Maybe Twitter isn't that popular... | Hacker News Reader

Maybe Twitter isn't that popular...

blog.chrisricca.com

22 pointsChrisRicca30 comments

To think about it, this isn't all that surprising

Facebook has to many users that a lot of people use google to get to Facebook (some just blindly clicking on the first link: http://www.readwriteweb.com/archives/facebook_wants_to_be_yo...)

To me it seems like Twitter just hasn't spread to the computer illiterate crowd, so this kind of graph is unsurprising.

Navigational search definitely has a lot to do with the graph. I don't think that's computer illiteracy, though. That's just how the normals use the internet ;) We techies are the fringe.

http://cdixon.org/2010/01/22/techies-and-normals/

I see the navigational search as an indication of mainstream adoption.

"navigational search"... I like that

Side note: As much as a I cringe at using google instead of the URL bar when you know the site you want to go to, I find myself doing it far more often on my phone. For example, on a mobile device, it's easier to google for "Hacker News" than type "news.ycombinator.com"

Agreed. Not to mention countless other problems with using Google Trends numbers as a proxy for "market adoption".

On a related note, I've been on the partial Twitter firehose and I've only seen around 5 million unique users over a couple of months. Granted this is the partial firehose, but I'd still expect the number of uniques to approach the true number of active users over a long enough time period. I'm seeing comparitively few new users at this point.

One other researcher I've talked to has similar results, anybody else gathering data on this?

I'm on the several twitter streams, totaling to around 20% of all the tweets. During a month I've got 8 million active users.

Very interesting — and precisely what I would have guessed beforehand. Here in Norway I see a lot of talk all the time about Twitter in the media, but nearly none of my friends use it at all. The only time I heard _everyone_ talking about Twitter was when living in San Francisco last summer.

You can't just use search engine queries as a proxy for popularity, especially on something like a website. Is it possible that Twitter serves a more technically savvy portion of users that knows how to use the address bar in their browsers? Or that people are more interested in researching about Facebook than Twitter even if they use both?

Even if you're right on all of those things, the difference in the numbers is staggering, and what you're talking about can't really account for that big of a difference.

That's true--I think Facebook is more popular than Twitter, but not necessarily to the magnitude search query traffic would suggest.

Also, Twitter is disproportionately popular among mobile users, many of whom don't even access it via a webbrowser. I'm not sure it's exactly the same as tech-savviness, either; I'll see if I can look it up, but I remember reading that Twitter is particularly popular among non-traditional internet users, especially African-Americans, who have a mobile device but rarely use a laptop or desktop computer. Facebook, by contrast, is mainly a website.

here's something else to put it into perspective. The other day ubuntu was a trending topic. I understand it was a big release and all. I'm not a stats pro by any means, but I just can't see how a "mainstream" application with 75 million+ users could have ubuntu as a trending topic.

Overall, the graph is correct. But...

Twitter population is about 10 times smaller than of Facebook. Also, many twitter-ers simply use a desktop/mobile client and not do anything Twitter-related on the web.

There is a secondary effect with respect to the platform too. I made a startup that served the Twitter platform market. It is very, very small compared to facebook's platform.

Here's an actual link to the trend - http://www.google.com/trends?q=facebook%2C+twitter&ctab=...

If you sort the Rank by Twitter, you'll see Brazil is the only place where Twitter outranks Facebook, which I would attribute to the Orkut dominance that pre-dated Facebook in that country.

Graphing COMscore data vs media mentions would a better indicator of hype/popularity differential than google query volume vs media mentions.

Comparing to Facebook is like comparing to people who breath air. Facebook is ginormous, if you're 1/100th their size you are doing well.

And users are not everything (and people searching google are NOTHING). Twitter has had huge impact News and News gathering. Facebook is/has revolutionized the game industry (or created the casual games industry) POV.

Google Trends is an extremely poor source of data if you're trying to extrapolate any real-world meaning.

I don't know where this presumption comes from that Trends numbers are an accurate proxy for such-and-such metric. Even the more skeptical posts here seem to admit that Trends for "twitter" vs. "facebook" would mirror those services' market shares if it weren't for those net-illiterates googling for the Facebook login page.

Well, I guess I'm even more skeptical then them because even removing people googling to get to a homepage, I think the numbers are only loosely correlated to traffic/popularity in any meaningful sense.

On a side note, with google "open" about so much, why doesn't the google trends graph show absolute numbers rather than this arbitrary scale? I know they're trying to protect their true business model (search+advertising), but it would nice to share a little bit.

Almost all Facebook users use the Facebook.com web interface, while the majority of Twitter users use a third party client or SMS. Comparing them by search activity isn't reasonable.

Many underestimate the difference caused by Facebook using Real Names while Twitter uses @shortnames.

And unless you explain this in some detail, many will continue to do that.

Well, it sets the 'tone' of the site. Just like the conversation will be different with the same group of people depending on whether they're all dressed in tshirts or if they're all dressed in formalwear... in each case, they're the same people 'underneath', but the image each sees of others and each's self-image will differ.

There's an additional layer of mapping added when using shortnames - the user has to remember a whole list of "@donkey2001,Joe Smith" tuples; most people (and especially non-geeks) like and operate better with fewer layers of abstraction.

The usage of a full name on Facebook makes explicit the roughly one-to-one correspondence between accounts and actual-people that is expected by users and required by the service; Twitter users openly create multiple profiles with different names and even identities, and the service and tools encourage that. In that sense, Twitter has more of a sense of a masquerade or Halloween party and less like a business meeting or high-school reunion.

Each has its strengths and its place, and the longer I use both, the less I feel like there's direct competition between the two.

Call me racist and you'd be wrong, but urban blacks LOVE twitter and it has become a serious part of their everyday social lives. Breakups, hookups, gang wars -- it's on Twitter, and it's real.

You know I've noticed this. Every single time I am convinced to check out Twitter I found the great majority of the posts on trending topics to have profile pictures of black people next to them. It is definitely an interesting phenomenon.

I've noticed that also, but I suspect it's mostly how people use twitter, rather than the overall population of twitter users; i.e. black twitterers are more likely than white twitterers to post in large public hashtags as a sort of chatroom.

A study (http://www.pewinternet.org/Reports/2009/17-Twitter-and-Statu...) did find that 26% of blacks vs. 19% of whites use "Twitter or another status update service", but it also found a large correlation with age. Since black demographics skew younger than white demographics, my guess is that the 7% racial gap would disappear if you controlled for age.

I wouldn't attribute that to race but rather to it working well for the needs of a small subset of a demographic (trend setters) and as they use it more, others join in.

It would be interesting to see twitter networks broken down by demographic and an analysis of what each group uses it for. Some people use it to convey social information ("I'm in the checkout line at Walmart"), others as a stunted form of IRC ("Oh no, you did not just say that!") and most of the people I follow use it as a substitute for email where the primary purpose is to send links around.

If anyone feels compelled to do the research, I'd love to read your findings.

Reply on news.ycombinator.com