Twitter accounts are like snowflakes – no two are the same.
So if you’re looking for an easy answer on the best time of day to tweet, you won’t find one*. I can, however, help you figure out the best times to tweet to maximize engagement on your particular account.
*Buffer, a great freemium social publishing tool for which my current and previous company have been paying customers, studied 4.8 million tweets to find the best time to post. I recommend checking out Kevan Lee’s findings.
Whether you’re building your personal brand or represent a company, I have both scientific and anecdotal data to share so you can optimize your posts.
Think about when people in the area you’re involved are most likely to be on Twitter. Do you tweet mostly about finance? Focus on weekday business hours, especially around opening and closing times of markets.
Ideally, most of your followers will be interested in what you tweet about most. If you’re running a brand account, this should be a given. For personal accounts, though, it’s more difficult, because people aren’t always able to separate you “the person” from you “the content creator.”
Regardless, figure out your existing audience and when they’re most likely to be online. Sports fans, for example, are most active on nights and weekends, when games/matches are being played.
Catering to your existing audience (i.e. your followers) is important, but you also want to think about who you’re trying to attract.
Perhaps when you launched your account, most of your followers were friends and colleagues. But unless they have a strong interest in what you’re trying to build your account around, don’t gear your content to closely to them. If you’re a brand, again, this should be a given from the outset, unless you pivoted your strategy along the way.
Is there a big sporting event on live TV? A political debate? Major breaking news? A Kanye West rant? Unless topics like these fall into your expertise, and therefore you regularly tweet about them, it’s best to lay low until the tweetstorm blows over. People aren’t likely to pay attention to you when 90 percent of their timeline is flooded with conversations about something else. Tweet volume will be higher in these situations, meaning anything you share will be pushed down in people’s timelines faster as more new tweets flood in.
You can’t plan for these, but it’s good to be aware of the topic du jour and its influence on Twitter conversation.
Data and Tools
Once again, Buffer has done a lot of work for you. Use its Optimal Timing Tool to figure out the best times to tweet based on engagement.
Twitter Analytics: Tweets
Go to Twitter Analytics and sign in to the account you want to analyze
Click on the “Tweets” tab in the top left
Select the date range you want to analyze (28 days should be fine if you tweet a lot, but you can select up to 90 days at a time)
Export the data to Excel
Select all the data from row 2 and below
Sort the data by the stat that’s most important to you
Impressions: If you just want to know how many people saw your tweet, this is the stat for you
Other: If those don’t suit you, sort by something else
Engagement: If clicks, retweets, likes, favorites, profile views and photo/video views are most important to you, sort by this category
Why is it important to think about which category matters most? Because they often don’t match.
Look for patterns in the times of day*
If you only tweet during certain times of day, your data will be biased toward those times. In order to get a true indication of the best times to tweet, you’ll need to spend a few weeks posting at all hours of the day then come back and analyze the data
*Check your timezone in your Twitter settings. As you can see, mine is GMT -5:00, which means I have to subtract five hours from the times listed in Excel to get my local time.
Narrow the times
Think of the day as a giant funnel. At the top you might have morning, afternoon, evening, overnight, based on your local timezone
Once you identify the best part of the day for your account, you can break it down by hours within those times
Once you know which hours of the day work best, you can start looking at your quarters – :00, :15, :30 and :45 past the hour
If you can get reliable data for times of day more specific than quarters, more power to you
Twitter Analytics: Audience
The "Audience" tab in Twitter analytics requires much less work. Simply navigate, and look at the demographics of your followers.
Take these with a grain of salt, as Twitter's user data isn't as comprehensive as say, Facebook's, but it's still a great look at who's seeing your tweets.
This is a great free tool that will probably give you more data than you can use, including:
Mapped location of your followers
Most active hours of your followers
Your most active hours
Word clouds based on your followers’ bios
“Social authority” of each of your followers
Of course posting at particular times doesn’t mean always being on Twitter at those times. Scheduling tools, such as Buffer and TweetDeck, will do the work for you. (If you use TweetDeck, don’t use the desktop app – go to TweetDeck.com, or if you’re on Google Chrome, install the TweetDeck browser extension.)
Now What?/Final Thoughts
You’re likely to find that there is no single time of day your tweets do well – there are probably a few, and don’t be surprised if they’re not next to each other on the 24-hour clock.
Once you find your best times to tweet, target your most important tweets around those times.
As your followers continue to grow, continue tweeting in your “off” times as well. It’s important to continue experimenting, auditing yourself once in awhile, and seeing if your ideal times shift. Remember, you can’t analyze data for times of day during which you haven’t posted.
If you’re goal is to get interaction – not just have people see your tweets – consider posting when the data says the fewest people see your tweets. This likely means it’s a less busy time for your followers, so they’re timelines aren’t moving as fast, and there’s a better chance they’ll see what you shared.
Have some tips I didn’t include? Spot an error? Please tweet them to me or share in the comments below.