Skip to main content

How to Use AI Candidate Match Scores

Automatically evaluate candidates against your job requirements using AI-powered criteria scoring. This feature analyzes each candidate's application data and shows what criteria they match or do not match.

Emilia Carvell avatar
Written by Emilia Carvell
Updated today

What are AI Candidate Match Scores?

Candidate Match Scoring is an AI feature that evaluates candidates against custom criteria you define for each job. The AI reviews candidate data (resume, application answers, etc.) and determines whether each candidate meets your specified requirements.


​

Key benefits

- Save time by automatically screening candidates against job requirements

- Get consistent, objective evaluations across all applicants

- See at-a-glance which candidates match your criteria

- Understand exactly why candidates did or didn't meet each criterion

Setting Up Criteria Matching


​Step 1: Open AI Fields

From your job's candidates view, click the AI Fields button in the toolbar.

*The AI Fields button is located next to Filters in the candidates toolbar*

Step 2: Create a New Criteria Score

Click + Create New, then select Criteria Score

![Create AI Field dialog showing Single Question and Criteria Score options] *Choose "Criteria Score" to evaluate candidates against multiple criteria*

Step 3: Configure Your Criteria

You'll see the configuration screen with several options:

AI Field Name: Give your criteria set a name (e.g., "Candidate Match", "Job Fit"). This will appear as a column header in your candidates table.


​Context: Select which candidate data the AI should analyze:

  • Application Answers

  • Resume

List Required Criteria: You have two options:

Option A: Extract from Job Description (Recommended)

Click Extract from Job Description and the AI will analyze your job posting to suggest relevant criteria automatically.


​

The AI suggests criteria based on your job description - select the ones you want to use


​

Review the suggested criteria, check the ones you want to include, and click Add Criteria in the bottom right of the drawer.

Option B: Add Criteria Manually

For each criterion, provide:

  • Criteria Name: A short label (e.g., "Technical Skills", "Experience Level")

  • AI Prompt Instructions for how the AI should evaluate this criterion

Tips for writing good prompts:

  • Be specific about what you're looking for

  • Reference the job description where relevant

  • Ask yes/no style questions (e.g., "Does the candidate have 3+ years of experience?")


Step 4: Test Your Criteria (Optional)

Before saving, you can test your criteria against a specific candidate to see how the AI evaluates them. Expand Test AI Instructions and select an application to preview results.

Step 5: Save

Click Save to create your Criteria Score field. The AI will begin evaluating all candidates in the background.

Viewing Criteria Results
​

Once configured, you'll see criteria results directly in your candidates table.

Candidates table with Criteria Match column showing scores and colored tags

Each candidate row shows:

  • Score: A ratio like "2/3" showing how many criteria the candidate met

  • Criteria Tags: Colour-coded indicators for each criterion

  • Green checkmark = criterion met

  • Red X = criterion not met

Click on a candidate's criteria score or any of the criteria tags to open the detailed breakdown drawer.

criteria Match breakdown drawer showing AI explanations

The drawer shows:

  • AI Score: Overall match ratio

  • Criteria Match: Each criterion with its result and the AI's explanation

  • Instructions: The prompts used to evaluate each criterion

  • Context: Which data sources were analyzed (e.g., "Answers and Resume")


​

The AI provides a detailed explanation for each criterion, helping you understand exactly why a candidate was marked as matching or not matching.

Best Practices


Writing Effective Criteria

1. Be specific Instead of "Good communication skills", try "Does the candidate demonstrate written communication experience through their cover letter or previous roles?"

2. Keep criteria measurable: Focus on objective requirements that can be verified from application data.
​

3. Limit to key requirements: 2-5 criteria works well. Too many criteria can dilute the signal.

4. Use job description language: Reference specific requirements from your posting.

Using Results Effectively

  • Filter by score: Use criteria scores alongside other filters to identify top candidates

  • Review explanations: The AI's reasoning helps you understand each evaluation

  • Provide feedback: Use the feedback link in the drawer to help improve the feature

Frequently Asked Questions


​

How long does it take for results to appear?

Results typically generate within a few minutes of saving your criteria. New applicants are evaluated automatically.


​

Can I edit criteria after creating them?

Yes, click the three-dot menu on your AI field to edit. Note that changing criteria will re-evaluate all candidates.


​

What data does the AI analyze?

Only the context sources you select (Resume and Answers). The AI doesn't access data outside your selection.


​

Are rejected candidates evaluated?

No, the AI skips candidates who have been rejected to conserve your AI quota.


​

Is there a limit to how many criteria I can add?

You need at least 2 criteria. There's no hard upper limit, but we recommend 2-5 for best results.

Need Help?

If you have questions about Criteria Matching or encounter any issues, contact our support team or use the feedback link in the criteria drawer to share your experience.

Related Articles
November 2025 New Features - AI Fields
Did this answer your question?