❓ Help & FAQ

Everything you need to know about using Candidate Evaluator

🚀

Quick Start Guide

▼

Getting Started in 5 Steps

  1. Upload Job Description

    Go to "+ New Analysis" and upload your JD (PDF, DOCX, or TXT)

  2. Review Criteria

    The system extracts criteria automatically. Review and edit them on the "Job Criteria" page

  3. Upload Candidate Resumes

    Upload candidate Resumes/CVs (up to 250 per batch). Jobs with more than 75 resumes automatically process in the background, allowing you to close your browser and receive an email when complete.

  4. Run Analysis

    "Analyse Candidates" to score all candidates against the criteria

  5. Review Results & Export

    View rankings and AI insights, and export reports

💡

Best Practices for Large Batches

▼

Handling Large Resume Uploads

When screening many candidates (100+), we recommend a strategic approach that gives you faster results, better reliability, and more control over the hiring process.

🚀 Automatic Background Processing

For jobs with more than 75 resumes, the system automatically uses background processing:

  • No timeouts: Large batches process reliably without browser timeout issues
  • Close your browser: You don't need to keep the page open - your job keeps running
  • Email notification: You'll receive an email with a direct link when your analysis is ready
  • Check progress: View real-time progress on the "Background Jobs" page (appears in top menu when jobs are running)
  • Typical time: 3-5 minutes for batches of 75 to 100 resumes

💡 Tip: While background processing handles technical reliability, the Two-Pass Strategy below is still recommended for better decision-making and cost efficiency.

🎯 Recommended Strategy: Two-Pass Screening

Our elite recruiters use a Two-Pass Executive Strategy to maintain surgical precision at scale. By processing candidates in batches, you create high-velocity feedback loops that allow you to calibrate your criteria in real-time before finalizing your shortlist.

  1. Process in smaller batches (50-100 resumes each)

    Break your candidates into manageable groups. This provides faster processing, more reliable results, and the ability to spot issues early. Each batch typically completes in 1-3 minutes.

  2. Identify top potential candidates from each batch

    Review the results from each batch and identify the strongest 5-10 candidates. Look for:

    • High overall scores (typically 65%+)
    • Strong matches on your critical requirements
    • Candidates who stand out from the rest of their batch
  3. Run one final comparison with only top candidates

    Once you've identified the strongest candidates from all batches, create a final job with just these finalists. This gives you a direct head-to-head comparison of your best candidates, making it easier to make your final hiring decision.

📖 Example Scenario

Situation: You have 300 applications for a Software Engineer role.

Approach:

  • Batch 1: Upload first 100 resumes → Identify good potential candidates (e.g., 8 are worth further consideration)
  • Batch 2: Use "Run New Batch with Same Criteria" button from Batch 1 results → Upload next 100 resumes → Identify promising candidates (e.g., 7 strong matches)
  • Batch 3: Use "Run New Batch with Same Criteria" button again → Upload final 100 resumes → Identify top candidates (e.g., 10 standouts)
  • Final Round: Upload just those 25 finalists → Run detailed comparison

Result: You've efficiently narrowed 300 candidates down to your top 25, then performed a detailed analysis on just the finalists who truly have potential.

💡 Key Feature: The "Run New Batch with Same Criteria" button (found on any historical job's results page) loads that job's exact criteria settings into your workspace. This ensures all batches are evaluated identically, making results directly comparable. Remember to update the job title (e.g., add "Batch 2", "Batch 3") to keep your batches organized in Job History.

Processing Applications as They Arrive

This same approach works brilliantly if you want to process applications progressively rather than waiting for all resumes to arrive. Use the "Run New Batch with Same Criteria" feature to ensure consistent evaluation:

  • Week 1: Process your first 80 applications, get early insights into candidate quality
  • Week 2: Load Week 1 job using "Run New Batch with Same Criteria" → Process the next batch that arrived, identify any standout candidates
  • Week 3: Load again with same criteria → Process final batch after closing date
  • Final: Compare only the top candidates from each week

Note: Using the batch feature ensures all weeks are evaluated with identical criteria, making scores directly comparable across time periods.

💡 Pro Tips

  • Large batches (75+ resumes): Run automatically in the background - check the "Background Jobs" menu item (appears when jobs are active) or wait for your email notification
  • Use the batch feature: Click "Run New Batch with Same Criteria" on any historical job to load its exact criteria for new candidates—ensures fair comparison across batches
  • Early insights: Processing batches early gives you a "sneak peek" at application quality before all resumes arrive
  • Adjust criteria: If early batches show unexpectedly low scores, you can refine your criteria before processing remaining candidates
  • Reduce costs: Only pay for detailed AI insights on your finalists, not all 300 candidates
  • Stay organized: Name your jobs clearly (e.g., "Software Engineer - Batch 1", "Software Engineer - Batch 2", "Software Engineer - Finalists")

Why Not Process All at Once?

While the system can handle larger batches (batches over 75 resumes automatically use background processing, so you won't experience timeouts), breaking them up still provides:

  • Strategic decision-making: Early batches reveal application quality before processing everyone
  • Progressive insights: Start reviewing top candidates while others are being processed
  • Easier management: Simpler to spot and fix any issues with specific candidates
  • Lower costs: Focus expensive AI insights on only the candidates who matter
  • Criteria refinement: Adjust your criteria based on early results before processing remaining candidates
📊

Scoring Explained

▼

How are the scores calculated?

Our AI evaluates candidates based on the depth of evidence provided in their resume for each criterion. It doesn't just look for keywords; it analyzes professional context and career impact.

🎯 The Three Pillars of a High Score

To achieve a "Strong" score (80%–100%), the AI looks for three specific markers:

  1. Skill Mastery: Does the candidate possess the required technical or soft skill?
  2. Professional Context: Has the candidate applied this skill in a recent professional role (e.g., as a CFO or Manager)?
  3. Proven Outcome: Does the resume cite a specific result, achievement, or metric (e.g., "Led budget preparation for a $10M portfolio")?

The more pillars present, the higher the score. A score of 60% might indicate the skill is listed but lacks professional context or specific outcomes.

Quick Reference Ranges

✅ Strong (80%–100%)
The candidate provides clear evidence of the skill, often backed by specific job titles, companies, and measurable results.
⚠️ Moderate (40%–79%)
The skill is mentioned, but the resume may lack recent context or specific details on how it was applied.
⛔ Weak (0%–39%)
There is little to no mention of the skill or relevant professional experience in the provided document.

Overall Score: This is a weighted average of all criteria scores, giving you an at-a-glance measure of candidate fit.

💡 The AI Co-Pilot

While our AI identifies evidence with expert-level precision, these scores are designed to be your 'Co-Pilot.' They highlight the strongest matches so you can apply your human expertise where it matters most: the interview.

Examples

Example 1 (Score: 100% - Strong)

Criterion: "Financial Modeling in Excel"

Resume: "Built complex multi-office budget models in Excel as CFO at SLF Lawyers, resulting in a 15% efficiency gain."

→ Score 100%

Why: Mentions the skill, a senior professional role (Anchor), and a clear outcome.

Example 2 (Score: 60% - Moderate)

Criterion: "Financial Modeling in Excel"

Resume: "Proficient in Microsoft Excel and financial analysis."

→ Score 60%

Why: The skill is mentioned, but lacks a professional anchor or a specific outcome.

How to use the scores

  • Treat scores as evidence strength indicators, not pass/fail thresholds
  • Use the Coverage Matrix to spot strengths and gaps across all candidates
  • Use Deep Insights to see Professional Anchors (job titles + companies) behind each strength
  • Review AI-generated Interview Questions to probe areas with moderate scores
🔒

Privacy & Data Handling

▼

🔒 Enterprise-Grade Security

Your data is processed in an isolated AI environment. Files are never used for public model training, ensuring total data sovereignty and zero cross-contamination.

View our Security & Data Privacy Standards →

How your data is handled

This application uses OpenAI's GPT-4o model for intelligent analysis of job descriptions and candidate resumes.

✅ Your data is NOT used to train AI models

OpenAI does not use data submitted via their API to train or improve their models (unless you explicitly opt in). We have NOT opted in to any training programs.

✅ Limited data retention

API data is retained by OpenAI for up to 30 days for abuse and misuse monitoring. After 30 days, your data is automatically deleted from OpenAI's servers.

✅ Secure transmission

All data is transmitted securely via HTTPS to OpenAI's API endpoints. No data is stored on external servers by this application.

⚠️ What this means for you

  • Job descriptions and resumes are temporarily processed by OpenAI's systems. They are not permanently stored by OpenAI and they are never used to train its AI models.
  • We have strict controls in place to ensure the privacy of all your data, but if you are unsure about your organization's requirements you should consult your compliance team.

📚 Learn more

💬

Frequently Asked Questions

▼

How does Candidate Evaluator work?

Upload your job description and candidate CVs. The app scores each candidate against extracted criteria, then (optionally) generates AI-powered insights for your top picks.

How much does it cost?

Candidate Evaluator uses a flexible credit-based model. Your first job is included as part of our Test Drive offer. For detailed pricing on Standard Jobs and bulk Credit Bonuses, please visit our Pricing Page.

How accurate is the scoring?

Our scoring engine uses human-level reasoning to understand context beyond simple keywords. While it is an incredibly powerful screening tool for identifying top-tier talent, we always recommend using the AI-powered Evidence Explorer to verify specific claims during your final review.

Can I edit criteria?

Yes! Visit the "Job Criteria" page to review and edit extracted criteria. You can add, remove, or modify criteria before running analysis.

What file formats are supported?

Job Descriptions: PDF, DOCX, DOC, TXT
Resumes: PDF, DOCX, DOC, TXT
All files are converted to text for analysis.

How do I export results?

After running analysis, click "Export Reports" to download executive summaries, coverage matrices (Excel/CSV), and individual candidate reports (PDF/Word).

Can I revisit past analyses?

Yes! All analyses are saved to your account. Visit the "History" page to view and reopen past jobs.

Why did a candidate score 60% if they have the skill listed on their resume?

A score of 60% indicates a "Moderate Match." This usually occurs when a skill is listed (e.g., in a "Skills" section) but isn't supported by detailed descriptions of projects or outcomes in their "Experience" section. For example, a candidate listing "Advanced Excel" may score lower than one who describes "using Advanced Excel to manage multi-office budgets at Wayne Enterprises".

What are "Professional Anchors"?

Our Deep Insights use "Professional Anchors" to link strengths directly to a candidate's history. This means every claim we make is tied to a specific job title and company found in their resume, ensuring the analysis is grounded in fact rather than generic AI summaries.

Can a candidate's score be improved?

Scores are based strictly on the provided resume. If a candidate has a lower score due to a brief resume, we recommend using our AI-generated Interview Questions to dig deeper into those specific areas during your screening call.

Where can I find the suggested Interview Questions?

Suggested interview questions are integrated directly into your candidate reports.

For High-Scoring Criteria: In the Deep Insights section, the AI may suggest "Discovery Questions" to help you verify the specific achievements mentioned in the resume.

For Low-Scoring Criteria: In the Detailed Scores section, if a candidate scores 0% due to missing evidence, the AI often provides a "Targeted Question". This allows you to quickly ask the candidate about that specific skill during a screening call to see if it was simply omitted from their resume.

How should I use these questions? Think of these questions as your "AI Co-Pilot" for the first interview. They are designed to help you dig past generic bullet points and find the Professional Anchors—the real-world companies and outcomes—that prove a candidate can do the job.

What happens to my data after I log out?

Your data is processed in a Siloed AI Environment. While OpenAI retains API request data for 30 days for integrity monitoring, it is never used for training. Your data belongs to you, and stays in your account until you choose to delete it.

🔧

Troubleshooting

▼

Analysis not running

  • Ensure you've uploaded both a Job Description and at least one candidate resume
  • Check that criteria have been extracted (visible on Job Criteria page)
  • Verify you have sufficient account balance

PDF text extraction issues

  • Some PDFs (especially scanned documents) may not extract text properly
  • Try re-saving the PDF or converting it to DOCX format
  • Very complex multi-column layouts may extract better as DOCX

Scores seem too low/high

  • Review criteria on Job Criteria page - overly specific criteria may score lower
  • Check the Evidence Explorer to see what text is being matched
  • Criteria Calibration: If scores seem lower than expected, your criteria might be overly specific. Try a 'Broad Pass' to see a wider range of talent, then use Deep Dive Insights to refine your top finalists.
📧

Contact & Support

▼

Need more help?

If you have questions not covered in this FAQ, please contact:

Email: [email protected]

About This Application

Version: 2.0 (January 2026)

Technology:

  • AI Architecture: OpenAI GPT
  • Framework: Flask (Python)

💡 Pro-Tip: You Haven't Missed Anyone!

Remember, we analyze every single resume against every single criterion, regardless of the tier you choose. The 'Standard' tier provides a Full Ranking of your entire candidate pool using our Intelligent Filtering logic. The 'Deep Insights' are simply the 'Deep Dives' we've prepared to help you get ready for your first round of interviews. You can always unlock more insights for any candidate later!