Friday, 10 April 2015

Shape Your Tweets with Twitter Analytics

The analytics package from Twitter can be used to increase the engagement and efficiency of your tweets, giving them more reach and engagement. Oh, and they’re free which is nice. That’s why it’s just not good enough to issue lots of tweets. It’s now essential to issue messages that are relevant and resonate with your followers so that they want to engage with your company.

The Service Basics
The service gives valuable insights as to how tweets are performing, instantly and the trends over time.
A key parameter is ‘Engagement Rate’ - simply the number of engagements a tweet has had divided by the views (twitter call them impressions) a tweet has had.
Tweets can be tracked in real time and messaging adjusted by the minute, hour or day.
It provides basic graphs and results on retweets, favourites and link clicks (maybe a link to an article on your website for example).
All of the above can be tracked by month, last 7 days, last 28 days, or you can customise a time frame.
By clicking on an individual tweet you can see detailed information for that tweet, such as time of day it was viewed (very useful this) and the link clicks, profile clicks, retweets and favourites. 

Does it Work?
My consultancy runs twitter accounts for a number of clients and the information provided by the analytics has proved to be extremely useful. The frequency, content and time of day messages are issued have all been shaped by reviewing and interpreting the reports.

That last point is key: understanding why something is working or not, so that we can do more of it if it is getting results and change tack if communications aren’t hitting the mark. So, believe what the statistics are saying but also work out what they mean. Although there is an ‘Export Data’ function, we tend to pass information to colleagues or clients, using a simple spreadsheet to quickly record the key performance indicators as an executive summary overview.

Rounding the Circle
To get a more complete picture of the power of content, we closely monitor the social media section on Google Analytics to see the referrals from social media accounts – LinkedIn, Google+, Facebook and twitter of course.  By clicking through, information is provided on which piece of activity resulted in the referral, so providing even more information to improve future strategy.

Make an Impression
By studying the data freely available to your business, it’s possible to improve the value of your tweets and see what resonates with your followers. Some examples we’ve found include:
-          - Time of day – the service allows monitoring of the time of day your audience are reading your messages, so you can issue tweets at the best time for them. We are seeing that because clients have different audience profiles, we now issue tweets both in and out of office hours. And don’t forget the weekends too!
-          - High impressions – To get your tweets noticed we try include business partner handles or news website links where a client company is featured. We also include links back to their  websites and check referrals on Google Analytics.
-          - You get the picture? - We have found that use of suitable photos and images have a real impact on engagements, but they have to be relevant to the story or promote the product/service. Brand imagery doesn’t work as well, but they do make your tweets look nice! 
-          - Who you gonna call? - Include a call to action in some tweets when you can respond online or take calls. For one client we saw an increase in phone calls during a promotion, so we worked with them to issue the tweets with a phone number when they had staff available to take calls.


In conclusion, these free statistics from twitter are a welcome addition to the science of social media. Combined with other analysis techniques they are a vital part of our consultancy tool box. However, time needs to be spent interpreting their meaning and taking the appropriate action, which means refining and developing key messages step by step.  

No comments:

Post a Comment