About a year ago I did a case study on the use of Twitter by BBGeeks.com – a site I own about BlackBerry phones. We were using Twitter as sort of a guerrilla marketing tactic to increase traffic to our site and more importantly, promote our brand.
The results were encouraging – we’d acquired 500 followers in a short period of time and found they were actually visiting our site and engaging with us as a result. The initial results were promising enough that we kept with it, kept learning, kept testing – and we’re very glad we did.
We went on building our brand via being an “online help desk” of sorts for BlackBerry issues, much like we did in the previous case study:
- One of the BBGeeks.com staff was the voice of @bbgeeks on Twitter.
- We continued in our belief that our goal should be for him to become a BlackBerry trouble shooter (.i.e. help people) first, promotional evangelist (i.e. drop links) for BBGeeks.com second.
- None of our tweets were automated, not even links to our blog posts – if we didn’t feel it was worth dropping by hand, we didn’t drop it.
- We continued to seek out BlackBerry users via various search methods.
We had built our following up to about 5000 people by February of 09. Our traffic from Twitter continued to grow from the time we’d last graphed it and we were starting to get excited about the possibilities even larger follower counts might bring:
In March of this year, we really “geared up” our efforts. We began to be more aggressive in seeking out followers:
- We began to seek out the Twitter accounts of our major competitors and follow their followers… if they were interested in the competition enough to follow them, chances were, they’d be interested in us too.
- We ran contests to get people to retweet about us to their followers.
- We sought out top twitterers we knew had an interest in BlackBerries.
- We continued to be extremely helpful to anyone would could, and in turn, those folks often told other folks about us when their friends had BlackBerry issues arise.
Within a month, our follower count had gone to five figures and over the next few months, it continued to rise steadily:
Now, if we go by Danny Sullivan’s theory that analytics under-counts Twitter traffic by as much as 500%, our actual traffic numbers from Twitter, in theory, may have more resembled the below during that timeframe:
At the time this post was written, BBGeeks has over 19,000 followers on Twitter. Based on internal data that I’m not willing to share (hey, you can’t expect the entire farm for free), I can tell you that Twitter traffic converts higher than any other social network we track to date, by leaps and bounds. Twitter traffic actually clicks on our affiliate links, on our contextual ads and uses our coupon codes (we run Twitter only coupon codes sometimes).
In addition to the direct revenue Twitter brings us, it also helps increase our brand, gives Google signals of our authority and trust and it has been directly responsible for many links we’ve acquired.
Twitter also helps us develop content that contributes heavily to our long tail search volume on the site from various search engines. As I mentioned in my last case study:
We get tons of content ideas from the various questions and problems we see our followers and the people we are following experiencing.
My theory is that if folks were looking for answers to specific questions and/or problems on Twitter, non Twitter folks were likely typing those same queries into search engines as well. So we started a dedicated section of our site we refer to as the Twitter help files.
In it, we round up the various questions we receive each month, the answers we’ve given and develop and entire post of “long tail” information that is helpful to search users. Additionally, the folks who help us out by asking us these questions get some promotion to our large readership as well.