Category Archives: Measurement

Count It. Twitter Makes Analytics Available to a Select Few, Hopefully All.

I am like many others who didn’t realize “tweet” wasn’t already an official word up until yesterday when the good people atThe Oxford English Dictionary quietly announced that it had added it as both a noun and a verb – along with other new words such as “big data” and “crowdsourcing.”

What I have known for a long time is that Twitter has never made performance analytics available to users. Those of us who manage personal and corporate accounts on behalf of employers and clients have been forced to cobble together data collected manually using the few things Twitter makes available, data from third-party services like TweetStats, click-thru data from Bit.ly, and data from paid services like SimplyMeasured and Hootsuite PRO.

Finally, the era of frustrating Twitter analytics may soon be over. Twitter hasannounced analytics for a select few. The analytics dashboard is a simplified version of what is offered to advertisers, but it is certainly a step in a better direction.

For a while I worried that Twitter was holding off until the company could figure out how to charge for the use of its analytics dashboard. Fortunately, this service will be free for all users as they have realized that only by knowing how Twitter content is performing, will brands invest money on the platform.

Below are highlights, including tips for social media managers:

  • Provides snapshot of activity, including mentions, follows and unfollows over a month’s time.
  • Users can see how many faves, retweets, replies and click-thrus each of their tweets receives.
  • Users can sort by “Best, Good, All” for at-a-glance ranking of tweet performance. “Best” being the top 15% tweets that received engagement in the last month, “Good” being the top 66% of tweets that received engagement in the last month, and “All” being all tweets.
  • Users can learn more about their new followers, including unique interests, demographic information (location and gender), and the top users also followed by your followers.
  • Users can export analytics as a CSV for later use and reference. This makes it easier for social media managers using templates for social channel metrics to accelerate reporting and even reallocate time spent counting tweets, retweets and mentions, to actually digging in deeper into insights.

twitter 2 twitter

What’s missing

  • Hashtag information, including click-thrus
  • Deeper “influencer” insights.
  • The whole thing still has some kinks to work out, but it is very promising indeed.

My agency works with several clients who are key players in big data and data-driven innovation. I would be remiss if I did not mention the importance of leveraging performance metrics to guide content strategy. As such, it would not be a bad idea to set up some A/B tests for content. If you have a launch coming up, try drafting tweets that include calls to action, link to multimedia, true/false questions, data points, etc. and determine which type of content your audience responds to.

To see if you have access, visit https://analytics.twitter.com and sign in with your Twitter user name and password. If you are one of the lucky few you will see an “Analytics” tab at the top. You should be able to toggle between “Timeline Activity” and “Followers.”

Do you currently measure Twitter performance? Why? Why not?

How do you use analytics to inform your content strategy?

What specific metrics would you like Twitter to add to its performance analytics dashboard before it becomes widely available to the public?

@NYTimes Project Cascade Visualizes Complex Twittersphere, Decodes Diffusion And Value

The sexiest thing I’ve seen today came to me by way of The New York Times. Project Cascade is the lastest technology developed by The New York Times R&D Group in order to better understand how a story goes viral on Twitter in order to change and improve the ways in which the newspaper delivers value to its readers.Project Cascade is exciting to me for a number of reasons:

  1. While other Twitter visualizations focus on connections and following/follower relationships, Project Cascade delves into the content of tweets and how it is disseminated throughout the Twittersphere.
  2. While other visualizations focus on retweets, hashtags, replies, mentions, and other specific parts of Twitter conversations,Project Cascade takes a more holistic view of the rather complex ecosystem that is Twitter.
  3. While many metrics and social media measurement platforms tend to emphasize the role of influencers and opinion leaders (or the people we have labeled as such), Project Cascade has the potential to show that Twitter conversations diffuse throughout the Twittersphere by way of everyday/average users.
  4. Project Cascade also has the potential to help us learn more about what messages have a longer life expectancy or chance of propagating throughout the Twittersphere. This could have very practical implications, not only for media companies but for brands, organizations and individual Twitter users. Learning what constitutes value on Twitter is a secret we have yet to decode, and Project Cascade shows a lot of promise in terms of helping us move in that direction.
  5. Project Cascade will soon be used to visualize conversations for other publications and content providers, which means that this is a project that will not only help The New York Times learn from its own readers’ social conversations, but other organizations using social media will be able to do the same with their readers, viewers, customers, etc.

As a visual learner, infographics and visualizations are immensely valuable and infinitely more meaningful when it comes to learning concepts and understanding how they work, how they can be applied, and how they interact with each other.Here is a fantastic article about Project Cascade from the Fast Company Design blog. Below is a video that explains Project Cascade in more detail. I’m looking forward to its continued development and applications.Long live The New York Times.What are your thoughts on Project Cascade? Is it really the first of it’s kind? What take aways would you most look forward to?

Whatever happened to surveys as part of the PR professional’s toolbox?

During this week’s #pr20chat, Justin Goldsborough (@JGoldsborough) asked how PR professionals can best identify and cater to the needs of their target audience. My response was something only a current public relations graduate student, fresh out of taking her comprehensive written and oral examinations and, appropriately trained in methods by her we-mean-business-when-we-say-research university would respond…and that response led to a comment by @JGoldsborough who shared that the practice of conducting surveys is not one that PR professionals employ as frequently as they should. After all, if the public is at the heart of public relations, why wouldn’t we ask them how they feel about certain issues that involve our organizations, clients, industry, and even current events that directly or indirectly affect us and them?

Read the full post on Shonali Burke’s Waxing Unlyrical

What Constitutes Value in Social Media Measurement?

measurementThis weekend, I read a few articles on public relations research that got me thinking (I think that’s what my professors were going for 😉 ) The general consensus is that public relations research is often limited to the evaluation of the products of short-term communication programs when what it should be doing is evaluating the products, processes, and the outcomes of both short- and long-term programs (Grunig & Grunig, 2001; Michaelson & Macleod; 2007). Of course, this statement assumes that organizations are conducting research in the first place… 

Social media ROI has been an ongoing debate ever since those who were dabbling in the space realized that they needed to come up with something measurable in order to convince reluctant executives that social media was a worthwhile investment of their organizations’ time and resources. Unfortunately, number of page views, clicks, conversions, tweets, mentions, fans, blog posts, etc. are all examples of products, not processes, not outcomes, and not necessarily long-term.

Processes refer to things like relationship-building and outcomes refers to the quality of those relationships (Grunig, & Grunig, 2001). 

A few days ago, my friend @AlanWeinkrantz tweeted:

The types of data we are used to capturing are numbers; we need to capture conversations and more importantly, streams.

I couldn’t agree more. I also think that words like “measurement” and “metrics” are stifling. I agree with scholars and think “research” and “evaluation” are a better fit. The idea is that organizations are conducting investigation that is purposeful, systematic and rigorous in nature and that they will use the results of said investigation to evaluate strategies and tactics. So I’m sorry for using “measurement” to title this post. Dang SEO keywords, rankings, and the very vicious circle I’m trying to argue against.

To clarify, I don’t think measuring products is a waste of time. I think we need to pay more attention to the processes and outcomes dimensions of social media research and evaluation. If we conceptualize social media as what lies at the heart of the Web 2.0 phenomenon – user collaboration, networking, sharing, and interaction; then it makes sense that in addition to counting the number of “X”, we research the quality of what is being said about our organizations’ efforts, programs and campaigns. Because quality is in the eye of the beholder, interpretative, and subjective in nature; the key is to know who you are, what you’re looking for and what you will do when you find it.

Super Bowl XLIV is in just a couple of hours… I’m SO VERY MUCH looking forward to Brand Bowl. The companies that spent gazillions of dollars on those commercials are going to get all sorts of ratings charts, bar graphs, and spreadsheets with numbers on them in the morning. The numbers will be huge and the companies may decide to buy ads in next year’s Super Bowl… But the kind of qualitative feedback they will get from listening to viewer’s *actual* conversations might make them reevaluate their approach (think Pepsi).