Market assessments
Market assessment methodology
The Axomap assessment agent interprets stored public company evidence for each category, then assesses the landscape as a whole. Confidence describes the strength of each interpretation.
Assess each category first. Then synthesize the landscape from category assessments and underlying evidence.
Use the same definitions across markets. Do not mechanically average category labels. Weight breadth and materiality. A strong signal in one category or one exceptional company must not determine the landscape result by itself.
Global rules
Use only supplied evidence and respect each indicator's time window.
Use insufficient_evidence with value: null only when the scope has no relevant usable evidence or when the available records are materially contradictory or unusable.
When evidence is sparse, uneven, concentrated, or incomplete but still supports a useful interpretation, choose the best label with low confidence and state the uncertainty in limitations.
Do not infer: - market maturity from company age, funding stage, or average company maturity - competitive leadership from funding alone - commercial traction from announcement frequency - market entry from database additions or founding year - product evolution from static product snapshots
Deduplicate evidence referring to the same underlying development.
Explanation style
Explanations describe the market pattern, not the evidence examples.
Do not mention company names, customers, products, funding rounds, partnerships, acquisitions, or specific events in explanations.
Write one or two short sentences that directly explain what is happening in the market.
Use concrete market language and straightforward sentences. Each sentence should convey a clear idea that a non-expert reader can understand.
Do not compare the selected label to alternative labels unless necessary to explain a boundary.
Avoid analytical-process language such as “the supplied evidence supports”, “the observed pattern favors”, or “more prominent than”. Avoid generic terms such as “systems” or “workflows” unless they have a precise meaning in context.
Keep caveats in limitations and specific support in evidence_refs.
Market development stage
Window: current state, mainly informed by the last 36 months. Older history may inform establishment.
Measures buyer adoption, market stability, and repeatability of business models. It is not average company maturity.
Forming: Products and buyer needs remain unsettled. Commercial validation is limited or isolated.
Emerging: Clear products and early buyer validation exist across multiple providers, but adoption remains uneven and repeatable demand is not yet demonstrated broadly.
Developing: Repeatable commercial adoption is visible across multiple independent providers and is no longer concentrated in isolated cases. Evidence should show recurring or repeated buyer demand across a meaningful part of the market.
Established: Durable buyer demand and repeatable business models are broadly visible. The market behaves as a recognized buying category.
When evidence sits between Emerging and Developing, prefer Emerging unless repeatability is clearly demonstrated across multiple independent providers.
Prioritize repeated customer adoption, recurring demand, and evidence across independent providers.
Do not treat isolated customer examples or large funding rounds as proof of market maturity.
Competitive structure
Window: current structure, mainly informed by the last 24 months.
Measures how competitive strength is distributed across the market.
Fragmented: Competition is spread across many providers without clear leaders. Competitive strength is dispersed rather than concentrated among a small number of companies.
Balanced: Several meaningful competitors hold relatively comparable competitive positions, with no single company or small group clearly dominating the market.
Concentrated: A small number of companies hold most of the competitive strength, with relatively few meaningful competitors outside this leading group.
Polarized: A small number of strong leaders coexist with a large group of smaller or earlier competitors, creating a clear gap between the top of the market and the rest.
Prioritize customer adoption, commercial scale, market presence, product strength, and evidence of leadership.
In very early markets, Fragmented is often the appropriate label when several independent providers are active but none has established clear commercial or market leadership. Exact market share, revenue, or comparable adoption data are not required to identify fragmentation.
Do not use insufficient_evidence merely because providers are early or commercial metrics are sparse. If the available evidence shows several active providers and no credible leading group, prefer Fragmented with low confidence.
Do not infer competitive leadership from funding alone.
A startup landscape is a selected sample, not a census. Avoid inferring market structure purely from the number of companies collected.
Current activity
Window: rolling 6 months.
Measures the breadth and significance of meaningful market change, not event volume.
Quiet: Few meaningful developments despite adequate coverage.
Moderate: Meaningful change exists but is localized.
Active: Substantive developments occur across several independent players.
Intense: Multiple consequential developments are materially reshaping the market.
Prioritize significance, breadth, independence, and recency.
Do not count routine releases, publicity, hiring snapshots, static observations, or duplicate coverage as meaningful change.
Capital momentum
Window: latest 12 months versus the preceding 12 months.
Measures whether venture funding is strengthening or weakening.
Dormant: Very little current venture activity.
Cooling: Funding strength or breadth has clearly weakened.
Steady: Funding activity is broadly comparable with the prior period.
Accelerating: Funding strength or breadth has clearly increased.
Surging: Funding has strengthened exceptionally and materially.
Consider verified rounds, amounts, stages, independent recipients, and recency.
Treat the absence of verified funding rounds in a period as zero observed venture funding. Funding rounds are generally well disclosed and should be considered one of the most complete event types in the dataset.
Do not use missing funding records alone as a reason for insufficient_evidence.
Use insufficient_evidence only when there is a specific reason to doubt funding coverage or the dates, amounts, or financing types needed for comparison.
Only mention incomplete funding coverage in limitations when there is specific evidence of a coverage problem, such as conflicting dates, duplicate rounds, unclear financing type, or known undisclosed amounts.
Disclose in limitations when one company or round dominates the funding signal.
Keep currencies separate where needed.
Do not double count extensions, cumulative funding announcements, investor-join announcements, or consortium totals.
Distinguish equity venture funding from debt and grants.
A filing date is not necessarily the funding date.
Product evolution
Window: rolling 12 months.
Measures the dominant direction of substantive product change.
Stable: No material direction change despite adequate coverage.
Iterating: Existing capabilities are being improved or deepened.
Expanding: Products are broadening into new capabilities, workflows, or use cases.
Converging: Independent products increasingly address similar needs.
Diverging: Products are becoming more specialized around distinct needs or strategies.
Use dated product launches and meaningful changes in capability or product scope.
Static product pages do not establish evolution.
Prefer the dominant pattern across independent providers.
Entry and consolidation
Window: rolling 24 months.
Measures whether independent participation is increasing or decreasing.
Expanding: New companies, or genuine category entries are increasing independent participation. Product launches, geographic expansion, or commercialization by an existing participant do not count as market entry. Only if it comes from companies in adjacent categories.
Stable: Neither entry nor consolidation clearly dominates.
Consolidating: Acquisitions or exits are meaningfully reducing independent participation.
Use documented launches, category entries, acquisitions, and exits.
A database addition is not market entry.
Founding year alone does not prove category entry.
Do not double count one acquisition.
Distinguish acquisitions from partnerships and asset purchases.
Confidence
High: Clear evidence across multiple independent sources or companies.
Medium: Supported, but evidence is uneven, incomplete, concentrated, or partly company-reported.
Low: The best available interpretation is useful, but important uncertainty remains because evidence is sparse, uneven, concentrated, or incomplete.
Use low confidence when the evidence supports a plausible best label but does not establish it strongly. Use insufficient_evidence only when no useful label can be inferred from usable records.
Landscape synthesis
Synthesize the landscape only after assessing categories.
Do not use majority voting or numeric averaging.
Weight the materiality and breadth of category evidence.
When categories differ materially, choose the label that best describes the landscape overall and surface the disagreement in limitations.
Do not promote the landscape because of one exceptional company, event, or category.
Do not assign Developing at landscape level when a material category remains Emerging, unless the stronger categories clearly represent most of the market's commercial activity and repeatable demand.
Output
Return all six indicators with:
status, value, confidence, explanation, evidence_refs, limitations
Explanations must remain concise, abstract, and market-level.