Article Summary
AutoRFP's Trust Scoring System categorizes generated responses into Trust Rankings (Exact Match, Near Match, High Trust, Low Trust, No Results Found) to indicate confidence levels. Use these rankings with Feedback scores to prioritize review efforts and continuously improve response quality.
Prerequisites
Active project with generated responses
Editor or Reviewer role assigned
Step-by-Step Instructions
Step 1: Understand Trust Rankings
Each response in your project displays one of the following Trust Rankings:
Near Match (99%+)
Indicates a near-exact match to a previous requirement
Uses a previous response with minimal differences
Very high confidence - quick review recommended
Exact Match (100%)
Indicates an exact match to a previous requirement
Uses a previous response verbatim
Highest confidence level—minimal review needed
High Trust
Response generated from highly similar sources
Suggests high reliability
Standard review recommended
Low Trust
AI found a response but has low confidence in accuracy
Requires thorough review and likely editing
May need additional source material
No Results Found
No relevant response found in content library
Displays icon with empty response field
Requires manual response creation using the response editor
Human Edited
A user has edited the AI-generated response
Trust score hidden after human modification
Indicates human judgment applied
Approved by Human
Response approved by a reviewer
Marked as reliable
Can be reused with confidence in future projects
Step 2: Sort Responses by Trust Score
Navigate to the Project Editor.
Click the Trust Score column header to sort responses.
Sort from highest to lowest trust to review most reliable responses first.
OR sort from lowest to highest to prioritize responses needing the most attention.
💡 Tip: Start with High Trust and Exact/Near Match responses for quick approvals, then focus editing time on Low Trust responses.
Step 3: Review AI Feedback Scores
Locate the AI Feedback score in the bottom right corner of each response.
Review the grade assigned (as a percentage).
If the score indicates room for improvement, click to view detailed feedback.
Read context and suggestions explaining why the score isn't higher.
What AI Feedback Evaluates:
Completeness of the answer
Response relevance to the requirement
Clarity and professionalism
Appropriate length and detail
Step 4: Use Scores to Prioritize Review Workflow
Efficient Review Strategy:
Filter or sort by Trust Score to group similar confidence levels.
Quickly approve Exact Match and Near Match responses with high AI Feedback scores.
Review and lightly edit High Trust responses as needed.
Allocate more time to Low Trust responses.
Manually create responses for No Results Found requirements.
Step 5: Improve and Approve Responses
Edit responses based on AI Feedback suggestions.
Address specific improvement areas mentioned in feedback.
Ensure response fully answers the requirement.
Submit for review once editing is complete.
Approve responses that meet quality standards.
Important: Each improved and approved response strengthens your content library, leading to higher Trust Scores and better first-draft quality in future projects.
💡Tips & Best Practices
Prioritize approving well-written responses to build library strength
Every approved response improves future Trust Scores
Consistently high-quality approvals lead to more Exact and Near Matches over time
Assign Low Trust responses to subject matter experts
✋🏼 Common Mistakes to Avoid
Dismissing Human Edited responses as automatically good—human edits still need review
Ignoring AI Feedback on High Trust responses—small improvements compound over time
Need Help?
💬 Live Chat: Available in-app
📧 Email: [email protected] or contact your Success Manager directly for urgent support.
📚 Learning Centre: learn.autorfp.ai/en