Search Report
This report lists the search queries made by the end users on the installed search clients. The report is segregated into four classifications and each classification covers four metrics.
| Search Classification | Metrics |
| All Searches | Query, Users, Sessions, Searches |
| Successful Searches (Clicks) | Query, Users, Sessions, Searches |
| Searches with no Click | Query, Users, Sessions, Searches |
| Searches with no Result | Query, Users, Sessions, Searches |
All Searches
All Searches is the first Search Classification. It displays a list of all the 50 latest queries looked up on a search client. To dig deeper, instance users can use pagination.
The table on the right has a searchable and three sortable columns:
| № | Column | Displays | Extra Functions |
| 1 | Query | Search queries. Example. At least one user searched react, analytics, content sources, and search client.
|
You can find a query in this column through the built-in search function.
|
| 2 | Users |
Number of unique devices (browsers) from which search queries occur.
Example: Four unique users searched react and 6 unique users searched search client.
|
The column is sortable. |
| 3 |
Sessions | Number of search sessions in which the queries occur. Example. Four unique users searched react in 8 different sessions and 6 unique users searched search client in 6 different sessions.
|
The column is sortable. |
|
You can find the ID of each session.
Example. content sources has been searched in 4 different sessions. To find the ID of each session, click 4. The dialog changes and you see a four-table column with the session IDs, the email of the user who searched the query (only for Internal users), and the time of search. The Session ID and Email columns are searchable. The Last Search column is sortable.
|
|||
| 4 | Searches |
Number of times search queries repeat.
Example. Four unique users searched react 10 times in 8 different sessions, six unique users searched search client 6 times in 6 different sessions, and only one user searched analytics 9 times in one session.
|
This column is sortable. |
|
You can find out if what facets and advanced search operators are used with a query by clicking Searches.
Example. analytics has been searched 9 times. Click 9 in Searches. A dialog opens where you can see the facets or advanced search operators applied.
|
|||
| 5 | Action (only for Searches with No Result) |
What has been done to address zero results queries.
Example. New content is required for python and someone has already made the changes to prevent Agent and product from being zero-result searches again.
The column is visible only when a search client has been specified in Select Search Client. |
The column has eight filters, All and the following seven:
|
|
From Example. To find queries that new content, select New Content. IMAGE NOTE. Filters are reset each time an SU Admin logs out, change the search client, or picks a new date range. Navigating inside the instance or refreshing the page through F5 has no impact on filter selection. |
). The default status. Each search query is zero results is set to Pending.
). An SU admin with access to the selected search client can change the status to In Progress when work has started on creating content or configuring tuning for the query.
). An SU admin with access to the selected search client can change the status to Update Content, when existing content needs editing to avoid zero results for the query again.
). An SU admin with access to the selected search client can change the status to New Content, when new content needs to be written to avoid zero results for the query again.
). An SU admin with access to the selected search client can change the status to Add Synonyms, when NLP Manager needs to be configured to avoid zero results for the query again.
). An SU admin with access to the selected search client can change the status to No Action Required, either the query return zero results because of a temporary hitch or nothing can be done to prevent zero results again. The latter can happen when a user looks up information on medicine on a video games website.
). An SU admin with access to the selected search client can change the status to Completed when the changes to prevent zero results for the query have been done.
, an SU admin can apply filters to find queries with a specific action taken. Successful Searches (Clicks)
List of queries which returned at least one result and at least one of the results is clicked
The table on the right has a searchable and three sortable columns:
| № | Column | Displays | Extra Functions |
| 1 | Query | Search queries. Example. At least one user searched react, analytics, content sources, and search client.
|
You can find a query in this column through the built-in search function.
|
| 2 | Users |
Number of unique devices (browsers) from which search queries occur.
Example: Four unique users searched react and 6 unique users searched search client.
|
The column is sortable. |
| 3 |
Sessions | Number of search sessions in which the queries occur. Example. Four unique users searched react in 8 different sessions and 6 unique users searched search client in 6 different sessions.
|
The column is sortable. |
|
You can find the ID of each session.
Example. content sources has been searched in 4 different sessions. To find the ID of each session, click 4. The dialog changes and you see a four-table column with the session IDs, the email of the user who searched the query (only for Internal users), and the time of search. The Session ID and Email columns are searchable. The Last Search column is sortable.
|
|||
| 4 | Searches |
Number of times search queries repeat.
Example. Four unique users searched react 10 times in 8 different sessions, six unique users searched search client 6 times in 6 different sessions, and only one user searched analytics 9 times in one session.
|
This column is sortable. |
|
You can find out if what facets and advanced search operators are used with a query by clicking Searches.
Example. analytics has been searched 9 times. Click 9 in Searches. A dialog opens where you can see the facets or advanced search operators applied.
|
|||
| 5 | Action (only for Searches with No Result) |
What has been done to address zero results queries.
Example. New content is required for python and someone has already made the changes to prevent Agent and product from being zero-result searches again.
The column is visible only when a search client has been specified in Select Search Client. |
The column has eight filters, All and the following seven:
|
|
From Example. To find queries that new content, select New Content. IMAGE NOTE. Filters are reset each time an SU Admin logs out, change the search client, or picks a new date range. Navigating inside the instance or refreshing the page through F5 has no impact on filter selection. |
Searches with No Click
Search queries for which results were found but no result was clicked. This data assists content managers identify documents whose titles and summaries can be rewritten to improve perceived relevancy.
The table on the right has a searchable and three sortable columns:
| № | Column | Displays | Extra Functions |
| 1 | Query | Search queries. Example. At least one user searched react, analytics, content sources, and search client.
|
You can find a query in this column through the built-in search function.
|
| 2 | Users |
Number of unique devices (browsers) from which search queries occur.
Example: Four unique users searched react and 6 unique users searched search client.
|
The column is sortable. |
| 3 |
Sessions | Number of search sessions in which the queries occur. Example. Four unique users searched react in 8 different sessions and 6 unique users searched search client in 6 different sessions.
|
The column is sortable. |
|
You can find the ID of each session.
Example. content sources has been searched in 4 different sessions. To find the ID of each session, click 4. The dialog changes and you see a four-table column with the session IDs, the email of the user who searched the query (only for Internal users), and the time of search. The Session ID and Email columns are searchable. The Last Search column is sortable.
|
|||
| 4 | Searches |
Number of times search queries repeat.
Example. Four unique users searched react 10 times in 8 different sessions, six unique users searched search client 6 times in 6 different sessions, and only one user searched analytics 9 times in one session.
|
This column is sortable. |
|
You can find out if what facets and advanced search operators are used with a query by clicking Searches.
Example. analytics has been searched 9 times. Click 9 in Searches. A dialog opens where you can see the facets or advanced search operators applied.
|
|||
| 5 | Action (only for Searches with No Result) |
What has been done to address zero results queries.
Example. New content is required for python and someone has already made the changes to prevent Agent and product from being zero-result searches again.
The column is visible only when a search client has been specified in Select Search Client. |
The column has eight filters, All and the following seven:
|
|
From Example. To find queries that new content, select New Content. IMAGE NOTE. Filters are reset each time an SU Admin logs out, change the search client, or picks a new date range. Navigating inside the instance or refreshing the page through F5 has no impact on filter selection. |
Searches with No Result
Search queries for which no result is produced. A search can be no-result for multiple reasons.
- Documents with the search keywords do not exist in your data repositories.
- The user does not have the permission to see documents.
- Your search client is not connected with the content source where the documents are stored.
It doesn’t end there. For every query where no result was found, a detailed user journey is provided when you click session ID. This helps you to deep dive into the filters that were passed, the advanced criteria that was used, or the keywords that finally end up producing no result on the search page.
The user journey captures: the activity type and details (like text searches, page views, search queries etc.), time of the activity performed, facets and values selected, and what kind of advanced queries are used (like Exact phrase, without words etc.). Using these details, you can explore more about the reasons why such searches are not generating results.
Download or Share
Check out Download and Share an Analytics Report
NOTE.
The downloaded or share reports for each classification are CSV files. Each CSV file consists of five columns: Search Query, Search ID, Facet Type, Facet Value, and #Searches, of which only Search ID is new. (#Search is the synonym of Count here.)
A user can apply multiple facets in a search query. Because each Facet Type and Facet Value occupies its own row, the Search ID helps readers immediately see which queries spread over multiple rows are essentially one. In the next image, row 4 and 5 refer to the same query because their Search ID is identical.
A report can contain data for up to 5,000 search queries.
Related
- Back to Overview
Last updated: Tuesday, November 9, 2021
Was this article helpful? Send us your review at help-feedback@searchunify.com