How to Manage Parent Companies, Subsidiaries, Plants, and Buying Locations
One parent can own several subsidiaries, plants, brands, procurement hubs, and ship-to sites. Treating all of them as one account hides applications; treating every address as independent creates duplicate pursuit. This guide builds a practical account graph that preserves ownership, operating, buying, delivery, and service roles.
DEFINITION
What is this about?
Manage complex accounts as a graph, not a flat company row. Create separate records for the ultimate parent, legal entities, operating facilities, buying locations, ship-to sites, and service locations; connect them with dated relationships such as owns, operates, buys for, specifies for, pays for, or receives for. Score application fit at the operating unit and coordinate commercial action at the actual buying level.
KEY TAKEAWAYS
- Parent, legal entity, establishment, facility, and buying location are different records.
- Application fit usually belongs to an operating site; contracting and payment may not.
- Model dated relationships instead of copying parent attributes to every child.
- Deduplicate outreach at the buying-motion level, not just by website domain.
- Keep unresolved ownership or procurement links unknown until verified.
Who this guide is for
This guide is for B2B teams targeting multi-site manufacturers, groups, distributors, contractors, institutions, or public buyers. It is useful when records share a domain, several subsidiaries use one brand, headquarters purchases for plants, a plant buys locally, or a service team needs a different location from invoicing and delivery.
Contents
1 Define each account node 2 Model relationships instead of copying attributes 3 Put ICP evidence at the correct level 4 Resolve the buying and delivery path 5 Create hierarchy and deduplication rules 6 Govern changes and unresolved links 7 Use the account-graph worksheet 8 Illustrative example 9 Common mistakes 10 What this does not mean 11 Common questions 12 Sources and methodology
1. Define each account node
Use separate node types: Node Working definition Typical evidence Ultimate parent Highest resolved controlling organization in scope Public filing, official corporate disclosure Legal entity Entity that can contract, own assets, invoice, or employ Registry, filing, procurement registration Operating facility Physical site where production or service activity occurs Facility page, permit, filing, verified document Buying location Unit that sources, evaluates suppliers, or places orders for one or more sites Tender, purchase document, first-party procurement statement Ship-to site Destination that receives goods Tender, logistics instruction, verified purchase record Bill-to or payer Entity responsible for contractual payment Contract, registration, verified account document Service location Site where installation, maintenance, or warranty work occurs Service scope, equipment location, contract These roles may collapse for a single-site business. They often diverge in a complex group. The Census Bureau defines an establishment as a single physical location and distinguishes single-unit from multi-unit companies. County Business Patterns methodology BLS similarly contrasts an establishment, commonly a single economic unit at one physical location, with a firm or company that can contain multiple establishments. BLS QCEW Concepts These are statistical concepts; a commercial account graph adds procurement, delivery, and service roles not supplied by the statistics.
2. Model relationships instead of copying attributes
Represent edges with a type, direction, scope, start date, end date, evidence, and confidence state:
• parent owns legal entity;
• legal entity operates facility;
• buying location buys for facility;
• engineering center specifies for facility;
• payer pays for order;
• ship-to site receives for order;
• service branch supports facility;
• brand is used by legal entity. Avoid copying parent revenue, industry, country, or status into each facility as if observed locally. Inheritance should be explicit and limited. For example, parent ownership can inherit down; plant application cannot safely inherit from a parent description. Statistics Canada describes different Business Register statistical levels, including enterprise, establishment, and location. Statistics Canada, Business Register Data Its definitions are for statistical production, but they reinforce the need to state which unit a field describes.
3. Put ICP evidence at the correct level
Assign every criterion a required node: ICP criterion Required node Why Product application Operating facility or equipment population Use occurs at the operating site Relationship type Legal entity or commercial unit OEM, distributor, contractor, or end-user role belongs to a business relationship Parent strategic scale Ultimate parent Consolidated context, not local proof Site conditions Operating facility Environment and duty are local Procurement eligibility Buying legal entity or buying location The contracting route applies here Delivery feasibility Ship-to site Freight, customs, and receiving are destination-specific Serviceability Service location and operating facility Support promise depends on both For public filers, SEC Search Filings provides company reports and other filings. SEC Search Filings Use a consolidated filing to establish what it actually states, such as subsidiaries or facilities; do not infer a plant's application merely from consolidated segment revenue. Unknown stays unknown. If the company states that procurement is centralized but does not identify the buying entity, record the centralized relationship as confirmed and the exact buying node as unresolved.
4. Resolve the buying and delivery path
Ask six different questions: 1 Which site has the application or installed base? 2 Which unit specifies technical requirements? 3 Which entity approves or contracts? 4 Which unit issues the order? 5 Which site receives and uses the product? 6 Which location needs service and warranty support? Do not force one address into every field. The path can change by product value, project type, geography, or business unit. For U.S. federal awards, SAM.gov distinguishes entity registration and a Unique Entity ID; direct bidding requires the applicable registration. SAM.gov Entity Registration USAspending Advanced Search exposes award filters including recipient, agency, NAICS, product and service code, and geography. USAspending Advanced Search Together they can support public-buyer and historical-recipient research, but they do not replace current tender or account-level validation. The detailed people who specify, influence, approve, use, or block the purchase belong in the Buying Committees module. This graph establishes which organizational nodes those people represent.
5. Create hierarchy and deduplication rules
Use stable identifiers for every node and do not rely on names or domains alone. Recommended identity keys include:
• internal node ID;
• official legal identifier where available;
• legal name and former names;
• physical address with source date;
• website or domain as an attribute, not sole key;
• parent link and effective period;
• facility aliases and brand names. Deduplicate at three levels: Identity deduplication Are two records the same legal entity or facility? Buying-motion deduplication Do several application sites share one procurement event or buying location? Contact-policy deduplication Would separate records cause overlapping contact to the same organizational buying group? Set a company-stop rule that can propagate across related nodes when evidence supports the scope. Do not automatically suppress an entire parent for a site-specific issue, or keep contacting subsidiaries after an enterprise-wide stop.
6. Govern changes and unresolved links
Ownership and operating relationships change. Store effective dates and source dates. Trigger review when:
• an acquisition or divestiture is announced or completed;
• a facility opens, closes, or changes operator;
• a buying center is centralized or decentralized;
• a brand or legal entity changes;
• a tender identifies a different contracting or delivery unit;
• returned mail, a reply, or verified research contradicts the graph. Do not delete the old edge. End-date it and create a new relationship. Active research, suppression, account ownership, and service obligations may need migration. Use evidence states for each edge: confirmed, contradicted, unknown, or expired. A high-confidence parent link does not imply a high-confidence buying relationship.
7. Use the account-graph worksheet
Node register Node ID Type Name Address Legal ID Evidence Observed date State [ID] Parent/entity/f acility/buying/ ship-to/servic e [Value] [Value] [Value] [Source] [Date] [State] Relationship register From node Relationship To node Scope Effective period Evidence state [ID] owns/operates/buys for/specifies for/pays for/receives for/supports [ID] [Product/site] [Dates] [State] Commercial route Record application site, specifying node, buying node, contracting entity, bill-to, ship-to, service site, account owner, contact cap, and suppression scope.
8. Illustrative example
Illustrative example - fictional. Fictional Atlas Process Group has one ultimate parent, three legal subsidiaries, seven operating plants, two procurement hubs, and one shared service coordinator. Research confirms the target application at three plants: P1, P2, and P3. P1 and P2 are operated by different subsidiaries but both buy through Hub H1. P3 buys locally through its operating subsidiary. The commercial count is therefore:
• Application-fit facilities = 3.
• Distinct buying motions = H1 for P1 and P2, plus local buying for P3 = 2.
• Ultimate parents = 1. Treating the parent as one flat account would hide three application sites. Treating each plant as an independent buying account would create three pursuits even though P1 and P2 share one buying motion. The graph produces two coordinated commercial routes with three site-level fit records. If the team contacts a maximum of two relevant contacts per buying motion in the bounded test, the cap is 2 motions x 2 contacts = 4 contacts, not 3 sites x 2 = 6. This is a fictional policy example, not a universal contact limit or performance claim.
9. Common mistakes
• Using domain as identity: several entities share one domain or one entity uses several domains.
• Copying parent fit to plants: consolidated context becomes site-level application evidence.
• Counting every address as an account: ship-to, office, warehouse, and plant roles are mixed.
• Assuming headquarters buys: the actual procurement hub or local buyer is unverified.
• Deduplicating only legal entities: shared buying motions still receive duplicate outreach.
• Losing effective dates: an old ownership link appears current.
• Making unknown relationships definite: weak sources create false routing.
• Mapping titles before nodes: stakeholders are assigned to the wrong organizational unit.
10. What this does not mean
An account graph does not require a complex enterprise-data platform. A controlled table can work if node types, identifiers, relationships, dates, and sources are explicit. It does not mean every parent should be pursued globally or every plant separately. It does not establish individual authority, buying intent, or guaranteed access.
11. Common questions
What should be the primary account record? Use the node that matches the decision. Application fit is often facility-level; contracting may be legal-entity or buying-location level; strategic ownership may remain parent-level. Can one facility have several buying locations? Yes. Different categories, spend thresholds, projects, or business units can follow different paths. Add scoped relationships instead of selecting one universal buyer. How should acquisitions be handled? Record announced and effective dates separately when relevant, verify legal and operating control, end-date old edges, and review active motions and suppression before migration. Are establishments and legal entities the same? Not necessarily. An establishment is an operating/statistical unit at a location; a legal entity may operate several establishments or share a location with other units. Where do contacts belong? Attach a contact to the organization and location supported by evidence, then map the person's role in the buying group separately. Avoid attaching everyone only to the ultimate parent.
12. Sources and methodology
Official sources define U.S. establishment and firm concepts, Canadian Business Register levels, public company filings, federal entity registration, and federal award search. SellersGPT created the commercial node types, relationship edges, deduplication layers, and graph worksheet as editorial tools for ICP operations. 1 U.S. Census Bureau, County Business Patterns Methodology. https://www.census.gov/programs-surveys/cbp/technical-documentation/methodology.html. Accessed August 24, 2026. 2 U.S. Bureau of Labor Statistics, Quarterly Census of Employment and Wages Concepts. https://www.bls.gov/opub/hom/cew/concepts.htm. Accessed August 24, 2026. 3 Statistics Canada, Business Register Data. https://www.statcan.gc.ca/en/statistical-programs/document/1105_D16_V3. Accessed August 24, 2026. 4 U.S. Securities and Exchange Commission, Search Filings. https://www.sec.gov/search-filings. Accessed August 24, 2026. 5 U.S. General Services Administration, SAM.gov Entity Registration. https://sam.gov/entity-registration. Accessed August 24, 2026. 6 USAspending.gov, Federal Awards Advanced Search. https://www.usaspending.gov/search. Accessed August 24, 2026.
Sources and Methodology
- census.gov: How to Manage Parent Companies, Subsidiaries, Plants, and Buying Locations
- bls.gov: How to Manage Parent Companies, Subsidiaries, Plants, and Buying Locations
- statcan.gc.ca: How to Manage Parent Companies, Subsidiaries, Plants, and Buying Locations
- sec.gov: How to Manage Parent Companies, Subsidiaries, Plants, and Buying Locations
- sam.gov: How to Manage Parent Companies, Subsidiaries, Plants, and Buying Locations
- usaspending.gov: How to Manage Parent Companies, Subsidiaries, Plants, and Buying Locations
Methodology
This article separates cited official or first-party facts from SellersGPT calculations, examples, frameworks, and recommendations. Source links and scope are retained from the publication pack.
Limitations
This material supports planning and decision-making. It is not legal, financial, regulatory, or performance advice, and it does not guarantee leads, meetings, customers, revenue, or return on investment.