New post