Twitter analytics

Since Twitter rolled out analytics a year ago, it evolved to be possibly the best social analytics tool online. The depth of detail, versatility and reliability of data provided by Twitter analytics can help bring valuable insights into your Twitter campaigns, whether free or paid. It’s a social marketing goldmine and I genuinely hope that it will set precedence in social analytics standards. I’d like to share examples from our experience at inboundli and illustrate how marketers can benefit from Twitter Analytics.

I assume that most marketers use some kind of content marketing or social media management software and like to have their analytics all under one roof. It’s convenient and it’s necessary, but it’s also inherently limited. Generally, social networks provide very few interesting metrics through their APIs and as of yet, there aren’t any tools that can access truly interesting data from across social networks. To be able to tweak social strategies to drive significant improvements, it’s still essential to check analytics tools that are built in within social networks.

Accessing Twitter Analytics

Twitter Analytics is free and can be accessed by following this link: http://analytics.twitter.com. The “Home” menu gives an overview of main metrics with changes over the last 28 days and a summary for current and previous months. Other menus that are in the scope of this article are “Tweets” with various engagement metrics and “Followers”, which offers an overview of your audience.
A view of Twitter Highlights in June

Twitter Highlight

Followers report

Audience interest

The most strategically significant and far-reaching chart in the tool is the follower’s interest. It allows you to check whether your chosen content strategy is in line with your followers. In other words: Are you connecting with the right people and are you sharing the right content to your chosen target audience? If you see that your audience is interested in topics you don’t post about, you might consider optimizing your following patterns or share content that attracts the target audience. Alternatively, you might need to share content that resonates better with your existing followers.

For example, in the following chart you can see that inboundli’s audience is interested in the exact topics we are sharing content about.

Audience Interest

Localization

Although Twitter is global in nature, it’s possible and also useful to localize marketing efforts and for many businesses–especially for local services and retailers, there is a strong case for focusing geographically. For that purpose Twitter offers country, region and language charts. Similarly to the interests’ chart, this helps to narrow down focus and make sure that you are communicating with the right people.

Currently we optimize sharing times to target North America and Europe, but only 70% of our audience is located in these regions (as shown in the image beneath). We have 2 possibilities: Either improves our targeting or start considering new locations when sharing content.

Audience in Countries

 

Other followers charts

For certain brands, a wedding dressmaker for example, the gender and marital status of their customers might be of high interest, while a boutique jewelry shop might want to reflect upon its audience’s income. Whichever it is, the followers report offers an insightful glimpse into your audience’s demographics, behavior and interests.

Consumer goods

Tweets report

Engagement types

Over the years we got used to measuring 3-4 engagement types across social networks, because only limited data had been available. On Twitter, the more commonly known engagements include favorites, retweets, replies and link clicks.

 

Tweet Activity

However, there are at least 9 more metrics offered by Twitter (more if you are using Twitter cards), the most important of those being:

  • Follows – Direct follows from a tweet
  • User profile clicks – Visits to the profile page from a shared post
  • Detail expands – Clicks on a post to see details about it
  • Impressions – Number of times a post was shown on Twitter
  • Engagements – The sum off all the interactions with a post
  • Embedded media clicks – Video plays or photo opening

Twitter has charts for 5 of these metrics within the analytics tool, while the rest is available via data export. The longest period of time for which you can export data is capped at 3 months, but it’s possible to select any 3 months so you can compile multiple quarters in a single spreadsheet.

Conveniently, all engagements have date and time attribution. The times are in UTC and would be most useful if converted to your or your audience’s local times.

The availability of such details helps answer complicated questions and bring more certainty into campaign planning. By being able to look at ratios and relations between the data, you can estimate how much you should tweet in the next campaign, how many impressions are needed to trigger an engagement or how many engagements are needed to generate a single lead.

You can also answer time and copy related questions such as the optimal text length for your posts, the number of hashtags to use or best times and days to tweet.

In the image you can see our best days for impressions and best times to post at within an hour.

Best Day. Best Time.

Bringing it all together

I appreciate Twitter for bringing a new level of detail into social analytics and hope that other social networks will follow suite. If you haven’t tried it already, I suggest starting now and it won’t be before long that you will see a return on the extra effort of having to use yet another analytics tool.

As demonstrated in the following image, we were able to improve every significant metric by optimizing audience targeting, Tweet timing, text length and hashtag usage.

Inboundli

Did you discover more awesome ways to use the tool or have suggestions how to improve the flows mentioned in the article? I’d love to hear all about it, so please do share in comments.

About the author: Gene Sobolev I’m a data-driven marketer with an engineering background. I also enjoy sales, data mining with Python and Frontend development. Co-founded and currently working on inboundli – a content curation platform for social media marketing.