Use Smart Import and AI Suggestions in the Repository

This article will walk you through how to use AI Suggestions in Repository, including Smart Import.

 

AI Suggestions in Repository, including Smart Import, are AI-powered features that simplifies the process of tagging your existing and uploaded contracts by utilizing AI technology to analyze and extract data from them. Each document is also full-text searchable in the Ironclad Repository through our best-in-class OCR technology.

 

There are two ways to use AI to extract data from your contracts:

AI Suggestions on Uploaded Contracts (Smart Import)

You can upload executed contracts and other documents directly to the Repository. This is useful if you have historical contracts that you are migrating to Ironclad or if you have a contract that was executed outside of an Ironclad workflow.

You can upload multiple documents at a time using Smart Import. The following are the limits based on file source:

• Local Drive: 2000
• Dropbox: 2000
• OneDrive: 999
• Box: 1000

Each file must be 100mb or less and one of the following file types: .pdf, .docx, or .doc. Smart Import has a limit of 100 pages per file.

Currently, the AI has limitations in accurately processing certain elements from tables, handwriting, watermarked paper, low quality scans, and overlapping fields.

To learn more, refer to Upload Files into Ironclad Using Smart Import.

 

AI Suggestions on Existing Records in the Repository

You can run AI on your existing records in the Repository. This is especially useful for analyzing contracts you've already negotiated in Ironclad and evaluating how changes to your custom AI clauses or properties affect your contracts. You can select up to 2,500 documents at a time to run AI on. If a record was previously added to the Repository using metadata import, AI suggestions will not be gathered on the record. When the analysis is complete, you'll receive an email with the results in a CSV file.

New predictions may be generated when Ironclad updates its underlying model or when you make changes to AI Clauses or AI Properties. AI predictions will not be provided for records where data was manually added through metadata import or record editing, or where you've selected "don't show again"

 

---

 

AI Suggestions on Uploaded Contracts (Smart Import)

You can import files from your local machine or from cloud storage providers OneDrive, Dropbox, and Box.com. If you import files from a cloud storage provider, refer to the Upload from Cloud Storage Providers section below.

If importing from your local machine, gather your documents on the local machine that you use to upload your files into Ironclad.

We recommend grouping your uploads by contract type and setting the record type with each group. For example, upload NDAs together and set the record type to be NDAs. This will save you the effort of setting the record type for each individual record later.

The default record type is Imported. If you select Imported as the record type, all records will automatically retain their filenames as their record names. Otherwise, if the counterparty name is detected via Smart Import AI, the record name will be "[Record Type] with [Counterparty Name]."

1. Gather your documents on the local machine that you will be using to upload into Ironclad. Ensure that there are no duplicates. Repository Upload does not detect duplicates at this time.
2. Click Repository > Upload Records.
3. If you want to upload records from your device, drag and drop, paste, or click to browse your files. The more documents you upload, the longer the upload takes. An upload of 2,000 records may take a few hours to complete if you have a slower internet speed. The Ironclad application needs to be open during the entire upload process, so we recommend breaking up large uploads into smaller portions if you’re not sure if you will have an uninterrupted, stable internet connection.
4. If you want to upload records from your cloud storage provider, select the cloud storage provider from which you want to import. You will be prompted to login to your cloud storage account and select the desired files for import.
   NOTE You can import from multiple providers or locations within a single import.
5. In the Set record type dropdown, select your record type. This step is optional but recommended if you can batch your uploads by specific contract types. This will save you the effort of manually setting the record type for each record after upload.
6. Click Upload.

 

Access and Search Your Uploaded Files

A new record is created for each uploaded document. Smart Import runs each uploaded document through our AI model to suggest contract metadata to tag as properties. Each record displays AI-suggested values in green text. Smart Import only extracts metadata for PDF and DocX files (DocX files are converted to PDFs).

After the upload is complete, each new record becomes full-text searchable in the Repository. To search the Repository, click All Records on the left panel to view all of your records, including imported ones. You can also apply search filters based on AI-suggested property values.

The AI-suggested property values are displayed on the record. Users have the option to select a suggested value and add it to the record, enter their own custom value for the property, or ignore the suggestion.

 

---

 

AI Suggestions on Existing Records in the Repository

1. In the Repository, select the records you want to run AI predictions on. You can select up to 2,500 documents at a time or select a view containing up to 2,500 records for AI.
2. Click the three stacked dots located in the top right corner, and then click Run AI.
3. Click Run AI.
4. When the analysis is complete, you'll receive an email with the results in a CSV file.

 

---

 

Access and Search Your Updated Predictions

To access and search your updated AI predictions, click Updated predictions located on the left. After the process is complete, each record becomes full-text searchable in the Repository. To search the Repository, click All Records on the left panel to view all of your records, including imported ones. You can also apply search filters based on AI-suggested property values. The AI-suggested property values are displayed on the record. You have the option to select a suggested value and add it to the record, enter your own custom value for the property, or ignore the suggestion.

 

---

 

Verify Records (Optional)

AI suggestions appear automatically in the Repository as filterable, searchable fields that can also be used to create insights and dashboards without requiring verification. However, since AI predictions cannot be guaranteed to be perfectly accurate, we recommend verifying your records if your business requires complete accuracy.

The process is to verify the AI-suggested property values on each record and add any important data not captured by the AI. Lastly, if you didn’t already set the Record Type in the upload step, it is important to do so in the Verification step.

Unverified records display a small circle icon next to the Record Name and show AI suggestions in green. To verify records, open each unverified record and review the AI-suggested property values. If you haven't set the record type during upload, you must set it during verification. You can also add any additional properties you want to track that weren't detected by the AI. When you verify a record, all AI-suggested property values that weren't rejected are automatically accepted.

1. If you used Smart Import to upload files, an import is created with all of the newly created unverified records from the uploaded documents. This allows you to easily find a set of records and uploaded files at a later time.

   

2. If you ran AI suggestions on existing records, you can see all of your unverified records across all imports by using the Verification Status filter.

   

3. Click on the record you want to edit, and then click Edit.

4. If the record type doesn't exist, add it, then click 'Save' in the top panel.

5. Review each property suggestion:

   1. If an AI suggestion is correct, select it, click Save, then click Save in the top panel.
   2. If you prefer to never receive AI suggestions for this property on this contract, click Don't Show Again, then click Save in the top panel.
   3. If the AI suggestion is incorrect, manually enter the correct property, then click Save in the top panel.

6. Review each clause suggestion:

   1. If an AI suggestion is correct, select it, click Save, then click Save in the top panel.
   2. If you prefer to never receive AI suggestions for this clause on this contract, click Don't Show Again, then click Save in the top panel.
   3. If the AI suggestion is incorrect, click Skip, then click Save in the top panel. Next, go to the document, select the correct clause, tag it, and click Save in the top panel again.

      1. NOTE In the top right corner, you can toggle between selecting text or selecting an area on the document to tag a clause. For easier clause tagging, we recommend switching to Area Selection Mode.

         

 

---

 

Add Subsidiary Information

You can add subsidiary information as you verify your records. This teaches the AI to instantly recognize those subsidiaries and disregard them from the counterparty name. When you add a subsidiary during Smart Import, all predictions in the current import are updated and all future imports will also respect the learned rule. You can also add addresses to these subsidiaries to keep the counterparty addresses accurate.

We highly recommend adding any subsidiaries and their addresses to Ironclad before doing a Smart Import in order to teach our AI to ignore those values for Counterparty Name and Address predictions.

You can add and edit subsidiaries from both the Repository and Company Settings.

 

From the Repository

1. In the Repository, click on a Smart Import record you want to add a subsidiary to.

2. From the panel on the right, select Counterparty Name or Counterparty Address. Click Add to Subsidiary Information indicated by a gray circle with a plus sign.

   

3. Enter the subsidiary’s information.

4. To edit an existing subsidiary, click the Existing Subsidiary tab and in the Legal Name dropdown, select the subsidiary you want to edit.

5. Click Submit. The subsidiary is added to the import.

6. After you add subsidiary information, you can refresh your records by selecting the records you want to refresh, and then click on the three stacked dots located in the top right. Click Refresh Company Properties. AI will refresh the predictions on imported records to recognize the added subsidiaries and disregard them from the counterparty name.

 

From Company Settings

1. Click on your profile icon located in the top right corner of Ironclad. Navigate to Company Settings > Profile.
2. Click the Show Subsidiaries button located towards the bottom right, and then click Add Subsidiary.
3. Enter the subsidiary’s information and click Update Profile. The subsidiary is added.
4. After you add subsidiary information, you can refresh your records by selecting the records you want to refresh, and then click on the three stacked dots located in the top right. Click Refresh Company Properties. AI will refresh the predictions on imported records to recognize the added subsidiaries and disregard them from the counterparty name.

 

---

 

Rename Imports

Your Imports are named with the date and time that your file upload took place. To rename an import, hover over the import on the left column and click on the pencil icon.

 

---

 

Delete Imports

To delete an import, click on the three stacked dots located on the left hand side of the Repository of the open import. You have three options to delete an import:

1. You can delete the import, but keep all of the records in the import, including unverified records. The records will still be accessible in your Repository through text search and filters. You can see all unverified records in the Repository by using the Verification Status filter.
2. You can delete the import and all unverified records, but keep the verified records in the Repository. The verified records are already fully tagged with the metadata you approved so they are easily searchable in the Repository.
3. You can delete the entire import and all of the records in the import. This is useful if you make a mistake on an import and need to restart from nothing.

 

---

 

Clean Up Completed Imports

Once all of the records in the import are verified, the records are fully searchable in your Repository and you no longer need the import. There are two ways to clean up your completed imports:

1. Convert the import to a saved view. This enables you to easily find all of the records associated with the import at a later time.
2. Remove the import. Your records remain fully accessible and searchable in the Repository, but the import is not displayed.
3. Delete duplicates in the import. This deletes any duplicate records that were detected in the import.
4. Update with metadata. You can update the metadata of imported records by uploading a .csv, .xlsx, or .jsonl file with the new values. In order to match your data with records within Ironclad you'll need to include a column of Record Names or Record IDs you want updated. Make sure each Record Name or Record ID appears only once on your spreadsheet.

 

---

 

Use Smart Import API

You can use the Smart Import API to utilize Smart Import.

To learn more about how to use the Smart Import API, refer to the following resources:

• Upload a Smart Import Record to an Existing Import
• Create a Smart Import Record

 

---

 

Smart Import and AI Suggestion Errors

Sometimes you may see an error icon next to a record's name. Hovering over this icon reveals the error name and steps to resolve it. See the table below for all possible errors and suggested actions for the error.

Title	Suggested Action
Attachment Not Found	There was a problem retrieving the attachment for this document. Please try re-uploading it.
Unsupported File Format	Ironclad currently supports PDF, DOC, DOCX file types. Please convert the file to a supported format and try again.
PDF Conversion Error	Could not convert document to PDF. Please try re-uploading this document.
Unknown Error Occurred	Please try re-uploading this document or contact support.
Invalid company ID	Please try re-uploading this document or contact support.