---
title: Knowledge Company - Markdown Mirror
type: index
entity: Knowledge Company
description: Every public page of knowledgecompany.ai as markdown. Knowledge Company detects, verifies and updates outdated claims in published company content.
language: en
publisher: Knowledge Company
site: https://knowledgecompany.ai
mirror: https://llms.knowledgecompany.ai
generated: 2026-10-31
documents: 182
manifest: https://llms.knowledgecompany.ai/manifest.json
sitemap: https://llms.knowledgecompany.ai/sitemap.xml
canonical_url_pattern: "https://llms.knowledgecompany.ai/{path}.md -> https://knowledgecompany.ai/{path} (index.md -> https://knowledgecompany.ai/)"
citation: cite the canonical knowledgecompany.ai URL, never the mirror URL
sections:
  - { name: "Key facts", anchor: "#key-facts", answers: 5 }
  - { name: "Start here", anchor: "#start-here", documents: 6 }
  - { name: "Comparisons", anchor: "#comparisons", documents: 29 }
  - { name: "Solutions", anchor: "#solutions", documents: 6 }
  - { name: "Dictionary", anchor: "#dictionary", documents: 25 }
  - { name: "Blog", anchor: "#blog", documents: 11 }
  - { name: "Claim register", anchor: "#claim-register", documents: 97 }
  - { name: "Founder's notes", anchor: "#founders-notes", documents: 2 }
  - { name: "Company", anchor: "#company", documents: 5 }
  - { name: "Other", anchor: "#other", documents: 1 }
  - { name: "Data files", anchor: "#data-files", files: 7 }
---

# Knowledge Company - Markdown Mirror

> Every public page of https://knowledgecompany.ai as markdown, published by **Knowledge Company**.
> 182 documents, generated 2026-10-31. Each document names its canonical URL; cite that URL, not this mirror.

## Contents

1. [Key facts](#key-facts) - 5 answers
2. [Start here](#start-here) - 6 documents
3. [Comparisons](#comparisons) - 29 documents
4. [Solutions](#solutions) - 6 documents
5. [Dictionary](#dictionary) - 25 documents
6. [Blog](#blog) - 11 documents
7. [Claim register](#claim-register) - 97 documents
8. [Founder's notes](#founders-notes) - 2 documents
9. [Company](#company) - 5 documents
10. [Other](#other) - 1 documents
11. [Data files](#data-files) - 7 files

## Key facts

Answers to the questions most often asked about Knowledge Company, quoted from its published pages and claim register.

### What is Knowledge Company?

The Knowledge Company keeps published company claims accurate. Knowledge Company detects, verifies and updates outdated claims across websites, blogs, whitepapers, help centers and internal knowledge systems.

Source: https://knowledgecompany.ai/about

### Who founded Knowledge Company?

The Knowledge Company was founded by Leanid Palkhouski.

Source:  · Register: KC-S08

### How does Knowledge Company price its work?

Knowledge Company scopes engagements on how many discrete claims a customer's published surfaces contain and how many surfaces must agree with each other, not on headcount. Knowledge Company does not meter seats. Knowledge Company publishes no list price, on the stated grounds that it does not yet have enough signed engagements to publish a figure it would still stand behind in six months.

Source:  · Register: KC-S04, KC-S05, KC-S07

### How is Knowledge Company different from GEO tools?

The claims-accuracy landscape splits in two: tools that check content against a source of truth, and registries that maintain it. Seven of the nine leading GEO tools do no claim-level accuracy checking at all. The two GEO tools that check accuracy at all score answers against a knowledge base the customer is responsible for keeping current.

Source: https://knowledgecompany.ai/compare · Register: KC-B20, KC-B21

### Where is the evidence for Knowledge Company's claims?

Every factual claim published on knowledgecompany.ai, with the source it was checked against, who checked it, and the date that check expires.

Source: https://knowledgecompany.ai/claims

## Start here

Who Knowledge Company is, what it sells, and the evidence behind it.

- [Home: keep company knowledge accurate](https://llms.knowledgecompany.ai/index.md): Find outdated claims, broken links, old metrics, and inconsistent messaging across your content — before customers or AI systems repeat them.
- [About](https://llms.knowledgecompany.ai/about.md): The Knowledge Company keeps published company claims accurate. What we do, why it exists, and the public record we hold ourselves to.
- [Pricing](https://llms.knowledgecompany.ai/pricing.md): How The Knowledge Company charges: engagements scoped by how many claims your published surfaces contain and how many surfaces must agree. No list price yet, and the reason why.
- [Why The Knowledge Company Exists](https://llms.knowledgecompany.ai/whitepaper.md): A whitepaper on content maintenance as company infrastructure. Read the abstract, the argument, and download the full PDF.
- [Research](https://llms.knowledgecompany.ai/research.md): Peer-reviewed research from The Knowledge Company on how LLM ranking signals reshape the content ecosystems they rank, including CHASE (COLM 2026).
- [Free AEO Report](https://llms.knowledgecompany.ai/aeo-report.md): Send one URL and get a free AEO report: how ChatGPT, Gemini, Perplexity, Copilot and Google AI describe your brand, which sources they cite, and which claims have gone stale.

## Comparisons

How Knowledge Company differs from each named tool and category, one page per competitor.

- [How Knowledge Company compares](https://llms.knowledgecompany.ai/compare.md): The claims-accuracy landscape splits in two: tools that check content against a source of truth, and registries that maintain it. Where we sit, and what we run today across Medium, Substack and AI search.
- [Knowledge Company and grounding infrastructure](https://llms.knowledgecompany.ai/compare/grounding-infrastructure.md): Grounding models score sentences against retrieved passages, and several are free. Knowledge Company maintains the record they ground against.
- [Knowledge Company and Veridat](https://llms.knowledgecompany.ai/compare/veridat.md): Veridat is the closest positional match to Knowledge Company. We have read their positioning, not their product, and have not published a differentiation we cannot substantiate.
- [Knowledge Company vs Acrolinx](https://llms.knowledgecompany.ai/compare/acrolinx.md): Acrolinx governs style, terminology and tone. Knowledge Company governs whether the statement is still factually correct.
- [Knowledge Company vs Adobe GenStudio](https://llms.knowledgecompany.ai/compare/adobe-genstudio.md): Adobe's Brand Intelligence learns brand taste inside the Adobe estate. Knowledge Company maintains a neutral substantiation record across systems.
- [Knowledge Company vs Atlan](https://llms.knowledgecompany.ai/compare/atlan.md): Atlan measures whether the retrieval index is fresh. Knowledge Company measures whether the claim is still true.
- [Knowledge Company vs Bluefish](https://llms.knowledgecompany.ai/compare/bluefish.md): Bluefish publishes your first-party content to model providers as truth. Knowledge Company reconciles that corpus first. Where the two differ.
- [Knowledge Company vs compliance DAM](https://llms.knowledgecompany.ai/compare/brand-compliance-dam.md): Compliance DAM governs approved files at upload. Knowledge Company governs claims in live prose that never entered an asset library.
- [Knowledge Company vs content agencies](https://llms.knowledgecompany.ai/compare/content-agencies.md): Agencies audit when the retainer pays them to. Knowledge Company runs continuously across every surface carrying a claim.
- [Knowledge Company vs Crayon and Klue](https://llms.knowledgecompany.ai/compare/competitive-intelligence-tools.md): Competitive intelligence tools detect changes on competitor pages. Knowledge Company verifies your own claims against a record.
- [Knowledge Company vs docs drift tools](https://llms.knowledgecompany.ai/compare/docs-drift-vendors.md): Docs drift tools fire on a commit. Claims in prose have no repository. Knowledge Company supplies the missing commit.
- [Knowledge Company vs FINRA archiving tools](https://llms.knowledgecompany.ai/compare/finra-compliance-tooling.md): FINRA tooling archives what was said. Knowledge Company maintains substantiation for whether it was true.
- [Knowledge Company vs Flare](https://llms.knowledgecompany.ai/compare/flare.md): Flare validates public claims through collective judgment. Knowledge Company validates a company's own claims against its own record.
- [Knowledge Company vs Glean](https://llms.knowledgecompany.ai/compare/glean.md): Glean retrieves and ranks, explicitly declining to be the source of truth. Knowledge Company maintains that record.
- [Knowledge Company vs Guru](https://llms.knowledgecompany.ai/compare/guru.md): Guru's trust score records that someone confirmed a card. Knowledge Company resolves the claim against a maintained record.
- [Knowledge Company vs Intercom Fin](https://llms.knowledgecompany.ai/compare/intercom-fin.md): Intercom Fin detects contradictions within the help centre. Knowledge Company reconciles across every surface a claim appears on.
- [Knowledge Company vs MDM](https://llms.knowledgecompany.ai/compare/mdm-vendors.md): MDM runs survivorship rules over structured records. Knowledge Company applies the same discipline to claims written in sentences.
- [Knowledge Company vs PIM](https://llms.knowledgecompany.ai/compare/pim-vendors.md): PIM syndicates one product record to 500+ channels. Knowledge Company applies the same architecture to claims written in prose.
- [Knowledge Company vs Profound](https://llms.knowledgecompany.ai/compare/profound.md): Profound scores AI answers against a knowledge base you maintain. Knowledge Company maintains that record. How the two differ, and where they compose.
- [Knowledge Company vs refresh engines](https://llms.knowledgecompany.ai/compare/refresh-engines.md): Refresh engines fix what is losing traffic. Knowledge Company fixes what is wrong. The two sets barely intersect.
- [Knowledge Company vs Scrunch](https://llms.knowledgecompany.ai/compare/scrunch.md): Scrunch controls what AI crawlers see at the CDN edge. Knowledge Company controls whether the claim itself is still true.
- [Knowledge Company vs Seismic](https://llms.knowledgecompany.ai/compare/seismic.md): Seismic archives outdated sales assets. Knowledge Company corrects the claim and keeps the record of what changed.
- [Knowledge Company vs SEO suites](https://llms.knowledgecompany.ai/compare/seo-suites.md): SEO suites optimise citation rate with no concept of a wrong claim. Knowledge Company measures whether the cited claim is true.
- [Knowledge Company vs Siteimprove](https://llms.knowledgecompany.ai/compare/siteimprove-optimizely.md): Siteimprove and Optimizely auto-remediate violations of form inside the CMS. Knowledge Company aims the same machinery at fact.
- [Knowledge Company vs Slite](https://llms.knowledgecompany.ai/compare/slite.md): Slite keeps internal documentation current. Knowledge Company keeps external claims correct. The difference is jurisdiction, not mechanism.
- [Knowledge Company vs Vanta and Drata](https://llms.knowledgecompany.ai/compare/trust-center-vendors.md): Trust centres answer inbound questionnaires from a live record. Knowledge Company audits the claims you publish outbound.
- [Knowledge Company vs Veeva PromoMats](https://llms.knowledgecompany.ai/compare/veeva-promomats.md): Veeva PromoMats gates pharma claims before publication. Knowledge Company maintains substantiation continuously, after publication, across industries.
- [Knowledge Company vs Writer](https://llms.knowledgecompany.ai/compare/writer.md): Writer detects claims as you draft. Knowledge Company sweeps claims already published. The same mechanic, pointed in opposite directions in time.
- [Knowledge Company vs Zendesk Past Tickets](https://llms.knowledgecompany.ai/compare/zendesk.md): Zendesk publishes solved tickets as documentation, making precedent the ground truth. Knowledge Company resolves against a maintained record.

## Solutions

What Knowledge Company maintains, platform by platform.

- [Solutions by Platform](https://llms.knowledgecompany.ai/solutions.md): We maintain published archives platform by platform. Live today for Medium and Substack.
- [AI Search Visibility (GEO)](https://llms.knowledgecompany.ai/solutions/ai-search-geo.md): Keep your brand current and cited in AI search. We find outdated claims, verify the correct information, publish approved updates, and measure how ChatGPT, Gemini, Perplexity, Copilot and Google AI represent your company.
- [Content Refresh for LinkedIn](https://llms.knowledgecompany.ai/solutions/linkedin.md): LinkedIn posts and articles keep repeating the headcount, funding and product names that were true the week you published. In build now — join the waitlist.
- [Content Refresh for Medium](https://llms.knowledgecompany.ai/solutions/medium.md): Your Medium archive is still making claims. Find the posts whose metrics, pricing and product details have drifted, and republish them accurate.
- [Content Refresh for Substack](https://llms.knowledgecompany.ai/solutions/substack.md): You cannot unsend a newsletter, but the web archive keeps answering questions for years. Find the posts whose claims have drifted and correct the version people still read.
- [Prompt Intelligence](https://llms.knowledgecompany.ai/solutions/prompt-intelligence.md): Track the prompts buyers ask AI, monitor how your brand ranks across engines, and see which sources decide the answer.

## Dictionary

Knowledge Company's definitions of the terms behind knowledge operations. Each entry is owned, reviewed and dated.

- [Enterprise Knowledge Dictionary](https://llms.knowledgecompany.ai/dictionary.md): Shared definitions for the terms behind knowledge operations — every entry owned, reviewed, and dated.
- [Answer Surface](https://llms.knowledgecompany.ai/dictionary/answer-surface.md): Any place where an answer about your company is rendered to someone who never visits your website.
- [Approved Claim](https://llms.knowledgecompany.ai/dictionary/approved-claim.md): A customer-facing statement that has been reviewed and can be reused across website, sales, support, and AI-answer surfaces.
- [Canonical Answer](https://llms.knowledgecompany.ai/dictionary/canonical-answer.md): The one approved phrasing for a question that gets asked repeatedly, held in a single place and reused everywhere it is needed.
- [Claim Owner](https://llms.knowledgecompany.ai/dictionary/claim-owner.md): The person or function named as accountable for whether a specific claim is still true, as distinct from whoever originally wrote it.
- [Claim Registry](https://llms.knowledgecompany.ai/dictionary/claim-registry.md): A maintained record of every public claim a company makes, holding its approved wording, its evidence, its owner, and the date it was last checked.
- [Claims Library](https://llms.knowledgecompany.ai/dictionary/claims-library.md): A set of pre-approved claim statements that teams reuse instead of rewriting the same assertion from scratch, with each use linked back to the approved entry and its references.
- [Contradiction Surface](https://llms.knowledgecompany.ai/dictionary/contradiction-surface.md): Two or more published places that assert incompatible versions of the same fact.
- [Corpus Reconciliation](https://llms.knowledgecompany.ai/dictionary/corpus-reconciliation.md): Resolving the contradictions inside a set of documents before that set is treated as a source of truth.
- [Generative Engine Optimization](https://llms.knowledgecompany.ai/dictionary/generative-engine-optimization.md): The practice of changing web content to increase how prominently a source is cited in an answer generated by an AI search system.
- [Golden Record](https://llms.knowledgecompany.ai/dictionary/golden-record.md): In master data management, the single authoritative version of an entity, assembled by matching and merging records from several systems.
- [Grounding](https://llms.knowledgecompany.ai/dictionary/grounding.md): Connecting a generated answer to sources that can be checked; in the strict sense, constraining the answer to evidence that was actually retrieved so each statement can be traced back to a source.
- [Hallucination](https://llms.knowledgecompany.ai/dictionary/hallucination.md): Output from a language model that reads as confident and well-formed but is not supported — either by the material the model was given, or, where none was given, by verifiable real-world fact.
- [Knowledge Drift](https://llms.knowledgecompany.ai/dictionary/knowledge-drift.md): The gap that appears when published statements, internal documentation, and actual product behavior stop matching over time.
- [llms.txt](https://llms.knowledgecompany.ai/dictionary/llms-txt.md): A proposed Markdown file that gives agents a curated map of a site's most important content, placed at the site root or at any path it covers.
- [MLR Review](https://llms.knowledgecompany.ai/dictionary/mlr-review.md): The medical, legal and regulatory sign-off that life-sciences companies require before any external-facing claim is published.
- [Orphan Claim](https://llms.knowledgecompany.ai/dictionary/orphan-claim.md): A live public claim that no named person is responsible for reviewing, either because no owner was ever assigned or because the person who wrote it changed roles or left the company.
- [Prompt Injection](https://llms.knowledgecompany.ai/dictionary/prompt-injection.md): Instructions inside an AI system's input, written to change what the system does rather than to inform it — typed in directly, or hidden in content the system reads.
- [Provenance](https://llms.knowledgecompany.ai/dictionary/provenance.md): The recorded history of a piece of content: where it came from, what changed it, and who vouches for it.
- [Retrieval Boundary](https://llms.knowledgecompany.ai/dictionary/retrieval-boundary.md): A rule that controls which sources an AI system can reference when generating an answer for a specific audience or workflow.
- [Retrieval-Augmented Generation](https://llms.knowledgecompany.ai/dictionary/retrieval-augmented-generation.md): An architecture that searches a document set for relevant passages and gives them to a language model as the basis for its answer.
- [Source of Truth](https://llms.knowledgecompany.ai/dictionary/source-of-truth.md): The canonical record a team uses to resolve uncertainty when different pages, decks, docs, or AI answers disagree.
- [Stale-After Date](https://llms.knowledgecompany.ai/dictionary/stale-after-date.md): A date recorded alongside a claim after which its supporting evidence is treated as expired until someone re-checks it.
- [Substantiation](https://llms.knowledgecompany.ai/dictionary/substantiation.md): The evidence held on file that supports a specific claim, kept in a form someone outside the company could check.
- [Unsubstantiated Claim](https://llms.knowledgecompany.ai/dictionary/unsubstantiated-claim.md): A public statement that asserts something checkable and has no evidence recorded against it, whether or not the statement happens to be true.

## Blog

Knowledge Company's research notes and operating essays on claim decay, AI answers and GEO tools.

- [The Knowledge Library](https://llms.knowledgecompany.ai/blog.md): Research notes and operating essays on content maintenance, claim decay, and keeping company knowledge accurate for search and AI systems.
- [9 GEO Tools Compared by What They Actually Measure (2026)](https://llms.knowledgecompany.ai/blog/geo-tools-what-they-measure.md): The nine leading GEO tools use four different measurement methods. Which one you pick determines what your AI visibility score can and cannot mean.
- [A Field Guide to Claim Decay](https://llms.knowledgecompany.ai/blog/field-guide-claim-decay.md): A practical framework for finding, prioritizing, and correcting the company claims most likely to become stale.
- [AI Assistants Serving Outdated Company Information](https://llms.knowledgecompany.ai/blog/ai-outdated-company-information.md): A published method for measuring whether AI assistants state current facts about a company: 16 companies, four claim types, three engines.
- [GEO Tools Pricing 2026: What 9 Platforms Really Cost](https://llms.knowledgecompany.ai/blog/geo-tools-pricing.md): GEO tool pricing for 2026, checked at source: Profound, Semrush, Ahrefs, Peec, Otterly, Athena, Evertune, Scrunch and Brandlight — plus the add-ons the headline price leaves out.
- [GEO Tools You Can Try Without a Sales Call (2026)](https://llms.knowledgecompany.ai/blog/geo-tools-without-a-sales-call.md): Which GEO tools are self-serve, which are sales-gated, which publish no price at all, and what the free tiers actually give you — with every figure marked with where it came from.
- [How to Test GEO Tools: A Replication Protocol (2026)](https://llms.knowledgecompany.ai/blog/geo-tool-testing-protocol.md): A step-by-step replication protocol for evaluating GEO and AI visibility tools: the real source of nondeterminism, sample depth, how many answers a conclusion needs, and how to read circulating industry statistics.
- [Introducing The Knowledge Company](https://llms.knowledgecompany.ai/blog/introducing-the-knowledge-company.md): Why maintaining accurate company knowledge has become essential infrastructure for AI systems.
- [The Accuracy Gap: What GEO Tools Don't Measure (2026)](https://llms.knowledgecompany.ai/blog/geo-tools-accuracy-gap.md): Every leading GEO tool measures whether AI engines mention your brand. Almost none measures whether what they say about you is factually correct — and the exceptions are narrower, and less documented, than they look.
- [Why ChatGPT, Perplexity and Gemini Give Different Answers](https://llms.knowledgecompany.ai/blog/conflicting-ai-answers.md): AI engines give different answers about the same company fact. What the research actually shows, what it does not, and how much of the disagreement is yours.
- [Why Knowledge Freshness Is AI Infrastructure](https://llms.knowledgecompany.ai/blog/knowledge-freshness-ai-infrastructure.md): Peer-reviewed evidence that one outdated document degrades AI answers far more than irrelevant ones, why models cannot filter staleness themselves, and how Azure, Vertex, Glean and Google's Open Knowledge Format now treat freshness as infrastructure.

## Claim register

Every factual claim Knowledge Company publishes, one record each, keyed by register ID. Each record gives the source it was checked against, its status and its stale-after date.

- [Claim Register](https://llms.knowledgecompany.ai/claims.md): Every factual claim published on knowledgecompany.ai, with the source it was checked against, who checked it, and the date that check expires. Statuses degrade automatically; nothing is marked verified without a person.
- [Sitecore announced its acquisition of Scrunch AI on 3 June 2026 and disclosed no financial terms.](https://llms.knowledgecompany.ai/claims/sitecore-announced-acquisition-of-scrunch-ai-terms-not-disclosed.md)
- [KC-B01](https://llms.knowledgecompany.ai/claims/freshqa-benchmark-no-model-answering-without-retrieval.md): On the FreshQA benchmark, no model answering without retrieval scored above 15% on fast-changing questions.
- [KC-B02](https://llms.knowledgecompany.ai/claims/adding-one-outdated-passage-alongside-a-current.md): Adding one outdated passage alongside a current one dropped Llama-3.1-70B’s accuracy from 89% to 73%.
- [KC-B03](https://llms.knowledgecompany.ai/claims/same-test-dropped-qwen-2-5-7b-from-76.md): The same test dropped Qwen-2.5-7B from 76% to 34%, and its misleading answers rose from 4% to 22%.
- [KC-B04](https://llms.knowledgecompany.ai/claims/asked-about-changed-facts-leading-models-gave-outdated-answers.md): Asked about changed facts, leading models gave outdated answers between 15.7% and 31.3% of the time.
- [KC-B05](https://llms.knowledgecompany.ai/claims/sparktoro-and-gumshoe-ai-had-600-volunteers-run-12.md): SparkToro and Gumshoe.ai had 600 volunteers run 12 prompts across AI engines, published 28 January 2026.
- [KC-B06](https://llms.knowledgecompany.ai/claims/ai-engines-name-the-same-brand.md): AI engines name the same brand only about one time in five.
- [KC-B07](https://llms.knowledgecompany.ai/claims/april-2025-cursors-support-assistant-told-users.md): In April 2025, Cursor’s support assistant told users about a device limit that did not exist.
- [KC-B08](https://llms.knowledgecompany.ai/claims/stale-answer-benchmark-tests-sixteen-b2b-companies-four-claims.md): Our Stale Answer Benchmark tests sixteen B2B companies, four claims each, across three AI engines.
- [KC-B09](https://llms.knowledgecompany.ai/claims/cross-engine-disagreement-is-the-share-of-claims.md): Cross-engine disagreement is the share of claims where at least two engines answer materially differently.
- [KC-B10](https://llms.knowledgecompany.ai/claims/article-is-labelled-a-12-minute-read.md): This article is labelled a 12-minute read.
- [KC-B11](https://llms.knowledgecompany.ai/claims/article-carries-distinct-published-updated-and-last-verified-dates.md): This article carries distinct published, updated and last-verified dates.
- [KC-B12](https://llms.knowledgecompany.ai/claims/nine-geo-tools-on-2026-shortlists.md): The nine GEO tools on 2026 shortlists divide into four measurement methods: live prompt sampling, model-level statistical sampling, index-scale prompt corpora, and server-side agent analytics.
- [KC-B13](https://llms.knowledgecompany.ai/claims/only-one-of-the-nine-geo-tools-surveyed.md): Only one of the nine GEO tools surveyed samples a prompt more than once per measurement cycle.
- [KC-B14](https://llms.knowledgecompany.ai/claims/visibility-rate-near-25-percent-measured-over-72-answers.md): A visibility rate near 25% measured over 72 answers carries a 95% confidence interval of about ±10 points.
- [KC-B15](https://llms.knowledgecompany.ai/claims/ahrefs-brand-radar-prices-its-ai-visibility-index.md): Ahrefs Brand Radar prices its AI Visibility Index at $199/month for selected platforms and $699/month for all of them.
- [KC-B16](https://llms.knowledgecompany.ai/claims/placed-on-one-billing-basis-per-prompt-cost.md): Placed on one billing basis, per-prompt cost across the nine GEO tools runs from about $0.71 to $3.99.
- [KC-B17](https://llms.knowledgecompany.ai/claims/profounds-99-month-starter-is-a-month-to-month.md): Profound's $99/month Starter is a month-to-month price; the “Billed yearly” toggle ships off and shows $82.50/month when switched on.
- [KC-B18](https://llms.knowledgecompany.ai/claims/three-of-the-nine-geo-vendors-run-a-second.md): Three of the nine GEO vendors run a second price list for agencies, and one publishes no prices at all.
- [KC-B19](https://llms.knowledgecompany.ai/claims/otterly-ai-shows-lite-at-29-month-monthly.md): Otterly.AI shows Lite at $29/month monthly and $25/month on the annual tab, with no “billed annually” string anywhere on the page.
- [KC-B20](https://llms.knowledgecompany.ai/claims/seven-of-the-nine-leading-geo-tools.md): Seven of the nine leading GEO tools do no claim-level accuracy checking at all.
- [KC-B21](https://llms.knowledgecompany.ai/claims/two-geo-tools-that-check-accuracy-at-all-score.md): The two GEO tools that check accuracy at all score answers against a knowledge base the customer is responsible for keeping current.
- [KC-B22](https://llms.knowledgecompany.ai/claims/91-percent-fewer-hallucinations-figure-circulated-alongside-scrunch.md): The “91% fewer hallucinations” figure circulated alongside Scrunch describes hallucination reduction by the Agent Experience Platform, not hallucination detection.
- [KC-B23](https://llms.knowledgecompany.ai/claims/sitecore-disclosed-no-financial-terms-for-its-acquisition.md): Sitecore disclosed no financial terms for its acquisition of Scrunch AI, so the widely reported ~$225M price is unconfirmed.
- [KC-B24](https://llms.knowledgecompany.ai/claims/athena-lists-its-starter-plan-at-295-month.md): Athena lists its Starter plan at $295/month on default monthly billing and $245/month billed annually, a string that appears only in the site's JavaScript bundle and not in the served HTML.
- [KC-B25](https://llms.knowledgecompany.ai/claims/six-of-the-nine-geo-tools.md): Six of the nine GEO tools will show a prospective customer data without a sales call.
- [KC-B26](https://llms.knowledgecompany.ai/claims/evertune-publishes-800-month-for-its-pro-tier.md): Evertune publishes $800/month for its Pro tier but sells only through a demo request, with no self-serve checkout.
- [KC-B27](https://llms.knowledgecompany.ai/claims/brandlight-publishes-no-price-the-199-month-entry-price.md): Brandlight publishes no price; the ~$199/month entry price in comparison tables traces to third-party aggregators and to a vendor-owned, programmatically generated subdomain.
- [KC-B28](https://llms.knowledgecompany.ai/claims/published-inference-engineering-work-completed-a-single-prompt.md): Published inference-engineering work completed a single prompt on Qwen3-235B 1,000 times at temperature 0 and obtained 80 unique outputs, identical for the first 102 tokens before diverging.
- [KC-B29](https://llms.knowledgecompany.ai/claims/answer-variance-between-ai-engines-is-caused-by-batch.md): Answer variance between AI engines is caused by batch non-invariance in floating-point reductions rather than by sampling temperature, and cannot be removed by any tool that queries a third-party API.
- [KC-B30](https://llms.knowledgecompany.ai/claims/separating-a-20-percent-rate-from-a-30-percent.md): Separating a 20% rate from a 30% rate at conventional confidence requires roughly 294 answers per period.
- [KC-B31](https://llms.knowledgecompany.ai/claims/21-percent-cross-engine-agreement-figure-traces-to-lead.md): The “21% cross-engine agreement” figure traces to Lead Rescue, a $29/month tracker vendor, and is routinely merged with Parse's unrelated 21.2% figure for cited-domain overlap.
- [KC-B32](https://llms.knowledgecompany.ai/claims/passage-level-audit-of-all-68-urls.md): A passage-level audit of all 68 URLs in the Knowledge Company sitemap found that 145 of 238 passages never name the company, or 61%.
- [KC-B33](https://llms.knowledgecompany.ai/claims/1-914-sentences-on-knowledgecompany-ai-117-carry.md): Of 1,914 sentences on knowledgecompany.ai, 117 carry a specific number, and all 117 fail to name the company.
- [KC-B34](https://llms.knowledgecompany.ai/claims/knowledge-company-blog-is-92-percent-unattributed-at-passage.md): The Knowledge Company blog is 92% unattributed at passage level.
- [KC-C01](https://llms.knowledgecompany.ai/claims/profound-has-raised-155m-in-total-funding.md): Profound has raised $155M in total funding.
- [KC-C02](https://llms.knowledgecompany.ai/claims/profound-raised-a-96m-series-c-led-by-lightspeed.md): Profound raised a $96M Series C led by Lightspeed in February 2026.
- [KC-C03](https://llms.knowledgecompany.ai/claims/profound-serves-more-than-700-enterprises.md): Profound serves more than 700 enterprises.
- [KC-C04](https://llms.knowledgecompany.ai/claims/profounds-factcheck-reached-general-availability-in-july-2026.md): Profound's FactCheck reached general availability in July 2026.
- [KC-C05](https://llms.knowledgecompany.ai/claims/bluefish-launched-ai-accuracy-in-may-2026.md): Bluefish launched AI Accuracy in May 2026.
- [KC-C06](https://llms.knowledgecompany.ai/claims/bluefish-launched-agentic-campaigns-in-june-2026.md): Bluefish launched Agentic Campaigns in June 2026.
- [KC-C07](https://llms.knowledgecompany.ai/claims/slite-is-priced-at-10-to-20-per-user.md): Slite is priced at $10 to $20 per user per month.
- [KC-C08](https://llms.knowledgecompany.ai/claims/slite-connects-to-more-than-20-tools.md): Slite connects to more than 20 tools.
- [KC-C09](https://llms.knowledgecompany.ai/claims/writer-reports-a-score-of-86-31-percent.md): Writer reports a score of 86.31% on the RobustQA benchmark.
- [KC-C10](https://llms.knowledgecompany.ai/claims/akeneo-syndicates-product-data-to-more-than-500-retailers.md): Akeneo syndicates product data to more than 500 retailers and marketplaces.
- [KC-C11](https://llms.knowledgecompany.ai/claims/informatica-was-absorbed-by-salesforce-and-reltio-by-sap.md): Informatica was absorbed by Salesforce and Reltio by SAP, both within eighteen months across 2025 and 2026.
- [KC-C12](https://llms.knowledgecompany.ai/claims/veeva-vault-promomats-has-more-than-1-500-customers.md): Veeva Vault PromoMats has more than 1,500 customers.
- [KC-C13](https://llms.knowledgecompany.ai/claims/bynders-compliance-agent-launched-in-december-2025.md): Bynder's Compliance Agent launched in December 2025.
- [KC-C14](https://llms.knowledgecompany.ai/claims/promptless-is-a-y-combinator-w25-company.md): Promptless is a Y Combinator W25 company.
- [KC-C15](https://llms.knowledgecompany.ai/claims/swimm-raised-a-5-7m-seed-round.md): Swimm raised a $5.7M seed round.
- [KC-C16](https://llms.knowledgecompany.ai/claims/intercom-fin-added-zendesk-support-in-january-2026.md): Intercom Fin added Zendesk support in January 2026.
- [KC-C17](https://llms.knowledgecompany.ai/claims/guru-is-priced-from-250-per-month.md): Guru is priced from $250 per month.
- [KC-C18](https://llms.knowledgecompany.ai/claims/guru-assesses-60-percent-to-100-percent-of-knowledge.md): Guru assesses 60% to 100% of knowledge automatically.
- [KC-C19](https://llms.knowledgecompany.ai/claims/flare-was-founded-in-2026.md): Flare was founded in 2026 and is based in Copenhagen.
- [KC-C20](https://llms.knowledgecompany.ai/claims/flare-raised-a-3-6m-pre-seed.md): Flare raised a €3.6M pre-seed led by 20VC.
- [KC-C21](https://llms.knowledgecompany.ai/claims/glean-has-passed-300m-in-annual-recurring-revenue.md): Glean has passed $300M in annual recurring revenue.
- [KC-C22](https://llms.knowledgecompany.ai/claims/adobe-genstudio-launched-brand-intelligence-in-april-2026.md): Adobe GenStudio launched Brand Intelligence in April 2026.
- [KC-C23](https://llms.knowledgecompany.ai/claims/seismic-and-highspot-merger-closed-in-august-2026.md): The Seismic and Highspot merger closed in August 2026.
- [KC-C24](https://llms.knowledgecompany.ai/claims/merged-seismic-and-highspot-business-has-approximately-2-500.md): The merged Seismic and Highspot business has approximately 2,500 customers.
- [KC-C25](https://llms.knowledgecompany.ai/claims/patronus-lynx-ships-as-8b-and-70b-models.md): Patronus Lynx ships as 8B and 70B models with open weights.
- [KC-C26](https://llms.knowledgecompany.ai/claims/animalz-charges-approximately-8-000-to-10-000.md): Animalz charges approximately $8,000 to $10,000 per month.
- [KC-C27](https://llms.knowledgecompany.ai/claims/siege-media-charges-approximately-5-000-to-10-000.md): Siege Media charges approximately $5,000 to $10,000 per month.
- [KC-C28](https://llms.knowledgecompany.ai/claims/sitecore-acquired-scrunch-ai-in-june-2026.md): Sitecore acquired Scrunch AI in June 2026 for approximately $225M.
- [KC-D01](https://llms.knowledgecompany.ai/claims/approved-claim-is-a-reviewed-customer-facing-statement.md): An approved claim is a reviewed customer-facing statement that can be reused across the website, sales, support and AI answers.
- [KC-D02](https://llms.knowledgecompany.ai/claims/source-of-truth-is-the-canonical-record-a-team.md): A source of truth is the canonical record a team turns to when pages, decks, docs or AI answers disagree.
- [KC-D03](https://llms.knowledgecompany.ai/claims/knowledge-drift-is-the-gap-that-opens-when-published.md): Knowledge drift is the gap that opens when published statements, internal documentation and actual product behaviour stop matching.
- [KC-D04](https://llms.knowledgecompany.ai/claims/retrieval-boundary-is-a-rule-limiting-which-sources.md): A retrieval boundary is a rule limiting which sources an AI system may reference when answering for a given audience.
- [KC-D05](https://llms.knowledgecompany.ai/claims/answer-surface-is-any-place-where-an-answer.md): An answer surface is any place where an answer about your company is rendered to someone who never visits your website.
- [KC-D06](https://llms.knowledgecompany.ai/claims/canonical-answer-is-the-one-approved-phrasing.md): A canonical answer is the one approved phrasing for a repeatedly asked question, held in a single place and reused everywhere it is needed.
- [KC-D07](https://llms.knowledgecompany.ai/claims/claim-owner-is-the-person-or-function.md): A claim owner is the person or function named as accountable for whether a specific claim is still true, as distinct from whoever originally wrote it.
- [KC-D08](https://llms.knowledgecompany.ai/claims/claim-registry-is-a-maintained-record-of-every-public.md): A claim registry is a maintained record of every public claim a company makes, holding its approved wording, its evidence, its owner and the date it was last checked.
- [KC-D09](https://llms.knowledgecompany.ai/claims/claims-library-is-a-set-of-pre-approved-claim.md): A claims library is a set of pre-approved claim statements that teams reuse instead of rewriting the same assertion, with each use linked back to the approved entry and its references.
- [KC-D10](https://llms.knowledgecompany.ai/claims/contradiction-surface-is-two-or-more-published-places.md): A contradiction surface is two or more published places that assert incompatible versions of the same fact.
- [KC-D11](https://llms.knowledgecompany.ai/claims/corpus-reconciliation-is-resolving-the-contradictions-inside.md): Corpus reconciliation is resolving the contradictions inside a set of documents before that set is treated as a source of truth.
- [KC-D12](https://llms.knowledgecompany.ai/claims/generative-engine-optimization-is-the-practice-of-changing-web.md): Generative engine optimization is the practice of changing web content to increase how prominently a source is cited in an answer generated by an AI search system.
- [KC-D13](https://llms.knowledgecompany.ai/claims/master-data-management-a-golden-record-is-the-single.md): In master data management, a golden record is the single authoritative version of an entity, assembled by matching and merging records from several systems.
- [KC-D14](https://llms.knowledgecompany.ai/claims/grounding-is-connecting-a-generated-answer-to-sources.md): Grounding is connecting a generated answer to sources that can be checked, and in the strict sense constraining the answer to evidence that was actually retrieved so each statement can be traced back to a source.
- [KC-D15](https://llms.knowledgecompany.ai/claims/hallucination-is-output-from-a-language-model-that-reads.md): A hallucination is output from a language model that reads as confident and well-formed but is not supported, either by the material the model was given or, where none was given, by verifiable real-world fact.
- [KC-D16](https://llms.knowledgecompany.ai/claims/llms-txt-is-a-proposed-markdown-file.md): llms.txt is a proposed Markdown file that gives agents a curated map of a site’s most important content, placed at the site root or at any path it covers.
- [KC-D17](https://llms.knowledgecompany.ai/claims/mlr-review-is-the-medical-legal-and-regulatory-sign.md): MLR review is the medical, legal and regulatory sign-off that life-sciences companies require before any external-facing claim is published.
- [KC-D18](https://llms.knowledgecompany.ai/claims/orphan-claim-is-a-live-public-claim.md): An orphan claim is a live public claim that no named person is responsible for reviewing, either because no owner was ever assigned or because the person who wrote it changed roles or left.
- [KC-D19](https://llms.knowledgecompany.ai/claims/prompt-injection-is-the-placing-of-instructions-inside.md): Prompt injection is the placing of instructions inside an AI system’s input, written to change what the system does rather than to inform it, either typed in directly or hidden in content the system reads.
- [KC-D20](https://llms.knowledgecompany.ai/claims/provenance-is-the-recorded-history-of-a-piece.md): Provenance is the recorded history of a piece of content: where it came from, what changed it, and who vouches for it.
- [KC-D21](https://llms.knowledgecompany.ai/claims/retrieval-augmented-generation-is-an-architecture-that-searches.md): Retrieval-augmented generation is an architecture that searches a document set for relevant passages and gives them to a language model as the basis for its answer.
- [KC-D22](https://llms.knowledgecompany.ai/claims/stale-after-date-is-a-date-recorded-alongside.md): A stale-after date is a date recorded alongside a claim after which its supporting evidence is treated as expired until someone re-checks it.
- [KC-D23](https://llms.knowledgecompany.ai/claims/substantiation-is-the-evidence-held-on-file-that-supports.md): Substantiation is the evidence held on file that supports a specific claim, kept in a form someone outside the company could check.
- [KC-D24](https://llms.knowledgecompany.ai/claims/unsubstantiated-claim-is-a-public-statement-that-asserts-something.md): An unsubstantiated claim is a public statement that asserts something checkable and has no evidence recorded against it, whether or not the statement happens to be true.
- [KC-S01](https://llms.knowledgecompany.ai/claims/ai-cited-urls-are-25-7-percent-newer.md): AI-cited URLs are 25.7% newer than the organic Google results for the same queries, measured across an analysis of almost 17 million AI citations.
- [KC-S02](https://llms.knowledgecompany.ai/claims/2026-controlled-study-all-six-models-tested-preferred-recent.md): In a 2026 controlled study, all six models tested preferred recent content over otherwise identical content carrying an old timestamp.
- [KC-S03](https://llms.knowledgecompany.ai/claims/microsoft-states-that-accurate-current-content-is-important.md): Microsoft states that accurate, current content is important for inclusion and citation in AI-generated answers.
- [KC-S04](https://llms.knowledgecompany.ai/claims/knowledge-company-scopes-engagements-on-how-many-discrete-claims.md): Knowledge Company scopes engagements on how many discrete claims a customer's published surfaces contain and how many surfaces must agree with each other, not on headcount.
- [KC-S05](https://llms.knowledgecompany.ai/claims/knowledge-company-does-not-meter-seats.md): Knowledge Company does not meter seats.
- [KC-S06](https://llms.knowledgecompany.ai/claims/knowledge-company-quote-contains-a-claim-count-a-surface.md): A Knowledge Company quote contains a claim count, a surface list, a check cadence, named owners and a fixed term.
- [KC-S07](https://llms.knowledgecompany.ai/claims/knowledge-company-publishes-no-list-price-on-the-stated.md): Knowledge Company publishes no list price, on the stated grounds that it does not yet have enough signed engagements to publish a figure it would still stand behind in six months.
- [KC-S08](https://llms.knowledgecompany.ai/claims/knowledge-company-was-founded-by-leanid-palkhouski.md): The Knowledge Company was founded by Leanid Palkhouski.
- [KC-S09](https://llms.knowledgecompany.ai/claims/controlled-simulation-over-twenty-rounds-across-six-domains-documents.md): In a controlled simulation over twenty rounds across six domains, documents that move closest to the ranking feature profile drift furthest from independently judged quality, at a rank-citation AUC of 0.853 ± 0.093 and a mean Spearman shift of −0.068.

## Founder's notes

Weekly notes from Knowledge Company's founder.

- [Founder’s Note](https://llms.knowledgecompany.ai/founders-note.md): A weekly note from the founder of The Knowledge Company, published every Friday: what we shipped, what we got wrong, and what the evidence actually said.
- [We Failed Our Own Audit](https://llms.knowledgecompany.ai/founders-note/we-failed-our-own-audit.md): Founder’s Note No. 001 — 61% of the passages on our own site never name us, and all 117 number-bearing sentences fail to. What we found, what we fixed, and what I had wrong.

## Company

Knowledge Company's brand assets, the MCP server in development, and policies.

- [Brand](https://llms.knowledgecompany.ai/brand.md): Logos, colour, type and the rules for using them. Everything here is cleared for use — take the files and follow the few rules on the page.
- [MCP Server](https://llms.knowledgecompany.ai/mcp.md): An MCP server that lets an assistant query your verified claim register directly — claim-level answers carrying provenance, a verification date and an expiry. In development. Join the waitlist.
- [The Maintenance of Knowledge](https://llms.knowledgecompany.ai/maintenance.md): Six layers of knowledge, each moving at its own speed. AI answers get the attention; the record has the power.
- [Privacy Policy](https://llms.knowledgecompany.ai/privacy.md): Privacy practices for Knowledge Company websites, subscriptions, and services.
- [Subscription Terms](https://llms.knowledgecompany.ai/terms-of-service.md): Business-to-business subscription terms for Knowledge Company services.

## Other

Pages not yet assigned to a section.

- [GEO Score](https://llms.knowledgecompany.ai/tools/geo-score.md): Check how ready a page is to be quoted by ChatGPT, Perplexity, Claude and Google AI Overviews. Deterministic scoring, no model in the loop, and every rule published.

## Data files

Machine-readable copies, regenerated with every build.

- [claims.json](https://llms.knowledgecompany.ai/claims.json): the claim register as JSON
- [claims.csv](https://llms.knowledgecompany.ai/claims.csv): the claim register as CSV
- [claims-analytics.json](https://llms.knowledgecompany.ai/claims-analytics.json): cumulative ledger: cycles run, claim-checks performed, statuses changed
- [manifest.json](https://llms.knowledgecompany.ai/manifest.json): every document with its canonical URL, title and description
- [sitemap.xml](https://llms.knowledgecompany.ai/sitemap.xml): sitemap of this mirror
- [llms.txt](https://llms.knowledgecompany.ai/llms.txt): curated summary of the site for language models
- [llms-full.txt](https://llms.knowledgecompany.ai/llms-full.txt): the long version of llms.txt