Network Analysis: Mapping the Hidden Connections Behind Australian Corporate Structures

Most corporate due diligence begins with an ABN lookup. That tells you whether an entity is registered, what type it is, and whether it holds GST registration. It does not tell you who controls it, what other entities its directors have run, whether those entities failed, or what relationships exist between this company and the twelve other entities at the same registered address. Network analysis maps what a single lookup cannot.

Why a single ABN lookup is insufficient

The Australian Business Register contains data about one entity at a time. ASIC's company register is similarly structured — look up a company by ACN, get back its registration details and current status. Neither register, queried in isolation, tells you anything about the ecosystem of entities surrounding the one you searched.

Consider a typical Australian SME corporate structure. There is an operating entity — say, a Pty Ltd company — that contracts with clients and employs staff. Sitting above it is a holding company that owns the shares. To the side are one or two trusts through which the principals extract profit. Offshore or interstate might be another entity holding intellectual property. Each of these entities has a separate ABN and ACN. Each can be queried individually. None of them, looked at in isolation, reveals the full picture.

Network analysis requires cross-referencing multiple registers simultaneously — pulling director data from ASIC, matching ABNs across entity groups, identifying shared addresses, and mapping relationships between entities that appear disconnected until you look at the names behind them.

The data sources that make network mapping possible

Gumshoe draws on several overlapping datasets to reconstruct corporate networks:

• ASIC company register — 2.9 million companies, with registration dates, deregistration dates, and company type. Deregistered companies (status EXAD or DISS) are retained in the register indefinitely, which means we can identify entities that were once connected to a current entity and have since been wound up.
• Entity officers (directors and secretaries) — scraped progressively from ASIC company extracts, giving us a director-to-company mapping that lets us ask: for every director of Entity A, what other entities have they directed?
• ABN register — 16 million records linking ABNs to entity names, types, and GST status. Matching ABNs to ACNs allows us to bridge the ABR and ASIC datasets.
• Address data — registered addresses from both the ABR and ASIC. When multiple entities share a registered address — particularly a serviced office address — the density of that overlap is itself a signal.
• Banned persons register — individuals disqualified from managing companies or from providing financial services. When a current director of your target entity also appears in the banned register, that connection is immediately surfaced.

What the address signal reveals

One of the most consistently useful signals in network analysis is the registered address. Legitimate operating businesses tend to have physical premises at their registered address — or at least a clearly explained reason for using a different address (accountant's office, registered agent). Entities that register at serviced office addresses are not automatically suspicious, but when 50 or 100 companies share a single address and those companies are connected through common directors, the pattern is worth examining.

The Phoenix pattern — where a failed business is quickly followed by a new entity at the same address, with the same principals, doing the same work — is one of the most common fraud patterns in Australian contracting. The ASIC register contains enough information to identify it, but you need to look at the full network, not just the current entity.

Director connections: what they reveal

A director running two or three companies simultaneously is entirely normal in Australian business. A director who has held directorships in 15 companies over eight years, of which nine have been deregistered, is a different matter — not necessarily fraudulent, but a pattern that warrants explanation.

The most valuable question to ask about a director's history is not "how many companies have they been involved in?" but "what happened to those companies?" Directors of failed companies are not banned from forming new ones. They are, however, required to disclose material prior insolvencies in some contexts, and repeated insolvency involvement is a documented risk factor for future insolvency.

Where a director connection becomes immediately concerning is when it links your target entity to a company that has faced regulatory action — an entity that appeared on the ACCC enforcement register, whose directors were banned by ASIC, or that was deregistered as part of an insolvency process while owing creditors. That linkage does not need to be direct. A second-degree connection (director of your target was also a director of Company X, which was run by a banned person) is still material.

A worked example: the Meridian Group

Consider a fictional but realistic corporate structure we will call the Meridian Group. The operating entity is Meridian Construction Services Pty Ltd, which tenders for government contracts and employs site staff. Its ASIC registration is current, its ABN is active, and it holds a contractor licence. On a single-entity check, it passes.

Network analysis reveals a different picture. The sole director of Meridian Construction also directs Meridian Holdings Pty Ltd (which owns Meridian Construction's shares), Meridian Plant and Equipment Pty Ltd (which leases equipment to Meridian Construction at above-market rates), and was previously a director of Consolidated Earthworks Pty Ltd — which was deregistered following an external administration in 2021 with A$2.3M in unpaid creditors.

None of this is visible from the Meridian Construction ABN lookup. All of it is recoverable from the public record. The related-party equipment lease is the specific concern here: it is a mechanism by which profit is extracted from the operating entity (reducing its apparent profitability and assets) before creditors can access it. Combined with a failed predecessor company run by the same director, this pattern has appeared in multiple ASIC enforcement actions over the past decade.

Network flags that precede insolvency

Gumshoe's network analysis is calibrated to surface specific patterns that appear consistently in companies that subsequently enter external administration:

• Director of a recently deregistered company — particularly where deregistration occurred within 36 months and the company was wound up with creditors
• Shared address with deregistered entities — especially at registered agent addresses that also hosted entities that failed
• Related-party transactions with entities the director controls — visible when the network map shows multiple entities with overlapping ownership and the same registered office
• Rapid entity cycling — new entity registered while a prior entity with similar name and identical director is in external administration
• Director appearing on ASIC banned register — current directors cross-checked against the banned and disqualified register; any match is an immediate escalation flag

What Gumshoe Capital Intelligence maps for you

The Capital Intelligence network analysis goes beyond the standard verification checks. For each target entity, we map the full director network across all ASIC-registered entities — current and deregistered — and present the results as a structured relationship map with timeline data showing when each connection was formed and, where applicable, how it ended.

For acquisition due diligence, this analysis typically identifies three to five connections that would not be apparent from a standard legal and financial review. In our experience, roughly one in four private company acquisitions shows at least one network connection that, properly understood, would have affected the deal structure, pricing, or decision to proceed at all.

Uncommon Insights

One often-overlooked aspect of network analysis is the importance of deregistered entities in identifying potential risks. ASIC's company register retains deregistered companies (status EXAD or DISS) indefinitely, allowing for the identification of entities that were once connected to a current entity and have since been wound up. This information can be crucial in assessing insolvency risk, as linked entities in financial distress can pose a significant threat to the stability of the network.

Under section 180 of the Corporations Act, directors have a duty to act with care and diligence, and a failure to do so can result in ASIC taking enforcement action. Network analysis can help identify directors with a history of poor management or compliance issues, allowing for a more informed assessment of operational risk. By cross-referencing director data from ASIC with other datasets, such as the ABN register and address data, it is possible to build a comprehensive picture of a director's track record and potential liabilities.

The ATO's use of data analytics to identify high-risk entities has significant implications for network analysis. The ATO's use of machine learning algorithms to identify entities with a high risk of non-compliance can help inform risk assessments and identify potential red flags. For example, an entity with a history of late lodgments or unpaid tax debts may be considered high-risk, and its connections to other entities in the network should be carefully scrutinized.

ASIC's enforcement patterns also provide valuable insights for network analysis. ASIC's focus on enforcing section 588G of the Corporations Act, which deals with insolvent trading, highlights the importance of identifying linked entities in financial distress. By analyzing ASIC's enforcement actions and identifying common characteristics of entities that have been subject to action, it is possible to develop more effective risk assessment strategies and identify potential vulnerabilities in the network.