The revenues of a US private company are one of the most requested items of competitive intelligence. They are also one of the hardest to find because they are, indeed private. Unlike, say, in the UK, private US companies are rarely required to file financial reports. There may be some ways of getting these revenues through primary research; but even secondary research can help you gain an idea of a private company's revenues. Certainly secondary research can help triangulate what might be a sensible revenue range. The comments below come with the usual caveats about adapting to the particular target situation, but are useful guidelines nonetheless.
- Number of employees. It is sometimes easier to find out how many employees a company has, than how much money it is making. Employees can be counted leaving a building, their cars can be counted outside a building, and in some states companies have to file the number of employees, or the size of their offices. Using secondary research, employees have LinkedIn profiles - for example, among the biggest US companies, 15% of employees have LinkedIn profiles. Companies may also post team photos on their websites. And the revenues of a company per employee are generally a few hundreds of thousands of dollars per year ($500K for Bank of America, $300K for Union Pacific Railroad, $80K for Starbucks). Private companies outside Silicon Valley are more likely to be around the lower end than the higher end, and (in a related way) smaller companies will also be at the lower end. In general for private companies, $200K is a good ballpark.
- Subscription databases. Databases such as Dun & Bradstreet, Kompass and Hoovers (and a number of more random websites databases that show up on Google) have their own guesses as to what a private company's revenues are likely to be. Methodologies vary but usually involve calling the company and asking. Of course, slightly suspect, but usually better than nothing. Certainly they tend to be right within an order of magnitude.
- Website traffic. Services such as Quantcast and Compete can provide some idea of a website's traffic. This in itself can help shape a revenue range - a company with website traffic of a thousand visitors a month is unlikely to be turning over $50M a year. If the website is transactional, a further step can be added, using industry data for ecommerce conversion - assume that ~2% of a website's visitors convert into customers, and that online sales represent ~5% of the company's overall sales.
- Local news stories. Private companies that are otherwise reticent about their revenues are often happy to talk to local press about how they are benefiting their community, and in doing so mention revenues, number of employees, investment amounts and other data that can be used to derive revenues. Such local stories may need to be found directly through the newspaper's website, or even physical archives, as Google News does not always index them well.
- Photos of facilities. Private companies may display, on their websites, photos of facilities that can be used to derive a revenue range. With experience, it will become possible to differentiate the factory of a $5M company from that of a $50M company.
Thinking more broadly, a lot of a company's activity can give an indication of its size, which in turn can be interpreted in terms of revenues. For example, how many job postings are open (on its website or through Indeed), or how many followers the company has on social media.
It is rare for the above to provide an accurate estimate of a company's revenues, but this kind of triangulation is what a lot of secondary competitive intelligence is about.