A Beginner's Guide To Twitter Analytics

Without any doubt, Twitter is one of the most important social media tools available for online businesses. It has the power of providing an excellent opportunity to directly engage with potential customers in real time. I myself use it for answering queries and for connecting with new people. Using Twitter is one thing and quantifying the benefits of using it is another. Here there are no page views or bounce rate to fiddle with. All you have is a constant stream of tweets from different sources. So how would you assess and measure your performance on Twitter? Well, it may seem difficult at first, but with some simple techniques you can develop your very own Twitter analytics system. Following metrics can be used and applied to find out how well you're using your Twitter account. You can refine the methodologies presented here to get a more personalized analytics solution.

Amplification Rate

This is one of the most important and common Twitter metrics to measure how well your updates are being received by the Twitter ecosystem. Amplification rate implies how often your tweets are retweeted by your immediate connections and beyond. Different people measure it in different ways. I use the following formula to measure the amplification rate.

Amplification Rate = (Number of retweets) → Per thousand followers

This metric can be further refined or dissected to get a more granular view of how well and how often your updates are shared among Twitter users. You can divide amplification rate data into two different parts. One part contains the retweets from your immediate connections and the second one contains retweets from users other than in your followers' list. An increase in the latter group clearly signifies wider reach of your Twitter updates. There are several good retweets tracking tools that can be used to calculate amplification rate of your account.

Sentiment Analysis

This is an interesting metric that shows how people are reacting to your updates. It includes measurement of all the positive, negative and neutral reactions to your updates. Now measuring this is not an easy task and requires complex linguistic analysis supplemented by a refined heuristic engine. As of now, there's no killer tool to give you 100% accurate Twitter sentiment analytics data.

But there are several good sentiment analysis applications that can give you a fair idea about how people feel and react to your Twitter updates. At the very basic level, these tools try to find out tweets mentioning you in a positive, negative or in a neutral way. If you're actively helping people on Twitter, you can expect a good rating while calculating sentiment level for your Twitter account. This metric is closely associated with the engagement rate which we are going to discuss below in one of the sections.

Engagement Rate

This is my favorite metric and I try to figure it out to see how proactively people are engaging with me on Twitter. Again, accurate measurement of engagement rate is a bit tricky. Contrary to the general engagement rate calculation methods, I use the follow two ways to find out how good I'm conversing with people on Twitter. This includes real conversations devoid of simple retweets of your update.

Twitter conversations (having more than 2 tweets) per month

Twitter conversations (having more than 2 tweets + no links) per month

Although you cannot get accurate data through these methods, but you can definitely reach above 90% accuracy through them. The first method marks and counts all those conversations which have more than 2 tweets within them. Although several retweets receives 'thanks' tweet from the user who originally tweeted the update and this cannot be considered as a real conversation. Still the first method filters out lot of simple retweets from the data set.

The second method is more interesting and effectively picks actual conversations happening with people on Twitter. This method picks those conversations from your account where the entire conversation consists of more than two retweets and each of these tweets is devoid of any kind of hyperlinks. When people ask a question or help each other, they are generally not including links within their tweets. This method filters out these real conversations and helps you get a more realistic figure about engagement rate for your Twitter account.

Follow/Unfollow Rate

And last but not the least are the follow & unfollow ratios for your Twitter account. Once you master the art of Twitter marketing, you're bound to get more followers than ever before. This also includes good number of unfollows from random Twitter accounts. Good number of followers count reduction can come from bot accounts suspended by Twitter spam cleaning team.

As the name implies, follow rate has everything to do with your new followers. It is generally calculated as follows.

Follow rate = Number of new followers per 1000 tweets

This metric is more or less directly proportional to the engagement rate. The more you engage the more followers you get. You can refine or modify this calculation method depending on the volume of followers you get in a given time frame. If you're getting high number of followers, you can reduce the number of tweets in the calculation formula. Unfollow rate is equally important although nobody wants to see it everyday. If you're experiencing significant depletion in your followers count, you may want to calculate your unfollow rate to get some idea about why it is happening.

Unfollow rate = Number of unfollows per 100 tweets

Again, depending on your account condition and the unfollow rate, you can modify and refine this formula to suit your requirement. Unfollow rate get be high due to several reasons. You may have followed large number of spam accounts which may later be suspended from Twitter, or you're simply tweeting everything under the sun. This annoys most the followers and triggers unfollow action. Make sure you follow all good Twitter practices to keep your unfollow rate low.