The first step in analyzing ticket data to determine content needs for your KB is to categorize your tickets.

Your ticket categories depend on the data you're collecting in your tickets. Take a look at the custom fields you've set up to collect extra information. Hopefully you're capturing some data about your customers, product, or issues that will help you. The best case scenario is that you have data about the major issue in each ticket so that you can easily categorize tickets according to issue.

For example, if you have multiple products (Game ABC, Game XYZ, and so on), or different user segments (novice, professional, developer, and so on), do you have a custom field for those items? Or do you have a custom field for selecting the feature related to the problem, such as "Drawing tool" or "Exporting?"

At Zendesk, we have a custom field calledAbout包含普通票Multibran话题”ding" or "Salesforce." Agents use this field to indicate the issue for each ticket. We also have a custom field calledHandle Timeto track the amount of time an agent spends handling a ticket. By comparing About and Handle Time, we identify topics that require a lot of time for our agents to handle. Then we can drill down to find common issues that we can document in our KB.

Your ticket volume and the number of issues you have will determine how long you need to wait before you have a statistically significant distribution of issues. Collect 500 tickets with the new custom field, then take a look. If some of the categories seem light (fewer than 30 tickets), you might want to wait a little longer, until you have 500-1000 tickets maybe. Also, consider whether you really need those light categories, (you might rethink the options in your custom field.)

If you are not currently capturing data that you can use to categorize tickets, set up a custom field where agents can indicate the issue for each ticket (seeAdding custom fields to your tickets and support request forms). Then you can build anInsights report for your ticket categories based on a custom multi-select "about" field using ticket tags.

After you categorize your tickets, you can drill-down into each major category to find specific, common issues. To do this, you cancreate a view of each category. If you have a lot of categories, you might want to start with the ones with the most tickets.

You might be able to find recurring issues by scanning ticket subject lines, but you'll probably have to dig in and read through tickets to pinpoint specifics. This will be a manual process.

To create a view for your ticket categories

1. InAdmin Center, clickWorkspacesin the sidebar, then selectAgent tools > Views.
2. ClickAdd view.
3. InMeet all of the following conditions, add conditions to show tickets tagged with a specific issue.
   • SelectTagsandcontains at least one of the following, then enter one of the tags associated with an issue in your About field.
   • Click thePlus(+) to add another condition, then selectStatusandless than Closed.
     Note:If you also want to review closed tickets, you can create another view. There is no way to see all tickets of any status in the same view.

     

4. SelectPreview match for the conditions aboveto test your conditions.
5. SetFormatting options.
6. In theAvailable forsection, selectMe onlyto make this a personal view.
7. ClickCreate view.

Now you are ready to dig in and find specific issues in this grouping that might need to be documented in your KB.

Establish a system for tracking the recurring issues you find during your ticket analysis.

A google doc or a spreadsheet that you share with others who might helping with the research, will probably work well. Later, you can use this document as a tracking sheet that you can update as you add items to your knowledge base. And you can also use it to make assignments where content needs to be developed.

As you drill-down into tickets in the view for a particular issue, you might find that some issues you thought were common are not actually common enough to document. And maybe some recurring issues don't lend themselves to documentation because there are too many variables in customer's environment. Your spreadsheet will evolve throughout your analysis phase as you add and remove issues as needed.

如果你使用宏,共同应对门票,you might want to take a look at your most used macros. You will probably find some common customer pain points in those macros. And you might be able to reduce tickets, and specifically tickets that get these popular macros, if you document the topics covered in those macros in your knowledge base.

The tags on tickets can also help you categorize tickets and find common issues. We've already talked about analyzing tags set by drop-down custom fields. You can also use other tags to help pinpoint common issues.

Take a look at your tag cloud to see the 100 tags that have been most applied to tickets in the last 30 days. This might give you some ideas about different groupings or issues where you can drill-down.

Or, better yet, if you have a system for manually tagging items that need documentation, then you can review those to find important items to document. But that's probably something you willstartdoing, but haven't been doing yet!

At Zendesk, when we set up our internal knowledge base, we relied on the tags set by our custom fields. But we plan to start manually a tag for doc_needed as another way to surface topics that need to be documented.

You should not overlook any existing content. This can be another source for determining issues and seeding your knowledge base. And in most cases, it's ready-made content that can go directly in the knowledge base with little or no editing.

Consider the following:

Be sure to listen to your internal community of agents and your external community of customers. It's a great way to directly learn about common customer issues and pain points.