Skip to content
InvestingBuddy
Sign in

Evidence-first research workspace

Evidence-first
investment research.

Discover companies, extract the facts from their own primary documents, challenge the thesis with several research agents, and turn scattered evidence into an auditable research report — one you review yourself before it means anything.

Analyze a companyDiscover opportunitiesView the research workflow

No ratings, no price targets, no return projections. The system assembles and challenges evidence; the investment judgement stays with you.

Two ways in

Start from a company, or start from an idea.

Both paths end in the same place: a cited, period-aware research report with its evidence attached.

Path A

Analyze a company

You already know which company you want to understand. Give it a ticker and the pipeline does the collecting, extracting and challenging.

  1. 01Company identity resolved from ticker and exchange
  2. 02Primary documents located on the issuer's own domains
  3. 03Latest annual and latest current-period reporting retrieved
  4. 04Financial facts extracted, with period, scope and currency
  5. 05Multi-year series reconstructed where the data supports it
  6. 06Regulated disclosures pulled from official venues
  7. 07Evidence pack assembled and cited
  8. 08Research agents analyse it from separate angles
  9. 09Red team argues against the emerging view
  10. 10Structured research report assembled
  11. 11You review the evidence and decide
Analyze a company

Path B

Discover opportunities

You have a theme, a market or a hunch. Describe it in your own words and the system builds a bounded universe to work through.

  • “European luxury companies”
  • “Small-cap European industrial automation”
  • “Nordic businesses exposed to data-centre investment”
  • “European companies benefiting from grid modernisation”
  1. 01Your description is parsed into theme, region, sector and market
  2. 02A bounded universe is generated from an auditable registry
  3. 03Candidates are screened and ranked, each with the reason it surfaced
  4. 04You pick the ones worth a full deep dive
  5. 05Full company research runs on the ones you choose
  6. 06Research packages stay comparable across candidates
Start discovery

How it works

Seven stages, and you can see inside each one.

Research is a pipeline, not a prompt. Every stage records what it found, what it could not find, and where each figure came from.

Turn a written research idea into a bounded universe of companies worth looking at.

  • Theme, region, sector and market parsed from your own wording
  • Universe generated from a curated, auditable company registry
  • Candidates ranked with the reason each one surfaced
  • Companies that were excluded — and why — stay visible

Why it is built this way

Design decisions you can check, not claims you have to trust.

Primary evidence first

Research starts from issuer documents and regulated disclosures, not from a model's recollection of a company.

Period-aware financials

Annual, half-year and quarterly figures are kept apart, and so are Group and segment. A part-year figure is never annualised to look comparable.

Multi-agent challenge

A thesis is examined by several agents with different jobs — including one whose job is to argue against it — rather than produced by a single assistant.

Auditable by design

Every figure keeps the document, page, period, scope and currency it came from, so a claim can be traced back to its source.

Honest about gaps

What could not be sourced stays visible as missing. Conflicting figures are surfaced rather than quietly reconciled.

You keep the judgement

The system does the research. It produces no rating, price target or return projection, and every report requires a human to review it.

Who it is for

The same workflow, read from your seat.

Research more companies without spending the evening collecting filings.

  1. Find the company
  2. Collect primary evidence
  3. Read current and historical financials
  4. Have the thesis challenged
  5. Review the report

A structured research workflow, with far less of the repetitive gathering and re-typing that usually consumes it.

Against the manual workflow

The gathering still happens. It just stops disappearing.

Doing it by hand

  1. Search
  2. Investor-relations site
  3. Download PDFs
  4. Copy figures into a spreadsheet
  5. Scattered notes
  6. News search
  7. Ask an AI chat
  8. Write the memo

The evidence exists, but only in your browser history and a spreadsheet. Six months later, nobody can tell which figure came from which document, or which period it covered.

With InvestingBuddy

  1. Research request
  2. Primary evidence located
  3. Structured, labelled facts
  4. Research council
  5. Report
  6. Human review

Spend less time collecting and reconciling evidence, and more evaluating the investment — with every figure still attached to the document, period and scope it came from.

What it is

A research workflow, not a data terminal.

InvestingBuddy orchestrates discovery, primary evidence, financial extraction, research agents and auditability into one pass. It sits before the decision, not at it.

It is not

  • a market terminal
  • a stock screener
  • a transcript database
  • a charting platform
  • a general-purpose AI chat window

It does not execute trades, connect to a broker, or manage a portfolio — and it never will as part of this product.

Start with one company, or with the idea behind it.

Research runs inside a private workspace. Sign in with your authorised account to begin.

Analyze a companyDiscover opportunitiesView existing research