Why is some information not extracted into the search engine from documents while indexing? – RChilli HelpDesk

When some information is not extracted into the Search Engine while indexing documents (resumes or JDs) using RChilli, it typically points to one of the following causes:

1. Parsing Is Not Performed Before Indexing

If you are indexing without parsing, the Search Engine won't have structured data like skills, experience, or education to index.

Solution: Use the ParseAndIndex API, which ensures the document is first parsed, and only then indexed with structured fields like skills, education, experience, etc.
Parse and Index API Documentation

2. Incomplete or Improperly Structured JSON

If you're submitting already-structured JSON data via the IndexDocument API, but it's missing key fields:

Those fields won’t appear in the Search Engine.
RChilli expects JSON to follow its specific schema.
Solution: Use the Resume Parser or JD Parser first to extract complete and accurate data. If using your own structured data, ensure it matches RChilli's schema.

3. Fields Not Enabled in API Settings

By default, some optional fields may be disabled in the parsing output, such as:

Detailed education or experience sections
Taxonomy alias fields
Certifications, publications, or achievements
Solution: Use dynamic API settings (in the apisetting node) to enable additional fields during parsing. For example:

"apisetting": {
"reqeducationdrill": true,
"reqexperiencedrill": true,
"reqskillsdrill": true
}

Dynamic API Settings Guide

4. Using the Wrong Environment (Demo vs. Live)

If you're viewing the search output in MyAccount, it may default to demo data.

Solution: Click "Use Your Key" in the MyAccount portal or call the API directly with your own userkey. This ensures you’re indexing and searching your real data, not sample data.

5. Preprocessing or File Quality Issues

Some documents may contain:

Poor formatting (e.g., scanned PDFs, image-based resumes)
Hidden text or unreadable fonts
Non-English content not supported in your parser settings
Solution:
- Ensure input documents are digitally readable, not image-based unless OCR is enabled.
- Consider enabling LLM Parser or PDF layout model (parsethroughpdflayout: true) for improved parsing of tricky layouts.

Summary Table: Common Causes & Fixes

Problem	Root Cause	Suggested Fix
Missing skills, education	No parsing or partial parsing	Use `ParseAndIndex` API
JSON has missing fields	Custom JSON not matching schema	Use RChilli Parser or validate schema
Extra fields not included	API settings not enabled	Set `apisetting` flags like `reqeducationdrill`, `reqskillsdrill`
Only sample data shows	Demo environment used	Use “Use Your Key” or real credentials
Poor quality files	Image-based, scanned, or noisy formatting	Use LLM Parser or OCR-enhanced parsing

Need Help Diagnosing a Specific Case?

Email support@rchilli.com for tailored help.