How many tea companies are there?

Several times, I have asked myself the question "How many tea companies are there in the United States?" or in the world, or in some other region. The answer to this is not straightforward. Manta.com publishes an interesting list of tea companies in the U.S., which currently has 927 companies listed. But browsing this list, it doesn't seem terribly selective and it looks likely to even contain some possible errors (unless Advance America Cash Advance serves tea with your payday loan).

But this list raises several difficulties about what exactly constitutes a tea company. Some questions that come up are:

• Do you count tea rooms? Or count some, but maybe not all tea rooms, depending on how "serious" they are about the tea, or whether they sell loose tea "to go"? And if you're going to count a tea room that serves tea but does not sell loose tea to take home, would you also count a bubble tea store?
  
• Is there a distinction between a tea shop that sells its own brand of tea, vs. one that does not? What if the tea shop does nothing but repackage and rebrand the tea with its own name?
• Do you count dropshippers, websites or mail-order outfits which "sell" tea only by placing an order through another company? And if you count dropshippers, do you count websites that set up a whole business website but just refer the sale through an affiliate link?
• Do you count a big company like Unilever which owns lots of brands, only once, or do you count its brands separately? Or...do you count some of them separately but not others? What about a brand like Red Rose, which has different ownership of the brand in different countries?
• At what point do you count a tea company as being "in" a given region? What about a company that is headquartered elsewhere but ships into the region? Or what about the numerous brands of tea companies which have no official presence in the U.S., but are available in specialty import stores? At what point do you start counting them?

There's no clear answer because there's no clear way to define the question. It's a lot like the old question: "How long is the coast of Britain?" This question is called the coastline paradox: it has no answer. Using different measures of width, and different standards about how to handle bays, rivers, and other estuaries, one comes to completely different answers, sometimes on the order of thousands of kilometers different.

Even if you set clear standards, there might not even be enough information available to accurately answer the question. For example, in many U.S. states, information about the ownership of private corporations is not widely available, so if you want to only count two brands once if they are owned by the same company, you will not be able to get a definitive answer. For example, until recently, there was much speculation about whether SpecialTeas was owned by Teavana. Although the evidence strongly suggested common ownership or a merger or a buyout of one by the other, there wasn't an unambiguously clear answer in the public record about what the exact relationship between the two companies was. You can see the evidence on RateTea's page on Teavana.

My answer about the number of tea companies:

From my research, my intuition is that, using most reasonable standards, counting all U.S.-based companies with their own brand, and all international companies with a more-than-esoteric presence in the U.S., excluding dropshippers and tea rooms without their own brand of tea, the answer to the question of how many tea companies there are in the U.S. probably numbers somewhere in the hundreds. How many hundred? I honestly have no idea. I think Manta's estimate, however, is probably too high as it seems to count a number of tea rooms that do not have their own brand of tea, as well as straight-up erroneous listings.

What do you think? How would you define tea companies to count them? And how many do you think there are?