Conga's Insights feature uses artificial intelligence to assess contract risks. Each extracted provision can be measured against predefined standards to create a risk score, and this score used to assess and mitigate contract risks for legacy or acquisition contracts (with Conga contract lifecycle management tools) or for drafting fresh agreements using X-Author for Contracts. When a contract is extracted, the AI identifies predefined clauses and analyzes language for salient field values or clause language. Once it identifies a passage containing such a value or term, the AI assigns it a risk score based on a predefined threshold. 

For example, an agreement may show a contract value of $1.5 million. On extraction, the AI first detects the passage identifying the total contract value. The AI then determines the stated value of the contract in this passage (a Total Contract Value field of $1,500,000) and compares it to a predefined risk threshold for total contract value. If the "high" risk threshold is set to $1,000,000, the user is warned that this contract carries a high risk. If so configured, the Insight feature may provide a suggestion for mitigating the risk, for example, to break the contract up into several lower-value agreements.

One might also assign risk based not on a field, but on the presence, absence, or nonconformity of a clause. For example, a contract may contain a jurisdiction clause that assigns the original contract to a state with an unfriendly legal climate. The jurisdiction clause setting may identify a set of jurisdictions with adverse legal rulings as high risk, other states as moderate risk (additional contract language may be needed), and other states/districts with friendly decisions as low risk. 

Each risk assessment is tied to a risk level assignment that enables stacked ranking of risks. 

To map Insights

1. Click Data Extraction Setup in the left navigation bar.
2. Click the name of the agreement type for which you are defining an insight to raise its Mapping window. 
3. Click the WORKSHEETS tab.
4. Assuming all fields and clauses are mapped for this contract, click Insights. If provisions remain to be mapped, it is a best practice to complete this task before defining Insights. 
5. Click the New button to map a new insight or click the pen icon on an existing definition you will edit. 

6. Select the type of insight. Your choices are Risk and Question. Choosing Risk indicates you want the AI to assign information it finds AI to a predefined risk class. Selecting Question enables you to ask predefined questions for the AI to interpret, with the answer appearing in the reviewer's results. For example, "What is the total contract value?" or "Which state court has original jurisdiction?" might yield "$150000" or "Delaware" in the Answer field.

7. For a risk, Enter a name, assign a risk scale score, assign triggering criteria, enter guidance for mitigating the risk, and define the provision in which the insight is to be found. You can program the tool to identify a risk using a Boolean assessment of comparative logic operations. Each condition compares the extracted data to preconfigured values, and acts when any, all, or custom conditions are met. For custom conditions you can apply Boolean AND or OR tests of multiple conditions: ( 1 AND ( 2 OR 3 ) ), e.g.

8. For a question, name the question in Question Title and enter a natural-language question in the Question field. You can also limit demands on the AI by selecting a provision from the Found in Provision field. 

   

   Tip

   You can use natural language to select non-standard operations. For example, "One month after the contract goes into effect" or "Does not contain an indemnification clause."

9. Click SAVE.

To map Insights with generative AI 

1. Follow the steps in To map Insights but activate the Gen AI toggle when prompted in the New Insight Mapping window.
2. Select the provision in which you expect to find the risk factor from the Found in Provision pull-down. 
3. Rather than entering criteria, enter natural-language prompts in the Identification Prompt and Mitigation Prompt text fields. The identification prompt reports the risk factor to both the AI engine and the user when a risk is identified. The mitigation prompt is offered to the user receiving the warning as a strategy for remedying the problem provision. 
   
4. Click Save.