--- title: "Get Connections" method: GET path: "/companies/{company_id_or_domain}/connections" tags: ["Connections Dataset"] --- # Get Connections `GET /companies/{company_id_or_domain}/connections` ## Request - Base URL: `https://predictleads.com/api/v3` - URL: `https://predictleads.com/api/v3/companies/{company_id_or_domain}/connections` - Auth: one of: - API key in header `X-API-KEY` and API key in header `X-API-TOKEN` - HTTP basic - OAuth 2 - Client credentials - Token URL: `https://oauth.predictleads.com/token` Returns a list of company's Connections. ## Path parameters - `company_id_or_domain` string, required ## Query parameters - `first_seen_at_from` string, date - `first_seen_at_until` string, date - `categories` union[] - union - 'partner': **Read as**: Company1 is a partner of Company2.
A partner relationship signals a collaboration between companies that goes both in two directions. You will find such connections using titles as "We work with", "Our partners", "The company we keep", and other titles similar in meaning. The implications of this category can be quite wide in their meaning, but in general all partnerships are a positive signal.
**NOTE**: Until February 2024 the `partner` category did not always signal a true partnership between two companies, but it also included a tighter knit relationship compared to the `other` category, which are not always as strong as the partner keyword suggests. The meaning is now correctly followed for all new cases or cases that have been last seen since and still have the `partner` category. - 'vendor': **Read as**: Company1 is a vendor to Company2.
The vendor category is much simpler in its meaning. It simply means that company1 is a supplier for company2 in some way. Some vendor connections signal a more positive relationship, depending on the source category. For example, a vendor relationship with a testimonial section/page source category is a stronger positive signal. On a webpage, such connections can be seen in lists such as "Our customers", "Trusted by", "Enabling businesses to", etc.
For example, to be a vendor to Microsoft, the company has to pass certain requirements as to company size, product reliability, safety, etc. This can be seen as a vetting process of some sort and increases the company's trustworthiness.
The vendor category can also be used to determine the supply chain of a certain company and evaluate supply chain risks. - 'integration': **Read as**: Company1 has an integration with Company2.
Most of the time an integration happens between a platform and a service.

Case examples: - 'investor': **Read as**: Company1 is an investor in Company2.
The investor category identifies relationships between companies making investments (Company1) and companies receiving investments (Company2).
The more investors the company has, the better. One could also track what companies/sectors competitors are investing in. - 'parent': **Read as**: Company1 is a parent company of Company2.
Parent category defines a parent-subsidiary relationship between websites. In some cases such websites are only localized versions for specific markets.

Case examples:
**NOTE**: Often it makes sense to not only check current hierarchy level e.g. "blizzard.de" but also check their parent connections e.g. "blizzard.com" and "activisionblizzard.com". Since sometimes company connections (customers, investors…) are also available at higher company hierarchical levels. - 'rebranding': **Read as**: Company1 is a rebranding of Company2.
Company1 being a rebranding of Company2 suggests that Company2 has undergone a process to change its brand identity. We detect rebranding based on domain change and similarities between the old and the new website.

Case examples: - 'published_in': **Read as**: Company1 is published in Company2.
The `published_in` category is usually found on early stage startup websites with fewer accomplishments on the market, but positive press coverage. You can see such connections under titles such as "As seen in", "Featured in", "Talking about us", etc.
A `published_in` connection is a positive signal, but after some months or years, we would expect to see them removed and replaced with other content proving the legitimacy of the company. - 'badge': **Read as**: Company1 has a badge of Company2.
The `badge` category shows that the Company1 has a certificate issued by the Company2 e.g. SOC 2 Type 2, ISO 27001, or some other common badges like GDPR, CCPA, etc.
Such connections may have a TechnologyLite reference present, describing the technology in some detail. Full information about that technology is available via the technology endpoint. - 'other': **Read as**: Company1 is connected to Company2.
This is the most general category. If we detected a relationship between companies, but we were unsure which category it belonged to, or it didn't belong to any of our categories, then we would categorize it as `other`.
Outgoing `other` category connections are a slightly positive signal, while incoming ones are quite more so. Thus a connection Company1 -> other -> Company2 is more positive for Company2, than for Company1. As Company2 was featured on Company1 website.
Sometimes the `other` category can have additional information via the source_category attribute depending on where it was found. An example of such a source_category would be `cookie_section`, which is explained further below. Such connections could be evaluated differently. - `page` integer - `limit` integer ## Response `200` Request was successfully processed. - ConnectionsDataset: Since 2019, PredictLeads has detected over 345 million business connections between companies globally. The Company Connections Dataset describes relationships between some 60 million companies.

Connections are sourced from different subpages on the website, such as Main, Customers, Partners, Integrations, Case Study, Testimonials, Suppliers, Portfolio, and identified through image recognition of company logos.

The Connections Dataset includes fields such as Connection Category, Source Category, Source URL, Context, First Seen At, Last Seen At, and other Connection information. - `data` Connection[], required (unresolved $ref) - `included` union[], required: Contains objects referenced in property `relationships`. - union - CompanyLite (unresolved $ref) - TechnologyLite (unresolved $ref) - `meta` object: Meta information about `ConnectionsDataset` object. - `schema_version` string, required: Represents schema version of objects. - `record_state` 'active', required: Represents record state of objects. - `count` integer: Total number of results.
**NOTE**: Only included in API responses when the request parameter `page` is provided, for performance reasons. ## Other responses - `400`: Invalid request body or parameters. - `401`: Authentication credentials are invalid or API subscription is not activated. - `402`: Exceeded the monthly request limit. - `403`: Request rate over limit. - `404`: Record not found. - `422`: Provided data failed to validate. - `429`: Too many requests. - `500`: Internal server error. - `503`: Service temporarily unavailable. ## Changes - **2026-10-09** `94475b6f17a8`: 2 warning - added the new `demo_page` enum value to the `data/items/attributes/source_category` response property for the response status `200` - added the new `registration_page` enum value to the `data/items/attributes/source_category` response property for the response status `200` - **2026-10-01** `113413aee4a2`: 1 info - api operation id `company_connections` removed and replaced with `get_api_v3_companies_connections` - **2026-09-01** `a56831a1efd4`: 1 warning, 2 info - for the `query` request parameter `categories`, the type of property `items/` changed from no type to `string` - added the non-success response with the status `503` - added the required property `data/items/attributes/confidence` to the response with the `200` status [Operation history](https://skmtc.dev/predictleads/apis/predictleads-api-reference/changes/companies/:company_id_or_domain/connections/get.md) --- [API](https://skmtc.dev/predictleads/apis/predictleads-api-reference.md) - [All operations](https://skmtc.dev/predictleads/apis/predictleads-api-reference/llms.txt) - [OpenAPI document](https://skmtc.dev/predictleads/apis/predictleads-api-reference/revisions/ccb9fec6a710?raw)