Technology: So how many people really use Twitter?

Like many marketing executives, I am forced to provide proof that dedicating resources to Twitter is worthwhile without noticeable ROI. It's a struggle many of us lose or by which we've at least been bloodied.

So in the process of putting together my argument for Twitter, I've spent a great deal of time researching data about the social networking site's users. This, too, has not been easy.

How many users does Twitter have? The data is not consistent.  And even if you have an agreed upon number, it's not clear really what that number means -- for example, how many of those users are dead accounts and how active are the remaining users?

I gave a presentation for our board a little over a week ago and indicated 18 Million monthly users in the U.S.  But how many are there total?  And how many have value?

In all, it's been reported that there are 50 Million registered users. Sounds great. But as a marketer -- and in an attempt to show value -- I need to determine how many of these users are, well, useful.

Several questions come to mind:

1. How many users are active?  If you aren't active, there's really no reason to count you.  Dead accounts provide no positive impact to the Twitter community.  Those who do not update should not be counted.  Even users who update with fewer than 10 followers provide an extremely limited reach -- therefore, they provide little value as followers.  So, let's define "active users."
• How many users logged in at least once during the past month?
• How many users have made at least a total of 10 updates?
• How many users have at least 10 followers?

2. So, we've eliminated the inactive users.  Now, of those who are "active," how many users are open to engagement?  If you are looking to build relationships -- either as a brand or as an active user -- you want people willing to engage in a conversation.  People may follow you, but many are bots or spammers who will not respond or retweet.  They are simply interested in spreading their own message.  While some may have "value" in an entertainment or other sense (ie, coupons), they provide marginal value to people looking for engagement.  Additionally, some "ego-centric" Twitterers have value in the information they can provide you (and that you may then pass on to others), but there is no reciprocation.  So for the purpose of furthering your message and interaction, they still provide little value.  In other words...
• How many users reply to updates?
• How many users retweet?

3. Number of followers should now be reassessed since totals are inflated by bots, spammers and others unwilling to engage.  Why eliminate these people?  As a brand, I want to communicate with followers who will spread my message to others willing to perpetuate that message.  If you are not followed by such people, your value to me as a marketer is close to zero.  And, to be honest, you also have no value to me as a personal user.

4. Now that we have eliminated those who provide us no value, what is left? How many are there, and what are their characteristics?
• Average followers?
• Breakdown of percentage of active users with number of followers?
• Average number of tweets per day?
• Breakdown of percentage of active users with each number of daily tweets?
  

So, these are the questions I inevitably want answered.  In the process of research, I found some very interesting data -- some directly or indirectly related.  Let's plow through it.

• A data visualization showing that, if you had 100 Twitter followers, they'd be broken down as follows:  20 dead accounts, 50 haven't tweeted in the past week, five with more than 100 followers, and five creating more than 75% of the tweets.  To me, it's really those five to 10 whom I care about.
  
• The top 10% of Twitter users create 90% of the content.  For typical social networking sites, the top 10% create 30% of the content. I understand.  They're different.  But Twitter relies an awful lot on that 10%.
  
• 60% of Twitter users abandon their account within the first month.
• Twitter's unique users dropped in September to 23,538,791.
• Over 75% of users have 10 or fewer followers, with as many as 80% reported having fewer than 10.
• 38% of users have never sent a single tweet.  Over 75% have sent fewer than 10 tweets -- reported to be as much as 78%. 
  
• Average Twitter user has 42 followers.
• One in four registered users tweets in a given month.
  
• More than 29% of accounts have 0 followers.

I'm not trying to go out of my way to bash Twitter.  Understand that what I want to do is determine the number of Twitter users of value who we'd have the potential of involving in a conversation.  The data I discovered today doesn't necessarily provide an exact number, but rather a road map to finding a ballpark.

The problem, of course, is overlap.  How many of the users with fewer than 10 followers made fewer than 10 tweets?  Probably a lot of them, but no way to know just how many.  And these numbers also don't consider what happens when you reassess the follower numbers -- ie, eliminate inactive followers from those totals, thereby increasing the number of users with fewer than 10 "active" followers.

A safe bet?  Of the 50,000,000 Twitter users, 85% are of little to no value to a marketer or individual looking to engage others in conversation.  This isn't taking into consideration whether the remaining 15% has any interest in your conversation -- these are simply those open to engagement.

We are now left with about 7.5 Million users.  Still a decent number of people, but the pool is shrinking fast.

While it is a large number, 7.5 Million is also incredibly modest when compared to Facebook's 300 Million active users.  Even if you want to argue a small percentage of those Facebook users have value to a marketer...  well, it's an uphill battle.

The Twitter proponents will point to the success stories, regardless of numbers.  There are success stories, no doubt about it.  And there are also valuable ways to use Twitter.  But the question becomes just how valuable, without ROI, and just how significant the impact can be expected for typical users (not Fortune 500 companies and celebrities) who use it properly.

And since we can't measure success in ROI, the success needs to be measured in interaction -- not followers.  Followers, of course, often reflects a misleading number.

Twitter is a nice tool, but it is important to peel away the layers of hype to get to the truth of its active user base.  Additionally, if participation truly has plateaued, we may need to wonder if the tool has peaked in its usefulness.

What do you think?  How many useful accounts actually exist on Twitter, and how do you measure "usefulness?"  Does it matter?