In addition to full text search, you can search using common search operators combined with data property keywords and values to narrow your results.
This article describes how Zendesk Support's advanced search works. It is aimed at administrators and managers with full access to the data in Zendesk Support. Use this article in conjunction with the Zendesk Support search reference, which contains tables listing and describing the full set of keywords and values you can use when searching tickets, users, and the like.
If you're an agent, or want to know more about basic search, start with Searching the data in Zendesk Support and refer back to this reference article if you want to perform more advanced searches.
Topics covered in this article:
Performing an advanced search
You can perform an advanced search from your Zendesk agent interface.
To perform an advanced search
- Click the Search icon () in the
upper-right of the top toolbar. You can also hover
your mouse over the +Add button and choose
Search from the drop-down menu.
Your most recent searches and viewed content appear in a menu. See Accessing your recent searches to learn more.
- Enter your search terms, using the terminology, operators, and formatting described in the sections below.
- When the search results appear, click the
Tickets, Users, Articles, or
Organizations tab to filter the results.
- If you currently have a ticket open and the similar tickets feature is turned on, filtering by Tickets shows you a Similar tickets header that includes any tickets related to the ticket you’re currently working on, if applicable.
Advanced search terms and terminology
Advanced search allows you to narrow your results by using data property keywords, operators, and search terms. In the context of an advanced search:
- Data property keywords indicate that you are restricting a search to one or more specific data properties.
- Operators are symbols used to modify the keywords and focus the search.
- Search terms are the words, phrases, or values you are searching for.
Search terms can include any string, including user, organization, or group names. For example, you can search for "vip" to list all users tagged as VIPs. However, the results will also include other data in your Zendesk Support instance that match the term "vip." For instance, if an organization has "vip" in its name, like "Midwest VIP Transportation" or "Viper Reptile Supplies," even if it does not have the VIP tag, it will appear in your search results.
The following is an example of a search string looking for anything with the tag "vip" that was created before May 1, 2015:
tags:vip created<2015-05-01
- tags is a keyword indicating you are searching only within a specific data property, in this case a tag.
- : is an operator indicating the tag property needs to match the subsequent search term. Note that there is no space before or after the :.
- vip is the search term referred to by the prior property/operator combination.
- created is a keyword indicating you are searching the created data property for items created relative to a certain date.
- < is an operator indicating you are searching for users, tickets, organizations, or articles created before a certain date.
- 2015-05-01 is a search term indicating the date you want to use.
Please upgrade my account
However, when you use a data property keyword multiple times in a query, there is an OR search across the values you specified. The following phrase returns results that contain either the tag "silver" or the tag "bronze".
tags:silver tags:bronze
To perform an AND search on multiple data property keywords, wrap the keywords in double quotes. The following phrase only returns results that contain both tags, "superman" and "is_awesome".
tags:"superman is_awesome"
Search operators
The following search operators can be used to build your search statements.
Operator | Description |
---|---|
: | The colon indicates that the given field should equal the specified
value.status:open |
< | Less than. status<closed |
> | Greater than. priority>normal |
<= | Less than or equal to. status<=pending |
>= | Greater than or equal to. priority>=normal |
" " | Double quotes. In a simple keyword search, this is referred to as a
phrase search and returns the exact words in the exact order; however,
punctuation characters are not
included."Please upgrade my account"
Note: In
the Japanese version of Support, this feature does not work as
expected. A simple keyword search that includes double quotes
returns results, but the results are not the exact words in the
exact order.
In a search including data properties, use double quotes to perform an inclusive AND search, returning results that include all properties in the search. tags:"superman is_awesome" |
- | Minus sign. Excludes items containing a word (or property value) from
the search results. For example, the following statement searches for
any tickets with the status 'pending', but excludes any tickets
containing the tag 'invoice' from the search results:
status:pending -tags:invoice |
* | The wildcard operator is useful when you want to search various forms
of a word. For example, searching for photo* returns
results that would include photography, photographer, photograph and any
other words that began with 'photo'. However, because of the
performance issues involved with doing wildcard searches,
unqualified wildcard searches are not currently supported. In other
words, you need to use a property keyword to make your search
specific to the data you're trying to locate.
subject:photo* |
Searching for properties that contain no data
none
as the search term, along with the
group, tags, via, organization,
or assignee keywords, as in this
example:assignee:none
This returns all unassigned tickets.
Searching by date and time
Date property keywords - created, updated, solved, and due date) can be combined with search operators to return data from a specific date, before a certain date, and after a certain date. To search dates in any locale, use the format YYYY-MM-DD. You can also use locale-specific formats such as MM/DD/YYYY in the United States.
created<2011-05-01
due_date>2010-01-10
solved:2010-01-10
You can also use the <= or >= operators to indicate less-than-or-equal-to and greater-than-or-equal-to respectively.
Searching with combined dates and times
created>2015-09-01T12:00:00-08:00 updated<2015-09-01T12:00:00Z
The first example above searches for anything created after September 1, 2015 at 12:00 p.m. (Pacific Standard Time).
The second example above searches for anything updated before September 1, 2015 at 12:00 p.m. (UTC).
Searching within a date/time range
created>2014-08-01 created<2014-08-05
You can also include specific times in your search range. The following example searches for anything created between August 1, 2014 at 11:59 p.m. (UTC) and August 2, 2014 at midnight (UTC):
created>2014-08-01T11:59:00Z created<2014-08-05T24:00:00Z
Searching with relative times
You can search for a time relative to the present time, using the time units hours, minutes, days, weeks, months, or years. The following search returns anything created in the last four hours:
created>4hours
Searching for tickets with a specific ticket form
You can search for a ticket form and get results for all tickets where that ticket form is applied. For example, the following search returns all tickets that use the Change Request ticket form:
form:"Change Request"
Sorting search results
You can sort your search results by field, in ascending or descending order, using the following keyword phrases:
-
order_by:field
-
sort:asc
orsort:desc
Sorting is available on the following fields:
created
commented
priority
status
ticket_type
Using the order_by
and sort
keywords is
equivalent to using the API parameters sort_by
and
sort_order
.
Using the 'type' keyword
For API searches, one of the tools you have available for narrowing your
search results is the type
keyword. It is
used to explicitly declare that you want to search for one of the
following types:
- ticket
- user
- organization
- group
- article (Help Center)
- entry or topic (forums)
Using the type
keyword means that you are explicitly
searching on the type you specify. For example, you can search for
all the users that belong to the customer's organization using this
search statement:
type:user organization:customers
If you instead searched for organization:customers
you
would also get all the tickets that have requesters who belong to
this organization. This is because searches that do not explicitly
specify type return results for all types, including tickets (and
organization is a ticket property).
Using type:user
, your search returns all users that
belong to the Customers organization. So, you're narrowing your
search to the user type and excluding tickets.
While organizations and groups are properties of the user object, they have their own properties that can be searched as well. The following query allows you search only for organization tags, excluding tags of the same name that may be used in other elements of Zendesk Support such as tickets and forum topics.
type:organization tags:premium
Suspended and verified users
You can search for users based on whether or not they are suspended, and whether or not they are verified. Users with any verified identity, not just an email address, are included in the search results. For example, users with verified phone numbers are listed.
To search for suspended users:
is_suspended:true
is_verified:true
Search FAQ
-
How soon can new data be searched?
When you add new data to Zendesk Support, it typically takes about a minute before it's indexed and can be searched. If you have the Zendesk Agent Workspace enabled on your account, you can also search live chat and messaging conversations, but it can take 10 to 12 minutes before a conversation is indexed.
-
How does punctuation affect search?
Punctuation characters are generally not included in searches.
-
Are there limitations to wildcard searches?
You can only do wildcard searches when combined with property keywords (
subject:photo*
). The wildcard must go at the end of the search term. -
Who can search what?
Administrators can search all the data in Zendesk Support. Agents can search the data that they've been granted access to. End-users can do full text searches of the knowledge base.
-
What languages are supported?
There is language-specific support for searching in the following languages:- Brazilian Portuguese
- Dutch
- English
- French
- German
- Italian
- Japanese
- Korean
- Russian
- Spanish
The support includes dictionary-based tokenization for Japanese, because words are not separated by spaces in that language. For the other languages, the language-specific support is primarily stemming, which allows different forms of the same word to match. In particular, the singular and plural forms of a word will generally match.