Alternative data is information from outside a company's financial statements and market prices that investors use to research it. Examples are job postings, web traffic, card transactions, satellite images and shipping records. Its value is timing and detail, and each source needs checks of its own for history, coverage and legal sourcing.
How it is used
Fundamental analysts use it to follow a company between reporting dates, for example by watching hiring data for a change in where it invests. Systematic funds turn it into signals and test them on history. Credit and private-markets teams use it to monitor borrowers and targets that report little.
All of them map each record to a security or a company, compare the series with reported figures and decide whether it adds anything to what the filings and prices already say.
What to check before you rely on it
- History. A signal needs enough periods to test, and a short history limits what a back-test can show.
- Mapping. Records must join to your security master on stable identifiers.
- Timing. Every value needs the date it became known, or a test suffers look-ahead bias.
- Coverage. Find out which companies and countries the data favours, and which it misses.
- Sourcing. Ask how the data was obtained in a due diligence questionnaire, and whether it could carry material non-public information.
What it cannot do
Alternative data measures activity around a company, not its results. A rise in postings or a new tool on a website can come before a reported number, but it can also be noise, so each series needs a test against reported figures, using only data that existed on each date, before it is trusted.
In Fokals data
Fokals data is collected from first-party company sources and public records and processed in-house: postings, tools added to websites, filings and announcements. Every observation is dated, written once and never revised, so each series is point-in-time by construction, and a listed company carries its ticker, MIC, ISIN, LEI and share-class FIGI. The investors page sets out identifier mapping and point-in-time design, and the sourcing statement sets out what is collected and how.
Related terms
- Hiring data: job postings as a dated signal.
- Point-in-time data: data stored as it stood on each date.
- Material non-public information: the main legal risk of a source.
- Due diligence questionnaire: how a buyer checks a vendor's sourcing.
Frequently asked questions
What are examples of alternative data?
Common examples are job postings, website and app traffic, credit and debit card transactions, satellite and aerial images, shipping and cargo records, web-scraped prices and product listings, and corporate announcements read at the source. What they share is that they reach the analyst from outside the financial statements, often sooner and in finer detail.
Is alternative data legal to use?
Whether a dataset is lawful to use depends on how it was collected and what it contains, not on its category. The risks lie in the sourcing: personal data the buyer has no right to hold, material non-public information, and collection that breaks a duty or a contract. Funds ask vendors to describe their sources in writing before they use a dataset.
How is alternative data different from traditional data?
Traditional data in investing means financial statements, regulatory filings and market prices. Alternative data comes from the activity around a company: what it hires for, what it adds to its website, what customers buy. It is usually less standardised, so more of the work lies in mapping it to securities and in testing whether it adds information.