This article is only applicable to users of Onna Discovery and Onna Enterprise.
While the basic search bar searches across all indexed text fields, advanced search allows you to specify metadata fields and use boolean logic to search through the files you have synced to the system.
To begin using advanced search, select Advanced Search on the left hand menu as demonstrated below
You’ll be asked to name your search first. All advanced searches are saved so you will be able to access this search at a later date.
Next, select the metadata field you would like to search ('Select property' in the dropdown). Our standard options include:
- Application Name: available primarily for Microsoft products, you can use Application Name to specify the application that was used to create a document. E.g Microsoft Office Word
- Author: creator of document or sender of email
- Company: Extracted from the "company" field within Office documents.
- Date Created: Date that the document was created or received.
- Date Last Modified: Date that the document was last modified or sent.
- Extension: The file extension of the document, eg. pdf or docx.
- Extracted Text: The text of the document. For image files, documents are OCR'd to obtain text. Proximity search can be conducted through extracted text by searching: term1 ~[distance] term2; eg. "test ~3 search"
- MD5 Hash: Unique value calculated from signature of the document to identify duplicates.
- File Name: The name with which that file was saved.
The complete list of options for searchable metadata fields can be changed through the admin preferences. Below is an example of an organization that has enabled multiple metadata fields from groups that include Confluence, Slack, Jira and Mail:
Once you’ve selected the metadata field that you’re looking to search through you can select a condition to further specify your search. Our conditions include:
- Contains - field value contains specified search term
- Does Not Contain - field value does not contain specified search term
- Regular Expression (only available for the extracted text metadata field) - allows for wildcard searching
- Is Set - the metadata field is populated by a value
- Is Not Set - the metadata field is not populated by a value
Conditions are automatically linked with AND. You can add multiple conditions such as:
[Extracted Text] [Contains] Negotiation Agreement
[Extension] [Contains] PDF
This example would look for all PDFs that contain the word 'Negotiation Agreement' in their extracted text.
To add OR statements into your search, add a new group. Groups are conditions that can be linked by AND/OR.
Each time you add a new condition or group, you’ll see an example of your search displayed below. This is also the search that you’ll see displayed above your results when you run your search.
Below is an example of what the query looks like if you want to find all PDFs that contain the words 'Negotiation Agreement':
Once you've created your search parameters, there are a couple of options you can choose:
- Test search: If you want to get a quick feel on how many results are matching your terms, use this button on the left hand side. This will give you a quick figure beside the text indicating how many files match your search.
We can see the search for PDFs that contain Negotiation Agreement brought up 8 results.
- Save as new & apply: this creates a new saved search and runs the search which will display the results.
- Save & apply: this modifies the existing saved search and runs the search which will display the results.
Exporting an Advanced Search
Beside the search name, you'll see an Export icon in the shape of a box. Clicking on that will open a modal to take you through the export workflow:
Sharing an Advanced Search
If you would like to share your advanced searching with another use, you can click on the add user icon:
A new modal window will display and walk you through the steps of sharing a search with another user.
Deleting an Advanced Search
As you continue adding new advanced searches deleting older ones no longer used can help clean up the dropdown menu of saved searches. To delete a search click on the delete icon:
After clicking the delete icon you are first asked to confirm the deletion. Once confirmed the search is no longer available.
Wildcard Search Syntax
A Wildcard search can be used in the ‘metadata - extracted text’ field by selecting ‘Regular expression’. Placing the period and the asterix (.*) at the beginning, middle and/or end of a word will look for all results that include the specified term plus any letters for any number of times in place of the wildcard.
The search in the screenshot below for cont.*t in extracted text will match with files that contain ‘content, contract, contact’ etc.
The other wildcard search for .*ree will match with files that contain ‘agree, disagree, pedigree’ etc.
At least three other characters should be included for an optimal experience. Onna’s wildcard search looks for results on the active index and does not use a preset dictionary - the search may take longer but there will be more accurate results.
Advanced Search Best Practices
- Create templates for the most common types of searches: Every time you run an advanced search it gets saved. If you’re frequently running searches using the same metadata fields, consider creating templates that can be modified as needed.
- Use the ‘Test Search’ button at the bottom when you’re adding different conditions to see whether they’ve affected the number of results your search is hitting.
Can you change the name of a saved search once it’s been run?
Yes, by selecting the search you’ve already run from the dropdown and clicking the edit icon, you may rename the search. Clicking “Save & Apply” will rename your search and click “Save As New & Apply” will add an additional search with the new name and keep your original search.
Can you name advanced searches that are run from Excel?
Yes, Onna will automatically generate a name for you. You can then click edit to change the name and follow the above instructions.
Can you group conditions beyond one layer?
Currently, grouping conditions beyond one layer is not supported within the user interface directly. If you would like to group conditions beyond one layer of nesting, we recommend using the Advanced Search Template first and editing as needed.