Alternative data: what investors read beyond filings and results
Quarterly results arrive weeks after the quarter is over. Alternative data is what analysts read in between, from footfall to what a company emails its customers.
Alternative data is information investors use beyond financial statements, company filings and press releases, such as credit card spending totals, satellite images or data from phones and apps. Funds use it to read a company between its quarterly results. In a 2022 risk alert, the US SEC’s examination staff said advisers should check the terms under which each dataset is collected.
On this page
What counts as alternative data
The US Securities and Exchange Commission’s examination staff gave a working definition in an April 2022 risk alert. It covers many kinds of information now used in financial analysis beyond traditional financial statements, filings and press releases, with these examples:
- Satellite and drone images of crop fields and retailers’ car parks.
- Aggregated credit card transactions.
- Social media and internet search data.
- Location data from consumers’ phones.
- Email data from apps and tools that consumers use.
What these share is timing. A fund sees a retailer’s footfall or a brand’s card spending this month, weeks before the company reports.
What diligence on an alternative dataset asks
The same alert said some advisers hadn’t assessed the terms, conditions or legal obligations attached to how a dataset was collected or provided. Some also had no way to decide when diligence should be redone after a provider changed how it collects. The alert sets out staff views with no legal force of its own, and it notes that alternative data doesn’t necessarily contain material non-public information. This is general information, not legal advice.
In practice a fund’s compliance team asks each vendor where the data came from and on whose permission, and keeps asking as the vendor’s methods change.
Alternative data on what consumer companies offer customers
One gap in most datasets is what a company shows only to its own customers. That includes how deep the discounts in its emails run and how often it sends them, its member prices beside the public ones and the offer when someone cancels. Those facts change week to week and sit behind a login wall, out of reach of search engines.
Obsession collects what companies show only to customers, step by step: their screens, emails and texts, member prices, trial limits, quotes and cancel offers, each dated with a screenshot. Teams and AI agents look up any company through the API or MCP, and a company not yet covered is added on request. Every email it keeps was sent to an account its own agents opened, so each series is point-in-time data from its first day. Funds and analysts read it as consumer promotions data.
Traditional data is what companies and markets publish: financial statements, filings, prices and press releases. Alternative data is gathered elsewhere, from transactions, devices, the web and what companies send their customers.
Not necessarily, as the SEC’s 2022 risk alert says. The alert asked advisers to have written policies for that risk and to check how each dataset was collected. This is general information, not legal advice.
A backtest can use only what was knowable at the time. A dataset that records the date each fact was seen, with its evidence, avoids look-ahead bias.
Point-in-time data, definedSources
In the order the page cites them. Every page was read on 11 Nov 2026.