You signed up for Twitter, added a short bio, uploaded an avatar and are Tweeting regularly, but still nobody’s following you.

Now what?

The way most Twitter users (especially new ones) build a base of Twitter followers is by following people themselves. Lots of people follow-back people who follow them, so by going out and following people you should be able to accumulate a lot of followers.

I recently spent some time using data from Twitter Grader to test this assumption. I broke up the database into “buckets” of users based on how many users they’re following. If you’re following around 100 users, you’re in the 100-user bucket, if you’re following close to 1000 users, you’re in that bucket.

The graph below shows the number of users in each bucket (the red line) and the average number of followers the users in each bucket have (the blue line).

The red line indicates that most users aren’t following a ton of people, which is expected given that most users aren’t Twitter-addicts. The blue line, however, tells a more interesting story: People who follow lots of people tend to have lots of followers themselves.

Let’s look at little closer at the follow-back assumption. The graph below shows the distribution of Twitter users at each following to follower ratio.

We see that most users have close to a 1:1 ratio of following to followers, meaning that many users follow-back those that follow them.

So does that mean you should go nuts and follow tons and tons of people? To answer that question, let’s look at how your following/follower ratio is related to the number of people that follow you.

The graph below shows the average number of followers of users based on their ratio. A ratio of 0.5 means that you follow half the number of people that are following you, and a ratio of 2 means you follow twice as many people as are following you.

This shows that users with a low following to follower ratio tend to have a high number of followers. That means that if your goal is to build a Twitter account with lots of followers, and we assume these factors have some sort of causal relationship, you should try to keep your ratio near or under 1 (following the same number of people as follow you or less).

Conclusion

The data shows that the best way to build a robust Twitter account is via a stepped approach. Follow a few people (a few of them will follow you back), then follow a few more. Don’t go crazy following thousands of people. Do it slowly and build up your followers gradually.

We’ve all heard people say “I won’t follow you back if you don’t have a link and bio” etc. But do people really mean it? Does having a bio and website link in your Twitter profile really make a difference in the number of followers you have? The answer is yes.

I dugg back into the data we’ve collected from Twitter Grader. We have a bunch of different information on over 1.6 million Twitter accounts, including bio, link and follower data.

The first “aha” moment from this slice of data is the sheer number of users without bios and links. Almost two-thirds of users don’t have a bio or link listed on their Twitter profiles.

Looking at the average number of followers shows a very distinct trend. Users with a bio have over 8 times more followers on average than users without a bio and users with a link have over 7.5 times as many as users without.

Beyond sheer number of followers, power users are even less likely to follow you if you don’t have a bio and link. (Power users are Twitter users with high Twitter Grades.)

Users with a bio have over 15.5 times more power followers than those without a bio. Users with a link have over 22 times more power followers than without.

And of course your Twitter Grader score is effected strongly.

The bottom line here is that if you haven’t already specified a site and a bio for your Twitter profile (and lots of you haven’t) go do it now.

Ever wonder how often you should be tweeting? How much is too much, and why do some people seem to do nothing all day but sit around posting to Twitter?

One of the great things about tools like Twitter Grader is that they allow you to do all sorts of interesting data analysis. Our database includes stats on more than 1.6 million Twitter user.

The average tweets per day (TPD) I measured was 4.422 and the average number of followers for users in the database was 103.39.

Below is a graph of the distribution of those 1.6 million Twitter users’ average number of tweets per day. Notice that most users fall towards the low end of the range, meaning that most users only tweet a few times a day (or less).

What is more interesting than that is what happens when we the graph average number of followers of users at various TPD levels. As you can see by the graph below, a sweet spot emerges. Users who tweet between 10 and 50 times per day have more followers on average than those that tweet more or less frequently. The “peak” of the curve below is at about 22 tweets per day.

There are a small but siginificant number of users who tweet more than 150 times per day on average, but when added to the above graph they only continue the downward trend to the right.

At the extremely high-end of the spectrum the TPD metric shows top users at a variety of different posting rates. The blue line on the first graph represents the follower counts of the 50 most followed users, and the red line represents their tweets per day. The table at the end of this post is the tweets per day of the 20 most followed users in our database.

User		Followers	Tweets Per Day
	barackobama	316651	0.38
	CNNBrk	263730	0.81
	stephenfry	226854	8.06
	twitter	205758	0.37
	britneyspears	191126	0.53
	kevinrose	177949	2.72
	Nytimes	174635	39.04
	lancearmstrong	164864	7.85
	algore	156223	0.12
	ev	148709	3.67
	the_real_shaq	144201	3.48
	aplusk	143299	12.13
	nprpolitics	142835	11.84
	mashable	136364	19.65
	techcrunch	129105	9.64
	Veronica	129017	3.98
	ijustine	128965	7.33
	wilw	128160	6.16