Request an Esitmate
Send Us an Email
RSS FeedFollow Us on TwitterFollow Us on TwitterSubscribe to the E-Blast  Follow

Twitter Statistics

Being Information Driven: Understanding the Meaning of Data

October 18, 2010
The fundamental question with reach and impressions is if people are even seeing the tweet.
More than any other data type, Twitter statistics should be viewed with extreme caution. There are many factors that contribute to the questionable reliability of Twitter metrics. Social media, more than any other marketing tool you use, should be seen from a perspective of goals and conversion rates.

Yes, it’s that word again, longitudinal. When looking at your data, you always want to look at it over time. You want to take a long term look at your data and then trends and meaning will begin to emerge. With social media statistics, taking the longitudinal view is very important.

We’ve been looking at social media metrics for a while, both in general, and in specific by looking at how our clients use social media and the results they are achieving. If we look at raw numbers for our clients we find a range of traffic brought to their websites by social media in the 1% to 6% range. These numbers have not changed since our article last year “Social Media: The Next Big Thing...Effort.”

What Makes Twitter Statistics So Unreliable
1. Tools: There are many tools you can use to monitor your twitter statistics, however they are questionable at best. Many of the tools don’t work, or don’t work reliably, or don’t work often.

2. Lack of a Paradigm: In researching this post we found that there really aren’t that many people willing to take a stand on what we should be tracking, how we should be tracking, and what the numbers mean.

3. Artificial Inflation: The numbers that the current tools present have serious issues with inflation. The way in which twitter works, and the tools that people use with twitter leads to an enormous inflation of numbers.

Twitter Statistics By The Numbers
There are several key metrics that can be tracked and have been suggested as a means for discussing twitter statistics. Among these are:

1. Reach
The number of unique people who saw your message.

2. Frequency
The number of times an individual is exposed to your message.

3. Impressions
Essentially, reach x frequency

With retweets you know that not only did someone read your tweet, but they liked it enough to share it with their followers.

5. Click Throughs
Assuming that 140 characters isn’t enough to convey your entire message, tracking how many people went to your website after viewing your tweet.

Why The Numbers Are Less Than Reliable
The thing that makes twitter and other social media so great is precisely what makes the numbers so much nonsense. Combine this with the way that users interact with twitter and how twitter tools like tweetdeck are used and you can quickly see that the numbers should be taken with a grain of salt.

For example, the strength of social media is that you can tell a bunch of your friends and they can tell their friends who then tell their friends. So in theory, you can reach a wide number of people very quickly.

Add this multiplying affect to the multiplying affect of constant refreshes of your twitter feed, and your numbers can go up exponentially. As your post is loaded with 10, 20, 50, or even hundreds of other posts, your numbers can go up and up.

The real question isn’t even if people are reading, comprehending or caring. The fundamental question with reach and impressions is if people are even seeing the tweet. When I fire up my copy of tweetdeck, instantly a hundred, or even more people get an impression. But how many of those tweets did I read? I go for a cup of coffee, take a phone call and tweetdeck pulls in another 50 or 100 impressions. How many of those did I actually read?

You can see how relying on these numbers is tenuous at best.

A Real World Example
We recently encountered some twitter numbers that indicated almost 1 million people saw a series of tweets with 10s of millions of impressions.

When we looked at the actual click through to their site the number totaled 331, which amounts to .0034% of impressions resulted in people going to their website.

Now this is an extreme example, but it does point to what we are illustrating here. Without knowing what your goals are, without having an actual conversion target to map against, it is very difficult to look at a set of twitter statistics and derive meaning.

What To Do About Social Media Statistics?
At the end of the day it can’t be denied that social media is an increasingly important part of the landscape. It also can’t be denied that at times it can be an extremely powerful tool. However the real issue is how to make use of this tool for you.

Our recommendations are:

1. Look at your social media statistics longitudinally. The answers and trends will become apparent over time.

2. Identify quantifiable goals and results you would like to achieve. Whether you have 10 million impressions or 20, if you made your sales goals then you know what you are doing is working.

CRM Customer Relationship Management involves managing relationships with current, former and prospective customers. The web is one of your primary contact points with your customers. Shouldn’t your website be a part of your CRM strategy? At GeekPAK we think it should, which is why we built our CMS from the ground up to do just that.
T:  1.888.GEEKPAK (888.433.5725)
F:  718.836.0509
346 N. Justine
Chicago IL 60607-1010
564 72nd Street
Brooklyn NY 11209
6031-F Curtier Drive
Alexandria VA 22310