According to hitwise: “share of US Internet visits to Twitter increased 24% on Friday, April 17, the day of Oprah’s first Tweet. Comparing visits with the previous Friday, visits were up 43%.” Oprah is well recognized for the impact she can have on a product or service and these numbers seem to validate that. I started wondering, however, if coverage volume (irrespective of the influence, UMVs or circulations of a given publication) can be proven to have some sort of effect on the growth of a service. I decided to try to reverse engineer Twitter’s growth by looking at coverage of Twitter since its founding and the rate of increase in unique monthly visitors to Twitter.com.
Through Compete I could get monthly UMVs for Twitter going back 12 months (I could have paid for 2 years, but I’m not going to do that). I decided to restrict the media search to the same time frame. I conducted the media search with Factiva, and searched for all articles in all types of publications mentioning the word Twitter – which means I probably picked up about 100 or so unrelated articles a month (based on a quick search I did covering years preceding Twitter’s launch).
Here’s what I found:
- Since April 2008 media coverage of Twitter has increased 2157%.
- Increase in Twitter traffic has increased 832% in the same time frame.
- The rate of increase of media coverage outpaced the rate of increase of UMVs to Twitter.com 9 out of the 12 months.
The following graph illustrates the findings:
So what does this mean? Um, probably not much. It does suggest that a highly covered service will have high traffic to its Website, but who really doubts that? It also suggests that the media is increasing its coverage of Twitter at a rate that outpaces Twitter’s growth, but I’m not completely comfortable with that conclusion as many people ‘experience’ Twitter without visiting the homepage.
I performed this analysis expecting to find that months of high Twitter growth would coincide with months of unusually high media coverage but that wasn’t always true. What I’m left with is the feeling that this type of analysis, if performed over enough years and enough companies, could begin to reveal some trends and tendencies that might allow us to prove a correlation between coverage and growth in traffic, but clearly this analysis is insufficient. It’s also worth noting that while Factiva does pick up a fair amount of online coverage, much was omitted. I did try using some blog search tools, but the results weren’t reliable enough to include in this post.