What you can learn from your best and worst tweets

With Twitter, as with many things in life, taking a long hard look at your most and least successful efforts can tell you a great deal about what kind of approach you should take in the future. What common elements lead you to success and which to a lackluster performance? While we can’t provide that kind of information for your next pickup basketball game, we can help you take a look at and learn from your best and worst tweets to improve your Twitter strategy going forward.

Got questions? Find us on Twitter @UnionMetrics.

Time for a Twitter audit

Don’t panic; this doesn’t have to be a large-scale, deep dive if you don’t have the resources for that. Even if you’re just looking at the last month or even week of tweets you can pick out some things to start working with (and avoiding) to improve your tweets.

Find your best and worst tweets using the tool of your choice- personally we recommend running a free snapshot report around an account if more in-depth Twitter analytics are currently out of your budget- and start there. Line up the best and the worst and start asking questions.

What kind of questions? We’re glad you asked.

Ask these questions

You want to tease out what you should replicate and what you should avoid going forward so start with these questions and keep rolling with whatever comes to mind about your brand’s particular strategy, values, voice and more:

  • What time did you post these tweets?
  • Does a certain time consistently get more responses?
  • Is your audience more responsive on a specific day?
  • How was the content presented?
    • Did you phrase the tweet as a question or statement?
    • Were you attempting humor?
    • Were you trying to comment on a current event (and was it relevant to your brand)?
    • Did you use any slang? (If so, did it match your brand voice?)
  • Did you add a link?
  • Is there other media— a photo, a video, a gif, emojis? Was the video live or pre-recorded?
  • What is your ratio of promotional tweets to non-promotional?

Start with these and you should have a solid list of things to change or try out for the next experimental period.

Test these elements

Once you’ve answered the questions above, you need to work out how you’re going to test the insights you get from answering them. For example, if you think the time you posted some of your poorer performing tweets is an issue because the content wasn’t that different from some that performed well, consider tweaking the message of a poorer performing tweet and tweeting it at one of those better times, then measuring the result.

If tweets with media in them are performing better, try to include media with more of your tweets and be sure to take note if it’s a specific type of media your audience enjoys best. If they like regular video, you might want to experiment with live video to see how they respond to that.

Keep repeating these measurements

Set aside a regular time to do a surface look at how things are going- we recommend weekly- along with a deeper dive. Do the latter at least quarterly so you’re not scrambling to pull together a whole year of data every December or January.

Just keep in mind that improvements can be slow, so be patient and keep an experimental mindset.

Want more engagement on Instagram? Our checkup analyzes the last 30 days of posts, likes and comments on your account to quickly give you key insights into how to improve.

Free checkup