Methodology

§1. Who this is for

ProductBeacon Research is open-web market research with a publishing standard that three audiences should be able to verify independently. We write to that bar and publish the rules so the bar is checkable. Read about the author for operator background and the author-independence disclosure.

For private-equity and hedge-fund analysts, methodology is what separates research from opinion. You need to know whether a Winner / Watch / Loser characterization traces to a public signal you can re-derive, whether the author has a position in any named security, and whether citation density matches claim strength.

For category-tracking operators and buyers, methodology is the difference between a vendor landscape you can take into a decision meeting and one you cannot. You need positioning paragraphs anchored to dated vendor URLs, category boundaries that follow the vendor's own current product page, and a commitment that we will not retroactively re-classify a vendor mid-research.

For fractional founders, CPOs, and product leaders, methodology is the portable part of the work. The frameworks below work for any market we eventually cover; markets come and go.

§2. Sourcing rules

Every claim in a ProductBeacon Research chapter traces to one of the following published surfaces:

• Vendor earnings call transcripts and investor materials.
• Named-outlet journalism (Bloomberg, Reuters, Fortune, TechCrunch, Calcalist, SecurityWeek, Dark Reading, WSJ, FT, NYT, The Information, plus equivalent regional and industry outlets).
• Analyst category creation or reclassification (Gartner Magic Quadrant, Market Guide, Hype Cycle; IDC MarketScape; Forrester Wave).
• Public vendor product pages, with dated capture (hero copy, headline capabilities, category labels).
• SEC filings (S-1, 10-K, 10-Q segment disclosures; 8-K material event filings).
• Funding rounds with named lead investor and valuation.
• Vendor LinkedIn self-positioning (useful for trend signals, never sufficient alone for casualty framing).

Pitchbook, Crunchbase-premium, and The Information subscriber-only data are used only via named journalistic republication. Direct subscriber-only database republication is out of scope.

What does not count: anonymous "industry source" claims, vendor self-reports without independent corroboration, future-tense roadmap statements treated as present capability, and paid analyst reports without primary-source backup.

When a citation URL breaks, we replace it with the stable Wayback snapshot at web.archive.org/web/2*/{original-url}. Evidence preservation is part of the publishing standard.

Each report ships with a master citations file regenerated at every chapter release. Citation density follows the cite-when-claiming rule below in §5. We cite when we claim, not when we speculate.

§3. Verifiable proxy rule

The hardest methodological commitment we make: every falsifiable test in the report grounds on a publicly observable signal the reader can independently re-derive. This applies to Pattern Claim Conditional predictions, Plays-Conditional outcomes, Campaign Ahead watchlist items, and the falsifiable-test footers on every Pattern Claim diagram.

This rule has teeth. In pre-publication audits, predictions that originally hung on "enterprise RFP language" as the falsifiable test get rewritten before ship. RFPs are private; a reader cannot independently observe what an enterprise puts into one. The rewrite grounds the same prediction on a public signal type from §2.

Worked examples of the substitution pattern:

• A prediction about vendor consolidation rewritten from "RFPs increasingly require [capability X]" to a named-outlet customer-win count plus the vendor earnings-call segment disclosure.
• A prediction about category displacement rewritten from "buyer RFP language shifts" to a snapshot-comparable change in product-page hero copy plus analyst category reclassification.
• A prediction about market absorption rewritten from "RFP line items" to a research-analyst category-reclassification event plus public funding-round framing.
• A prediction about cross-category dynamics rewritten from "private win-loss intel" to two named-outlet enterprise customer-win disclosures within a 90-day window.

Predictions grounded on private data cannot be falsified by the reader; predictions grounded on public signals can. The reader who returns in 2027 should be able to check whether we were right or wrong using the same evidence categories we used to make the call.

§4. Disclosure framework

Author conflict. Yohay Etsion is Head of Product (Fractional) at AXIA, which competes in the Data Loss Prevention (DLP) segment. Where DLP appears in ProductBeacon Research coverage, this disclosure appears on the chapter cover statement, in the footer of the relevant Research page, and again here. AXIA's scope is DLP only; the disclosure is narrow to that segment and does not transfer to other segments or markets.

No vendor sponsors. No paywalled data. No analyst-firm reuse. We do not accept vendor sponsorship, republish paywalled analyst datasets, or reuse paid analyst-firm conclusions. The frameworks are our own.

No vendor receives pre-publication review of how they are covered. Pre-publication vendor "review" is the mechanism through which industry analysis quietly becomes vendor-flavored content. We do not run that loop. Named vendors receive the published chapter at the same time as the rest of the audience.

Post-publication factual-correction window. Within five business days of publication, any named vendor may request a factual correction by writing to the published research contact. We will correct factual inaccuracies (misquoted product page, wrong funding number, mistaken acquisition date). We will not negotiate positioning, change Winner / Watch / Loser characterizations, or remove Pattern Claims under vendor pressure. The correction window is for facts, not framing.

Methodology applies equally to AXIA-adjacent vendors and to non-adjacent vendors. There is no parity exception.

Author independence

The author serves as Head of Product (Fractional) at AXIA. AXIA had no editorial input and no review rights on any chapter. No vendor pays for inclusion or placement. Full disclosure at /research/author.html#disclosure.

§5. Citation discipline

Every ProductBeacon Research chapter ships with a master citations file regenerated at every chapter release. The file lives on the report's main page; each report indexes its own citations and cross-references.

Per-chapter density follows the cite-when-claiming rule: we cite at the point in the prose where the claim is made. We do not cite to decorate, and we do not pile citations at the end of a section as a substitute for in-line attribution.

§6. Refresh cadence

ProductBeacon Research operates on a Quarterly Refresh cadence. Every three months we re-audit each published chapter against three signal types: (1) new earnings disclosures, (2) named-outlet customer-win disclosures that materially shift a vendor bucket assignment, and (3) analyst category-reclassification events.

A v0.X bump indicates a material change: a Pattern Claim is added, removed, or its falsifiable test is rewritten; a vendor moves between Gravity, Attention, and Wildcard buckets; a Winner / Watch / Loser characterization changes; or a new vendor is added. A Quarterly Refresh note is incremental.

Reader notification. Readers on the ship-event email list are notified when a chapter ships, when a Quarterly Refresh adds material content, and when a new market opens. Expected volume: six to ten emails per year. Ship events only.

§7. Per-market scoping

This methodology is domain-agnostic. The sourcing rules, Verifiable Proxy Rule, disclosure framework, citation discipline, and refresh cadence above apply to any market that ProductBeacon Research eventually covers.

Each market may add a per-market scoping addendum below this section specifying market-specific vendor selection rules, analyst-firm authorities (the relevant Magic Quadrants and Waves differ by category), and bucket-definition criteria. The structure reserves the slot; the canonical methodology page does not require a v2.0 rewrite when a new market opens. Markets covered to date are listed on the /research/ umbrella page; new markets are announced via the ship-event email list.

(No per-market addenda are published at first publication. Markets covered to date follow the canonical methodology without per-market modification.)

§8. Evidence preservation

We commit to URL stability for published chapter URLs and citation URLs for as long as ProductBeacon Research is published. Chapter URLs persist at their published location; we do not redirect, rename, or remove published chapters.

When a citation URL breaks, we replace it with the corresponding Internet Archive Wayback Machine snapshot, with the snapshot date visible. The claim does not change; the citation surface gets stabilized.

Snapshot discipline. Vendor product-page leads are evaluated against dated HTML capture taken at the moment a chapter enters research. Mid-research rebrands do not re-classify a vendor within the front.