The other day a colleague gave me a report to review on a recent Twitter chat she participated in on behalf of one of our clients. It was filled with some very impressive numbers — especially the number of impressions generated by an hourlong Twitter chat. That’s when the red flag appeared.
I asked where the numbers came from, and was told they were provided by the host of the chat. I asked for some background on how the numbers were generated before sending them to the client. The response was, let’s say, testy, but we learned the numbers came from a free online analytics site called Hashtracking.
Here’s how it works: you input a hashtag and the site determines the number of tweets and retweets associated with the tag. Then it calculates the number of impressions by multiplying the number of followers times the number of tweets for each Twitter address that used the tag to determine the number of impressions.
My experiment with a hashtag produced these results: 488 tweets generated 1,780,081 impressions, reaching an audience of 231,997 followers within the past 24 hours. Wow! Big numbers, right?
The logic behind the calculation is tantalizing: if a person tweets using a hashtag, then every one of the person’s followers potentially is exposed to it in their Twitter feed. Makes sense, but to assume that every person reads every tweet in her Twitter feed is just not realistic.
Those kinds of numbers will impress clients who are still susceptible to determining PR values by counting column inches, determining advertising equivalency and multiplying by three. But we assume our clients are more savvy than that and are unlikely to be so easily impressed by inflated impressions. So we removed the numbers from the report and concentrated instead on the quality of the conversation that revolved around the client.
We know, it’s not analytical, but at least it’s meaningful.
One Thought on “Twitter Impressions Can Impress, But Actually Mean Little”
Buck, we’re so glad you’re aware of Hashtracking. We believe that an impressions number does help to understand potential exposure to a particular message or hashtag, and that impressions only one of many metrics needed to understand a Twitter conversation.
It sounds like the report you saw was our free report; our premium service, currently in limited Beta, generates more comprehensive metrics than simply impressions, including who participated in the conversation, who contributed the most to the conversation, and the timeline for the tweets tracked. Future plans for our service include much more detail about the participants and their participation, moving the analysis far beyond impressions. Please let us know if you’d like to participate in our Beta, we’d love to have your thoughts. Thanks!