Editorial Methodology

Source selection

SSP analyzes public source material — talks, published essays, recorded interviews, conference presentations, and strategic frameworks — from marketing and strategy experts.

Source classification

Each source is classified by provenance to help readers assess reliability:

Primary — Official channel

Uploaded directly by the expert or their official organization (e.g., an expert's own YouTube channel, official publication).

Third-party — Verified outlet

Uploaded by an established event organizer, media outlet, or conference host (e.g., a conference organization's official YouTube channel).

Clip / Repost

Edited, clipped, or reposted by a third party. Lower provenance confidence. Used only when primary or verified third-party sources are unavailable.

Unknown provenance

Source origin cannot be verified. Flagged explicitly in the analysis.

Transcript handling

Transcripts are produced via AI-assisted transcription from source audio or video. Key claims, predictions, and tactical recommendations are reviewed for accuracy against the original source. Transcripts are not published directly — they serve as the input layer for structured analysis. No creative paraphrasing is applied.

Analysis generation

Articles are produced using LLM-assisted extraction under deterministic editorial schemas. The pipeline extracts core thesis statements, testable predictions, tactical recommendations, and timestamps — then structures the output into a consistent editorial format.

Final output is subject to human editorial review and zero-tolerance link quality assurance before publishing. The goal is structured intelligence, not generative content.

Evidence attribution

When an expert references external data, market trends, or third-party research, our pipeline requires a verifiable citation. The following rules apply:

• Every external claim requires a linked citation
• Broken links are actively stripped during quality assurance
• Hard-paywall sources are blocked — only open, canonical sources are permitted
• Claims without available evidence are explicitly marked as unverified

Uncertainty handling

When evidence is mixed, when a prediction's evaluation window is still open, or when a claim cannot be fully verified, SSP flags the uncertainty explicitly in the analysis. We do not present ambiguous evidence as settled conclusions.

Prediction tracking

Predictions extracted from expert source material are tracked using an objective status system:

Confirmed

Supported by third-party data, platform features, or observable market evidence.

Partially Confirmed

Directionally accurate, but magnitude or timeline was materially off.

Evolving

The evaluation window is still open and evidence is mixed or insufficient.

Wrong

The timeline passed without the predicted event occurring, or evidence contradicts the claim.

Updates and corrections

Articles are updated when new evidence emerges, when a factual error is identified, or when the source material changes. Corrections are noted in the article. Prediction statuses are updated as evaluation windows close and new data becomes available.

If you identify an error, contact us at contact@kymatalabs.com.

Source removal and unavailability

When a source video is removed, made private, or becomes otherwise unavailable, the analysis is preserved with a clear notice that the original source is no longer accessible. The editorial content remains available because the analysis has independent value as structured commentary. Dead source links are flagged and the video status is updated in the article metadata.