Skip to main content
Free-tools objection

Why Not Just Use ChatGPT or Free EDGAR?

An honest answer to the free-tools question: EDGAR is free and primary, chatbots are fast, and what ThesisCheck adds is the verification layer with receipts.

The question

EDGAR is free and ChatGPT can read filings, so I want to know what a paid thesis check adds before I spend anything.

What this answers

You can do this for free: EDGAR is the primary source, and a chatbot reads it quickly. The trade is time and audit discipline. ThesisCheck automates the verification layer: it splits your written thesis into claims, checks each against dated filings with receipts, forces a bear case, and marks unsupported claims as gaps. The public sample report shows exactly what that buys.

Direct answer

The free workflow is real, and the report links into it

EDGAR is free, primary, and authoritative, and ThesisCheck receipts point into the same public documents. The product does not sell access to sources; it sells the verification labor between your thesis and those sources.

  • EDGAR full-text search covers SEC filings at no cost and is the right starting point for any manual check.
  • A chatbot reads filings quickly, but a fluent answer can blend supported and unsupported claims unless every citation is audited by hand.
  • What you pay for is the audit: span-checked receipts, a forced bear case, a coverage audit, and explicit gaps.
Comparison

Manual EDGAR reading vs an automated verification layer

WorkflowFree workflow (EDGAR plus a chatbot)ThesisCheck
Cost

Free: EDGAR search plus a chatbot subscription you may already have.

One credit per check, CHF 49 pay as you go.

Time per thesis

Hours to find, date, and quote the relevant filings by hand with consistent discipline.

One run returns a claim-level ledger with receipts already attached.

Output

Notes you structure yourself; quality depends on your own audit habits under time pressure.

Span-checked receipts, a forced bear case, a coverage audit, and explicit evidence gaps.

Failure mode

A fluent chatbot answer can mix supported and unsupported claims without flagging which is which.

Unsupported claims are marked as gaps instead of being narrated away.

Do it by hand

How to replicate the check manually

The method is not a secret. If you keep this discipline for every thesis, the free workflow serves you well.

01
Search EDGAR per claim

Use EDGAR full-text search to find the filings relevant to each individual claim in your thesis, not just the ticker.

02
Build a source ledger

Record one row per claim with a dated source and a locator into the document, and keep quotes exact.

03
Force the bear case

Search for the filing-backed evidence that pushes against your thesis, even when conviction is high.

04
Log what you could not support

Write down every claim that found no supporting evidence. ThesisCheck automates these four steps and keeps the receipts inspectable.

Evidence boundary

What this comparison does and does not claim

Mixed2026-07-06

A careful manual EDGAR workflow and a ThesisCheck run produce the same kind of evidence; the product automates the discipline and keeps the receipts inspectable.

Evidence summary: The public sample shows the automated output side by side with its sources, so the free-versus-paid trade can be judged on the artifact itself.

ThesisCheck sample report · 2026-07-06 · Evidence boundary shown in the public sample
Inspect this in the sample report
FAQ

What am I actually paying for?

Verification labor, not data access. The credit buys the claim-splitting, the span checks against source text, the forced bear case, the coverage audit, and the gap accounting, delivered as one inspectable artifact. The public sample shows the full output before any purchase.

Sources

Public sources referenced for this comparison

Non-advice boundary. ThesisCheck provides descriptive, source-checked company research only. It does not provide buy, sell, hold, rating, sizing, or price-target recommendations.