Why “BloomReach?” BloomReaching the BloomFiltered Web

Today, BloomReach officially exits stealth mode and to mark this occasion, I’d love to share with you the story behind our company name. We wanted a name that would reflect the problem we’re trying to solve, connote positive feelings for the customer and at the same time wouldn’t cost us an arm and a leg.

But before I get into my story, I want to provide some context on Bloom filters—from which our name originates.

Bloom filter is a well-known data structure in the big data community. It is a space-efficient probabilistic data structure that is used to test set membership for an element. It can lead to false positives, but not false negatives–e.g. it says either “element is possible in the set” or “element is definitely not in the set.”

A common application for a Bloom filter is a check for password strength. To do this perfectly, one would have to load the contents of an entire dictionary in memory. Not only is this very memory inefficient, it’s just not feasible to do so on every browser. The standard way to solve this problem is to build a Bloom filter on top of all the words in dictionary and load them in the browser. Google Chrome uses a Bloom filter to store a set of malicious sites. Most search engines use Bloom filters to store lists of porn phrases, misspellings, bad pages etc.

But back to our story.

Back in early 2009, we were so busy working on the product and technology that we didn’t pay too much attention to finding a good name. We were in Boston, getting ready to pitch our company to investors, and a mere 18 hours before the presentation we realized that we need a name. We started scrambling and going over all possible and available names. Most good names were parked domains and something that we weren’t interested in.

Suddenly, we realized that as consumers of web content, we treat search engines and social networks as Bloom filters. When we perform a search on a search engine, any result outside of the top ten is considered non-existent; not many people go to second page. However, results that show up in the top ten may or may not be relevant to us. On a social network, if a shared item is not in your current news feed or Twitter feed, you assume it wasn’t shared. Not all shared items are relevant to you either.

As a web publisher, you don’t want to be hidden behind this vast sea of data. You want your content to bloom and reach every single person for whom it’s relevant. Voila!!  And that’s how the name BloomReach was born.