As you might have noticed I did a little Twitter Survey this morning. It wasn’t for any specific project and in fact was based entirely on several conversations I’d had in the past few weeks combined with simple curiosity. There wasn’t a ton of background, but everything on it I’d talked with someone about recently so I felt the need to include.

One thing to point out, two of the questions I asked required people to sort things in order of most to least in terms of how that related to them. Several people told me there was a bug with this because they could only list one thing in a column. This wasn’t a bug, it was by design. The goal of the question wasn’t about each item on it’s own, but rather how it related to the others. You can only have one most and one least, so people had to chose one that stood out on top most of all, as well as which one was the very least likely, even if they had never done several of them. It’s kind of a trick, but I wasn’t as interested in any specific answer as much as how they stacked against each other. If the results came back with 3-4 things all at 100% for least that would have been useless to me, so I forced a ranking on purpose though the rank itself is kind of meaningless. Knowhatimean?

That said, I’ll get on with the results. Some nerdy data for you to start – I had a little over 5100 followers when I started this, so that’s how many people got my tweet. I used a shortlink from which tracks click throughs and 282 people followed that link. Interestingly enough 142 of those people were on Firefox, 43 on Safari and only 17 on IE. 101 of those people actually took the survey. I used a free survey from Survey Monkey which allows for 100 responses before automatically closing, so I guess there’s an error allowance of +/- 1. Also, I gained 14 followers over the course of this. SCORE! Anyway…

#1: What is the reason you decide to begin following someone on Twitter?
45.3% – I know or have just met them in real life
41.0% – They make me laugh or provide useful information

That was really the point of this question. “Hoping they will follow me back” “I’m a bot” “They have a cool avatar” and “Research on someone you plan to meet” all ranked extremely high in the Least category as well. Additionally “Someone retweeted something they said that was interesting/amusing” ranked fairly high on most, with 28.3% in the 2nd column second only to the standout Most winners listed above.

#2: Which statement is more accurate?
43.9% – I use twitter mostly to stay in touch with friends
38.8% – I use twitter mostly to follow people I don’t know in person
17.3% – I use twitter mostly for business purposes

#3: If you could only follow one more person ever, would you choose:
42.0% – Someone who is consistently informative
23.0% – Someone who is consistently funny
23.0% – Someone who tweets all the time about many different things.
12.0% – Someone I know, but who might not be that funny or informative.

#4: How often do you use the Direct Message feature?
94.0% – Less often then I publicly tweet.
4.0% – About 1:1 with my public tweets.
2.0% – More often then I publicly tweet.

#5: Assuming you use the Direct message feature, which is most often the reason why you choose to do that instead of @replying in public?
87.6% – Content of message is personal and not for the public.
7.2% – Worried that I talk to someone too much and will annoy people who follow me by publicly @replying to them so much.
5.2% – Trying to keep my daily # of tweets under a certain amount.
(other answers got zero votes – however it should be noted I didn’t just make those up, I’ve heard them all as actual reasons so it’s just safe to assume those represent an exceptionally small % of people)

#6: What is the reason you decide to stop following someone on Twitter?
28.7% – They tweet too much
27.4% – They frequently tweet about things I’m not interested in
17.6% – They are boring
8.5% – Most of their tweets are @replies

36.4% – They swear or say things that are NSFW too often
21.5% – They don’t follow me in return
16.0% – Stopped being friends with them in real life

Interestingly enough “Stopped being friends with them in real life” also ranked high as a reason people stopped following someone, but “They never @reply to me” was ranked rather low. “They expressed a viewpoint I don’t agree with” also scored consistently low, just not the very lowest.

#7: True/ False
Question / True / False (majority is bolded)
I use twitter more than email / 37.4% / 45.5%
I use twitter more than instant messenger / 63.0% / 30.0%
I access twitter mostly through 3rd party apps / 46.0% / 46.0%
I send messages to twitter mostly via SMS / 11.0% / 80.0%
I read people’s twitter pages who I don’t follow / 54.5% / 40.4%
I subscribe to the twitter RSS feed for people I don’t follow / 8.1% / 86.9%
Tweets are better when they are all content and no links / 19.0% / 59.0%
My follower count is very important to me / 16.0% / 71.0%
I won’t follow someone who doesn’t follow me back / 3.0% / 94.0%

Additional Comments: 25 people added something here, most were just shout outs (frequently to someone who retweeted my tweet) or complaints about one question or another, but here are a few of the interesting ones:

“I subscribe to RSS feeds of search.twitter results a lot.”

“i will follow people that talk about politics but i do not follow/ and will unfollow people that talk about their religion all the time.”

“People tweet too much about stupid stuff like, “I’m going to sleep now.””

“If I could change one thing about Twitter, it would be to group people into categories – by city, friends that I don’t want to stop following because we like to DM, but don’t want in my stream, etc. so I can turn off certain categories when I need to”

“the ranking questions UI is almost as difficult as sudoku puzzle. there should also be questions: “T/F : people who think twitter is their blog are annoying”, “T/F : people who think twitter is dodgeball/brightkite/foursquare/loopt are INCORRECT.”, “T/F : twitter is not IRC/AIM” (all true, btw)”

“that @seanbonner guy is really annoying with his tweets! ;)”


One thing to keep in mind with all of this is that these answers are not really indicative of everyone who uses twitter but rather people who people use twitter AND have chosen to follow me, or follow someone who follows me. I point that out because some of the above match what I would have guessed, but are also things I’ve talked about in public which might have attracted people who agree with me. It might not have, but it’s something worth noting. It would be interesting to see these exact same questions asked by 2-3 other people with entirely different follower circles.

I’d talk about what was the most interesting part to me, but really with only a handful of questions I didn’t ask anything I didn’t think was interesting. I was pleased to see people follow people that they know and that are informative and rarely because of research or hoping for reciprocation. It was cool to see that folks rarely dump people because of the things they don’t agree with or NSFW content, but rather if it’s interesting or not. I don’t know that this answered any real big questions so much as it gave me some more things to think about, which was kind of the idea. Hope you dug it!