Analyzing the metrics that matter to improve customer support

How many tickets are we solving?

The number of tickets solved gives you a quick indication of how your team is doing. It's most useful for examining overall trends, as some tickets are quickly solved while others take longer. Monitoring this metric over time helps you evaluate your team's performance and decide your resourcing needs.

Viewing prebuilt reports (Lite and Professional)

The Tickets tab of the Zendesk Support dashboard gives you an at-a-glance view of your solved, unsolved, created, and reopened tickets.

Additionally, the Assignee activity report on the Assignee activity tab of the Zendesk Support dashboard shows your top ticket solvers, along with a wealth of other information.

Creating your own reports (Professional)

The Tickets dataset contains the metrics and attributes you need to create your own reports about solved tickets.

The following Explore recipes give you some great starting points for creating your reports:

For a full list of recipes, see Explore recipes reference.

What to look for

As you look at your reports, focus on the following information:

The Solved tickets metric for both the whole team and individual agents.
The average values over a period of time (for example, a month). Some days, agents will be over or under the target.
The number of solved tickets compared with the number of open tickets. This shows how well the team is keeping up with the ticket queue.

How to use the results

The following table gives some examples of how you can use solved ticket reports to help improve your team's efficiency:

If you see this	Do this
Ticket volumes are consistently high.	Consider hiring more agents. Look for patterns of particularly busy or quiet times and staff accordingly. Consider more training for your agents. Look for common problem areas in tickets. Meet with product managers to see if the product could be improved. Consider setting up an About field (see What are the most common ticket areas?). Consider setting up a knowledge base to enable customers to serve themselves. A great knowledge base means that many of your customers will find the answer to their problem without ever opening a ticket.
Ticket volumes are consistently low.	You might be able to reduce your number of agents.
Some agents have low solve rates.	Check which tickets the agent has been working on. They could be particularly time-consuming problems. If an agent has a consistently low solve rate, they might require more product training.

What are the most common ticket areas?

Feedback from customers can help you improve your products and your customers' experiences. Take advantage of this feedback by categorizing your support requests into product areas. Then, you can easily see which product areas generate the most support issues.

To categorize your support requests, add a custom field to your tickets and support request form. This field can contain as many values as you want, and you can call the field whatever you want. In this example, the field is called Product and contains three values.

To see how Zendesk categorizes tickets, see The 'About' Field.

Viewing prebuilt reports (Lite and Professional)

Explore doesn't include custom fields on prebuilt dashboards. However, you can report on custom ticket fields by creating your own reports.

Creating your own reports (Professional)

The Tickets dataset includes each of your custom fields as an attribute. For more information, see Reporting with custom fields.

To create a report showing areas with the most tickets

In Explore, create a new report using the Support - Tickets dataset.
In the Metrics panel, select Tickets > Tickets.
In the Rows panel, select Ticket custom fields > customfieldname.
customfieldname is the name of the custom field you created to categorize your tickets.

The following Explore recipes give you some great starting points for creating your reports:

What to look for

As you look at your reports, focus on the following information:

The number of tickets in each area compared to the overall ticket volume.
The average time it takes to solve a ticket in each area.
The average CSAT rating for tickets in each area.

How to use the results

The following table gives some examples of how the category can help improve your customer support:

If you see this	Do this
One category generates significantly more tickets than others.	Work with product teams to identify whether the product area could be improved. Create a knowledge base with information to help customers use that area of the product.
Many tickets are not assigned to a category.	Train your agents to categorize tickets by selecting from the list of predefined categories. Make the category field mandatory, meaning a category must be chosen before you can proceed. Add the field to your support request form so your customers can choose the relevant category when they submit a support request.

How much work do we have?

Your ticket backlog provides a general pulse on the health of your support team. Ticket backlog is defined as all tickets currently in a new, open, pending, or on-hold state. In other words, it's your unsolved tickets that need work done on them.

This is an important metric to watch because it provides insight into your incoming ticket volume and how well you keep up, given your resources.

Viewing prebuilt reports (Lite and Professional)

The Backlog tab of the Zendesk Support dashboard includes prebuilt reports that measure your daily and weekly ticket backlog.

Creating your own reports (Professional)

The Backlog history dataset contains the metrics and attributes you need to create your own reports about your ticket backlog.

The following Explore recipes give you some great starting points for creating your reports:

What to look for

As you look at your reports, focus on the following information:

Ticket backlog volume cross-referenced with ticket age and first reply time. A large backlog isn't necessarily a bad thing if your team is capable of a high throughput.
Ticket priority cross-referenced with ticket status. A large backlog may or may not be bad, but one with many high-priority tickets probably is.
Historical backlog trends. For example, what are your busiest months? Which of your agents are performing the best?
Whether there are multiple support requests from the same customer, or in the same categories.

Remember that speed is important, but not at the expense of quality. Sometimes, support issues take longer to solve than the customer expects or longer than the performance targets you’ve set for your team. However, the longer it takes to solve a customer’s issue, the more likely customer satisfaction will suffer.

How to use the results

The following table gives some examples of how examining your ticket backlog can help improve your customer support:

If you see this	Do this
Your backlog is too big.	This means that customers are waiting longer for help. Your team can help by keeping the customer informed about the status and progress of their request. Are a lot of tickets in a pending state waiting for a customer response? Consider creating an automation to remind the customer of the pending status of their issue. If they don’t respond after a set number of days, you can automatically resolve the ticket to remove it from the backlog. For an example, see Zendesk on Zendesk: Bump Bump Solve. A backlog consists of both unassigned tickets and assigned tickets that have not yet been solved. The two combined equals your total ticket backlog. Try to ensure that tickets are assigned to someone as quickly as possible.
You have a lot of high-priority tickets in your backlog.	Consider creating a trigger that escalates high-priority tickets to make them visible more quickly. See Creating triggers for automatic ticket updates and notifications. Triage incoming tickets from customers to ensure they've set the priority field appropriately. Make sure your agents aren't setting the priority field incorrectly. If so, they might need further training in how to categorize requests.
You see multiple support requests from the same customer in the same category.	Engage directly with the customer to find out what's wrong.
You get multiple support requests from different customers for the same problem.	Consider writing a knowledge base article about the problem. Engage with the community in your knowledge base. Talk to your product development team. Could a product redesign help to stop problems?

How long do customers wait for a first reply?

First reply time (FRT) is the amount of time from when a ticket is created to when an agent makes the first reply to the customer. In other words, it measures how long it takes a real human, not an automated reply, to contact the customer. First reply time is a good indicator of the efficiency of your team and how well they hold up in handling fluctuating incoming ticket volumes.

Your performance target for this metric should align with your industry and customer expectations for first reply times. For example, you might use:

24 hours for support requests submitted using email and online forms
60 minutes for requests submitted using social media

Of course, exceeding these expectations is even better from the customer’s perspective.

Viewing prebuilt reports (Lite and Professional)

The Efficiency tab of the Zendesk Support dashboard includes prebuilt reports that measure your first reply time.

Creating your own reports (Professional)

The Tickets dataset contains the metrics and attributes you need to create your own reports about first reply time and efficiency.

The following Explore recipes give you some great starting points for creating your reports:

What to look for

As you look at your reports, focus on the following information:

Look at your ticket volume whenever your first reply time changes. A change might correspond with a temporary change in ticket volume. Start by looking at the Tickets and Efficiency tabs in the Zendesk Support dashboard. If you had a temporary surge from a new product launch or a major service incident, then this might not be a worrying trend. If you can predict these surges, that's a great time to bring a bit of extra help onboard.
If your first reply time is 24 hours on email-submitted requests and your median number of agent interactions is eight per ticket, then your average reply time median should be 4 hours or less. Looking at tickets with an average reply time greater than 4 hours should highlight opportunities for improving agent performance (such as training or adding self-service documentation to your knowledge base).
What are your customers telling you about your reply time? Look at your CSAT ratings and comments.
Monitor first reply time for each of your support channels to ensure that expectations are being met.
Monitor first reply time by product area if you're using a custom field as described earlier in this article (see What are the most common ticket areas?).
Monitor average reply time. This is the average amount of time for all replies to the customer when solving a support request.

How to use the results

The following table gives some examples of how examining your first response times can help improve your customer support:

If you see this	Do this
First reply times are high.	Look at when you get most of your tickets. An Explore report using the Tickets metric and the Ticket created - Hour attribute is a quick way to help make sure you have support online when your customers are. However quickly your team solves emailed tickets, sometimes live channels can be a faster way for customers to get help and for you to solve tickets. When something goes wrong, it's critical to find and address the issue as quickly as possible. Letting customers know you're aware and working on a problem is key to turning a bad situation into a good interaction with your company. If your customers serve themselves, then you don't need to worry about first reply times! Set up a knowledge base and write articles that address your top customer pain points. Stay present in your knowledge base and always answer questions from customers.

If you see this

Do this

First reply times are high.

Look at when you get most of your tickets. An Explore report using the Tickets metric and the Ticket created - Hour attribute is a quick way to help make sure you have support online when your customers are.
However quickly your team solves emailed tickets, sometimes live channels can be a faster way for customers to get help and for you to solve tickets.
When something goes wrong, it's critical to find and address the issue as quickly as possible. Letting customers know you're aware and working on a problem is key to turning a bad situation into a good interaction with your company.
If your customers serve themselves, then you don't need to worry about first reply times! Set up a knowledge base and write articles that address your top customer pain points. Stay present in your knowledge base and always answer questions from customers.

How long do tickets take to resolve?

Resolution time is important not just to you and your team. Customers also want their problems solved quickly. There are several key metrics that indicate how long it takes for agents to resolve issues:

Resolution time: The time it takes for a support issue to be solved. A ticket might get solved more than once because it can be reopened before it’s closed. The time it took to solve the issue the first time is called first resolution time.
Full resolution time: The time it took to finally solve the ticket.
Agent replies: The number of replies it takes for your agents to solve the ticket. The faster you can solve a ticket, the happier the customer will be.
One-touch resolution: Support issues that were resolved in a single interaction such as a call, ticket reply, or live chat. This has a positive impact on customer satisfaction and helps reduce support costs.

Viewing prebuilt reports (Lite and Professional)

The Efficiency tab of the Zendesk Support dashboard includes prebuilt reports that measure how long your tickets take to solve.

Creating your own reports (Professional)

The Tickets dataset contains the metrics and attributes you need to create your own reports about first reply time and efficiency.

The following Explore recipes give you some great starting points for creating your reports:

What to look for

Metrics that can help you understand the time and effort that went into solving a ticket include:

Handle time: The time that an agent spends working on a single support interaction. In Support, handle time is captured using the Time Tracking app. Explore can report on this data using the custom fields Total time spent, Avg time spent per ticket, and Avg time spent per update. For details, see Time Tracking app: metrics you need to be measuring.
Agent touches: An update that an agent makes to a ticket. This includes changes of ticket status, adding comments, and others. With Explore, you can report on agent touches using the Agent replies metric and Agent replies brackets attribute.
Requester wait time: The time a customer waits for their issue to be resolved. This is the time that a ticket spends in the New, Open, and On-hold statuses. During these periods, it’s the support team’s responsibility to resolve the issue. It's a good indicator of the customer’s experience of the support interaction.

How to use the results

The following table gives some examples of how examining your full resolution times can help improve your customer support:

If you see this	Do this
Tickets are taking too long to resolve.	This has a direct effect on customer satisfaction: People don’t want to wait a long time to have their problems resolved. However, focusing solely on resolution time isn’t necessarily a good idea. Speed doesn’t always equal quality. (For more, see How often are solved tickets reopened?) In addition to resolution time, finding the actual time that the agent spent resolving the issue is important. Total resolution time doesn’t do that because it includes stages in the ticket lifecycle when the agent isn’t working on it. For example, when it’s sent back to the customer for information and put in a Pending status.
You have a high proportion of one-touch resolution tickets.	This might indicate that you have lots of low-complexity support issues that could be better handled with self-service. Agents should work on more complex issues, not those that a customer can easily solve on their own by reading FAQs.
Tickets have a high proportion of agent touches.	This might indicate that the agent needs more training because it takes them more touches than the team average to solve tickets.

How often are solved tickets reopened?

A reopen occurs when a ticket's status is changed from Solved back to Open.

When tickets are reopened, this might indicate that agents aren’t fully solving the customer’s support issues. This might be because agents are concentrating on first-touch resolutions and speed over quality. It’s good practice to routinely monitor your team’s ticket reopens.

Viewing prebuilt reports (Lite and Professional)

The Tickets tab of the Zendesk Support dashboard contains information about reopened tickets.

Creating your own reports (Professional)

The Tickets dataset contains the metrics and attributes you need to create your own reports about reopened tickets.

What to look for

The total number of reopens
The percentage of tickets with reopens

How to use the results

The following table gives some examples of how examining your ticket reopens can help improve your customer support:

If you see this	Do this
The percentage of tickets being reopened is high.	Is this an escalated issue? Ticket reopens are more likely when agents deal with more complex support issues, so reopens might be higher for escalated tickets. Are your agents sufficiently trained? They might not have sufficient knowledge to solve more complex problems. Was there missing information in the ticket that caused the agent to misdiagnose the issue? Consider collecting more information in the customer ticket form if this is the case.

If you see this

Do this

The percentage of tickets being reopened is high.

Is this an escalated issue? Ticket reopens are more likely when agents deal with more complex support issues, so reopens might be higher for escalated tickets.
Are your agents sufficiently trained? They might not have sufficient knowledge to solve more complex problems.
Was there missing information in the ticket that caused the agent to misdiagnose the issue? Consider collecting more information in the customer ticket form if this is the case.

How satisfied are customers?

Every time you interact with a customer, consider measuring their satisfaction with a Customer Satisfaction Survey (CSAT), a short survey sent to the customer after their problem is solved.

Tracking CSAT helps you spot trends that might be affecting customer satisfaction. Satisfaction ratings are also key performance indicators for your agents because each rating received is associated with the agent who solved the customer’s problem. These can be averaged to give each agent an overall CSAT score.

At Zendesk, we simply ask if the interaction was good or bad. If customers want to give more feedback, they can add a comment.

Examples of metrics you can track include:

A customer’s CSAT rating over time
CSAT ratings by channel, product, or service
Average CSAT ratings for agents and teams

Viewing prebuilt reports (Lite and Professional)

The Satisfaction tab of the Zendesk Support dashboard includes reports that help you analyze your customer satisfaction scores.

Creating your own reports (Professional)

The Tickets dataset contains the metrics and attributes you need to create your own reports about customer satisfaction.

The following Explore recipes give you some great starting points for creating your reports:

What to look for

Compare ticket stats on tickets rated bad to those rated good.
Look for satisfaction ratings with comments.

How to use the results

The following table gives some examples of how examining your customer satisfaction can help improve your customer support:

If you see this	Do this
You get one or more bad ratings.	Notify managers when you get feedback. You can automate this using a trigger to notify a manager or group of managers when a ticket receives a bad rating. Use the following two trigger conditions: Ticket: Satisfaction > Changed to > Bad Ticket: Satisfaction > Changed to > Bad with comment Give your agents more information about your customers. Tickets with more touches often have a lower CSAT rating. Making more customer information available to your support teams with CRM integrations or custom user and organization fields can help your team know more about customers up front. If you can eliminate initial questions on tickets, that leads to faster resolution times and more one-touch ticket solves. Did your agent provide inaccurate information? Was there an issue with resolution time? Consider categorizing badly rated tickets based on the reasons why the ticket was rated the way it was. If it's purely about ticket handling, resolution time, or workflow issues, then those are some great projects to work on going forward.

If you see this

Do this

You get one or more bad ratings.

Notify managers when you get feedback. You can automate this using a trigger to notify a manager or group of managers when a ticket receives a bad rating. Use the following two trigger conditions:
- Ticket: Satisfaction > Changed to > Bad
- Ticket: Satisfaction > Changed to > Bad with comment
Give your agents more information about your customers. Tickets with more touches often have a lower CSAT rating. Making more customer information available to your support teams with CRM integrations or custom user and organization fields can help your team know more about customers up front. If you can eliminate initial questions on tickets, that leads to faster resolution times and more one-touch ticket solves.
Did your agent provide inaccurate information? Was there an issue with resolution time? Consider categorizing badly rated tickets based on the reasons why the ticket was rated the way it was. If it's purely about ticket handling, resolution time, or workflow issues, then those are some great projects to work on going forward.

More resources

By focusing on the metrics that matter, both your organization and your customers will benefit. All of the reports you've learned about in this article can be used to help you avoid ongoing issues.

If you're new to Explore and want to learn the basics, read Getting started with Zendesk Explore.

For some great ideas to help you get started with reporting, check out Explore recipes reference.

Finally, to see all of the Explore documentation, start with Zendesk Explore resources.

Join the conversation!

What tools do you use to help measure the success of your support organization? How have you managed to improve these metrics? Let us know in the comments below.