Dun & Bradstreet alternatives for company signals

This guide is for a data or revenue operations lead whose accounts are keyed to Dun & Bradstreet (D&B) records and who is now asked a question those records were not built to answer: what did this company do last week? It sets out what D&B covers according to its own pages, which sources fit each part of it, and how to put dated company signals beside D&B data while you keep the key you already use.

The two are built for different questions. D&B issues the D-U-N-S Number and holds company data sourced from public registries, websites and trusted partners, together with business credit scores, payment experiences, corporate family trees and contacts. It is built for verifying a legal entity, setting a credit limit and tracing legal ownership. Fokals is built for what a company does: the technologies it adopts and drops, the roles it opens, what it announces and files, and what those actions signal by topic, every observation dated and delivered on one company index.

What Dun & Bradstreet covers, as its pages state

D&B's D-U-N-S page describes the D-U-N-S Number as a nine-digit identifier that is unique to a business, and says other organisations may use it when they assess a business's credit profile, keep their own records, onboard the company or meet compliance duties. It is the company identifier at the centre of D&B's data: the D&B Direct+ page says the number powers what D&B calls its Commercial Graph, which connects business identity, ownership structures, hierarchies and relationships. If your procurement or finance systems already store it, that alone is a reason to keep D&B.

Its data cloud page says D&B draws its data worldwide from public registries, tens of millions of websites and partners it trusts, and states data on 650M+ entities. The same page lists business scores and ratings, trade payment experiences, mapped corporate family tree relationships, professional and consumer contacts, and 15+ years of historical data.

D&B's guide to business credit scores describes the PAYDEX Score as a measure of payment performance on a scale from 0 to 100. It describes the Failure Score as a prediction of how likely a business is, over the next 12 months, to seek legal relief from its creditors or to stop operating before all of them are paid in full, and the Delinquency Predictor Score as a prediction of how likely a business is to fall behind on its payments.

For delivery, D&B's master data page lists D&B Direct+ APIs, batch and flat-file delivery, and data-warehouse feeds into cloud warehouses including Snowflake, Databricks and BigQuery. The D&B Direct+ page names firmographics, hierarchies, ownership structures, relationships and risk indicators as the data that enriches your records, and says organisations reach D&B's Data Blocks through it. D&B Hoovers is its product for sellers, with company intelligence, contact data, news, trigger events and buying signals.

Why teams look for an alternative

The reasons are about scope and fit.

• A different kind of data. A registry record and a credit score say what a company is and how it pays. A question about what it did, such as a tool added to its website, a role opened or an acquisition announced, needs observations that each carry a date and a source.
• Company-level data throughout. A product, a client contract or an internal policy written for data about companies, with the people left to the systems that already hold them.
• Research use. A back-test needs daily and weekly rows that are written once and left alone, keyed to ISIN or FIGI.
• Licence fit. Terms for embedding in a product or for redistribution, agreed directly with the vendor.

Dun & Bradstreet and Fokals side by side

The D&B column repeats what the pages linked above state. The Fokals column follows the data dictionary.

Adding dated signals beside D&B records

You do not have to choose between the two. Keep the D-U-N-S Number as the key of the account record, and add a second source for what changed.

Fokals keys each company by one stable company ID, and the bridge from a D-U-N-S-keyed account is the website: Technology Stack holds each website domain beside the company ID. A listed company also carries its ticker, MIC, ISIN, LEI and FIGI, so an account table that stores an LEI or an ISIN can join on those instead. The glossary entry on domain matching covers how to normalise a domain before you compare it.

This query returns every dated signal from a day you choose, here 28 September 2026, for accounts held under a D-U-N-S Number. The account table is yours, and the other two are in the data dictionary.

with domain_map as (
  select distinct domain, company_id
  from company_technologies
)
select
  a.duns_number,
  a.account_name,
  s.observed_at,
  s.source,
  s.kind,
  s.topics,
  s.weight
from my_accounts a
join domain_map m
  on m.domain = a.website_domain
join company_signals s
  on s.company_id = m.company_id
where s.observed_at >= timestamp '2026-09-28 00:00:00'
order by a.duns_number, s.observed_at;

Each row is one company action from the Company Signals dataset, with the time it was observed, its source (the website, its domain records, the careers page, a regulatory filing or an announcement), its kind and its weight. For a number per topic, Intent Scores in the intent dataset holds a weekly score from 0 to 100 for each company and topic, across a curated taxonomy of buying topics in nine groups, with a surge flag and the five strongest signals as evidence.

Take Acme Robotics, an illustrative company. Its D&B record tells a credit analyst which legal entity it is, who owns it and how it pays. In the same week Company Signals might hold a platform migration on its website, a leadership role opened on its careers page and an announced acquisition. The two records complement each other: one identifies the account, and the other says what it did and when.

Five points decide how to read the join.

• Many records, one website. Several of your records can share a domain, for example a parent and its branches. They all match one company ID, so decide whether signals attach to each record or to the top of the family.
• Brands. A brand or subsidiary with its own website has its own company ID and carries the identifiers of its listed parent. Group on ISIN or FIGI to roll its signals up.
• Match rate. The index holds listed companies worldwide, the brands they own and verified private companies. Count the match rate on your own accounts first, and keep unmatched accounts in their own queue: an unmatched account is not the same as a quiet company.
• Baselines. A first observation sets a baseline and is never counted as a change, so every signal in the feed is a change that was observed, never an existing state mistaken for one.
• Signals beside credit. A fall in open postings or a restructuring announcement is a dated public record of change. How it bears on credit risk is for your model to judge, and the guide to company signals in credit research works through how to read such signals beside a credit file.

Other alternatives to Dun & Bradstreet

Each part of what D&B covers has its own kind of source.

• Legal-entity records. OpenCorporates offers company records that its site says are drawn from over 140 government registries and other official sources, to search on the site or to take by API and in bulk.
• An open identifier. GLEIF publishes the data behind the Legal Entity Identifier. Anyone can search it free of charge and without registering, and it is also available as downloadable files and through an API.
• Business credit reports. Creditsafe offers business credit reports with credit scores, credit limits and trade payment data, and a business data API named Creditsafe Connect.
• Web-sourced company profiles. Veridion builds company records from web crawling, registry ingestion and public data assets, and its delivery page lists APIs, batch files and third-party delivery. The guide to Veridion alternatives compares it with Fokals.

Where each fits

D&B fits when the question is who a business is: its legal identity, its place in a corporate family, its credit standing and the people to contact there.

Fokals fits when the question is what a company did: which technologies it runs and when each was adopted or dropped, what it is hiring for, what its actions signal by topic, and what it announced. Every dataset sits on one company index, each record carries the time it was observed, and daily and weekly datasets are written once after the period closes and are never revised. A back-test or a model reads the record point-in-time, exactly as it stood on any day it covers.

Fokals is delivered direct, by REST API and as bulk files, which you load into your warehouse with its own loader. The delivery page describes both routes. Licensing is by written agreement for internal use, embedding in a product or redistribution.

Frequently asked questions

The queries and code on this page are examples to adapt. Test them in your own environment before you rely on them.

What this page says about the products it names was checked against their public documentation on 4 October 2026. Product and company names are trademarks of their owners. Fokals is not affiliated with them or endorsed by them.