Altsets Supply-Chain Data Methodology

Altsets Research uses supply-chain relationship data to study how companies depend on suppliers, customers, manufacturers, and other entities across time. This page explains how those relationships are assembled, why entity matching is difficult, what point-in-time preservation means, and how readers should interpret the results.

Altsets currently covers more than 90,000 unique entities, more than 400,000 unique relationships, and over 20 years of history. Coverage is broad, but it is not complete. The absence of a relationship from Altsets should never be interpreted as proof that no relationship exists.

What does Altsets Research measure?

Altsets Research examines connections between companies and other economic entities. Depending on the question, an analysis may focus on suppliers, customers, shared counterparties, geographic exposure, historical changes, or the concentration of multiple holdings around the same dependency.

A relationship is directional. One company may supply another, while the other company is the customer. That direction matters because the same relationship can have very different implications for each side.

For example, a customer may represent a meaningful share of a supplier’s revenue while the supplier represents only a small share of the customer’s costs. The reverse can also occur. Altsets therefore treats relationship direction and the perspective of each company as essential context.

The relationship graph, mapped connections, values, and metrics presented by Altsets are estimated research outputs. They should be treated as research inputs rather than audited transaction records, contractual disclosures, or a complete ledger of commercial activity.

Which sources contribute to the relationship graph?

Altsets draws from company filings, investor presentations, press releases, shipping manifests, and other sources that can help identify or characterize commercial relationships.

Each source type reveals a different part of the network:

• Company filings may describe major customers, suppliers, concentration, dependencies, or risk factors.
• Investor presentations may name partners, customers, manufacturing relationships, or strategic dependencies.
• Press releases may announce agreements, launches, supply arrangements, or commercial partnerships.
• Shipping manifests may provide evidence of physical trade between entities.
• Other sources may add context, aliases, dates, locations, or relationship direction.

No single source should be mistaken for a complete view of a company’s supply chain. Public disclosures are selective. Shipping records do not capture every form of trade. Announced relationships may evolve after publication. Different sources may also describe the same relationship using different names, dates, subsidiaries, or levels of specificity.

Altsets combines these fragments into an estimated relationship graph designed for research. The graph is not a claim that every relevant relationship has been discovered or that every discovered relationship has equal economic importance.

Why is entity matching a major part of the process?

The same company can appear under many names across different documents. A filing may use a legal entity name, a press release may use a brand, and a shipping manifest may use a subsidiary, abbreviation, local-language name, or shortened consignee name.

Entity matching is the long process of deciding when those references belong to the same underlying organization and when they do not. It can require reconciling:

• legal names and trading names;
• parent companies and subsidiaries;
• renamed, merged, or reorganized entities;
• abbreviations and document-specific shorthand;
• punctuation, spacing, and formatting differences;
• local-language names and translated names;
• inconsistent country or address information;
• common names shared by unrelated companies.

This work is especially difficult when documents transliterate Chinese names differently. Hanyu Pinyin and Wade–Giles conventions can produce different Latin-alphabet versions of the same name, and source documents can contain mixed conventions, spelling variations, or translation mistakes. A company may therefore look like multiple entities until its aliases, location, ownership, and surrounding context are reconciled.

Normalization attempts to reduce duplicate representations without collapsing genuinely different entities. It is necessarily imperfect. Some aliases remain unresolved, and some entities that appear similar may be kept separate when the available evidence does not support a safe match.

How are supplier and customer relationships represented?

Altsets represents relationships in a supplier-to-customer direction. The labels describe the role each entity plays within that specific connection, not the company’s entire business.

A company can be a supplier in one relationship and a customer in another. It can also participate through a subsidiary, contract manufacturer, distributor, or other related entity. Readers should therefore inspect the named entities and relationship context rather than assuming that a single edge summarizes the entire commercial arrangement.

When Altsets Research discusses a “shared supplier” or “shared customer,” it means that the normalized relationship graph connects more than one analyzed company to the same counterparty. That overlap can reveal a common dependency, but overlap alone does not establish materiality, exclusivity, substitutability, or a likely market impact.

What do the relationship metrics mean?

Altsets may display three estimated relationship metrics:

What is supplier revenue percentage?

Supplier revenue percentage estimates how much of the supplier’s revenue is associated with the customer relationship. It helps answer a supplier-side question: how economically important might this customer be to the supplier?

A higher estimate can suggest greater customer concentration for the supplier, but it does not prove that the relationship is contractually secure, exclusive, or permanent.

What is customer cost percentage?

Customer cost percentage estimates how much of the customer’s costs are associated with the supplier relationship. It helps answer a customer-side question: how economically important might this supplier be to the customer’s cost base?

A higher estimate can indicate a meaningful input relationship, but it does not directly measure replacement difficulty, operational criticality, inventory coverage, or switching time.

What is relationship size?

Relationship size is an estimated monetary representation of the commercial relationship. It provides another way to compare relationships, but it should not be interpreted as an audited purchase total, revenue disclosure, or contract value.

Each metric answers a different question. A relationship can be large in monetary terms while representing a small percentage of a very large company’s costs. It can also be small in absolute terms while representing a meaningful share of a smaller supplier’s revenue.

Missing metric values remain missing. They should not be interpreted as zero.

Why does point-in-time preservation matter?

Supply-chain research is vulnerable to hindsight. A relationship that is obvious today may not have been publicly visible several years ago. A company may later rename a supplier, disclose a customer, revise a presentation, or publish information that changes how an earlier relationship would be interpreted.

Altsets generally avoids retroactively rewriting historical results solely because later evidence becomes available. This helps preserve a point-in-time view: what the relationship graph represented using the information and processing associated with that period.

That choice has an important consequence. Historical outputs can differ from what a researcher might reconstruct today using all information now available. This is intentional. The goal is to reduce look-ahead bias rather than create a perfectly backfilled history with the benefit of hindsight.

Point-in-time preservation does not mean historical records can never change. Technical corrections, entity-resolution improvements, or clearly identified errors may require adjustment. When an Altsets Research article depends on a historical date, the article should state the data-as-of date and describe any material revision that affects its conclusions.

How should dates be read in an Altsets article?

Research pages may display three different dates:

• Published date: when the page first became public.
• Updated date: when the article, analysis, or presentation was materially revised.
• Data-as-of date: the historical point represented by the Altsets data used in the analysis.

These dates are not interchangeable. An article published today may analyze an earlier point in time. Updating the prose does not automatically change the data-as-of date, and updating the underlying analysis should be disclosed rather than silently replacing the original frame.

How should missing relationships and missing values be interpreted?

Missing data is not a negative finding.

If Altsets does not show a supplier, customer, percentage, or relationship size, the correct interpretation is that the item is not available in the analyzed Altsets output. It does not mean the value is zero, the relationship never existed, or the company has no exposure.

Coverage varies across entities, countries, industries, periods, and source types. Public companies that disclose major customers may be easier to analyze than private companies with limited public information. Physical-product relationships may leave different traces than software, licensing, distribution, or service relationships.

Research should preserve these gaps instead of filling them with assumptions. Charts and tables should distinguish unavailable values from measured zeros, and conclusions should be limited to the relationships and estimates actually present in the analysis.

Does a relationship prove operational dependence?

No. A mapped relationship can show that two entities are connected in the Altsets graph, but it does not by itself establish that one company cannot replace the other.

Operational dependence can be affected by factors that may not be visible in the relationship data, including:

• the availability of substitute suppliers or customers;
• product qualification and switching requirements;
• contract duration and termination rights;
• inventory levels and lead times;
• geographic redundancy;
• manufacturing capacity;
• pricing power;
• regulatory restrictions;
• the strategic importance of a specific component or service.

For that reason, Altsets Research separates network overlap from material dependency. Shared counterparties are useful candidates for deeper investigation. They are not automatic proof of concentrated risk.

What are the main limitations of Altsets Research?

Altsets Research is designed to make economic relationships easier to investigate, not to provide a complete or definitive model of every company’s supply chain.

Important limitations include:

• Source coverage is incomplete and uneven.
• Public disclosures can be delayed, selective, ambiguous, or later revised.
• Shipping records capture only certain types of commercial activity.
• Entity names and ownership structures can be difficult to resolve.
• Every relationship value and metric is estimated.
• A mapped relationship may not reflect current purchasing activity.
• Relationship presence does not establish exclusivity or replacement difficulty.
• Missing data does not establish the absence of a relationship.
• Historical point-in-time preservation can leave older views different from a hindsight-based reconstruction.
• Network proximity does not by itself establish financial materiality or causation.

Every research article should state its scope, selection rules, data-as-of date, and relevant exclusions. Readers should be able to distinguish what the analysis observed from what it inferred and what remains unknown.

Is Altsets Research investment advice?

No. Altsets Research provides estimated data, analytical frameworks, and examples for informational purposes. It does not recommend buying, selling, or holding any security, and it does not account for an individual reader’s objectives, financial condition, or risk tolerance.

Supply-chain relationships are one input among many. Company fundamentals, valuation, market structure, regulation, contracts, management decisions, and changing economic conditions can all affect how a relationship matters in practice.