HASHTRACKER

X and Twitter analytics 3 min read

How to compare hashtags on Twitter without fooling yourself

To compare hashtags fairly you need the same period, the same query rules and more than one number. Volume, impressions and people often disagree.

Comparing two hashtags looks like a job for one chart. Put the counts side by side and the bigger bar wins. Most of the time that chart is wrong, or at least it is answering a question nobody asked.

See what a report gives you

Start with the same window

The most common mistake is also the dullest one. One hashtag is measured over the week of the launch, the other over a month, and the month wins.

Fix the dates first. If the two campaigns ran at different times, compare equal stretches from the start of each, the first seven days against the first seven days. It looks less tidy on a slide, and it is the only version that survives a question.

Use the same query rules for both

If one query excludes retweets and the other does not, you are comparing two different things. The same goes for spelling variants. Write both queries with the same operators and check them side by side before you run anything.

#HashtagA -is:retweet
#HashtagB -is:retweet

Which measure of popularity do you mean

Popularity can mean three different things, and they often disagree.

Volume is how many tweets used it. Impressions is how many times those tweets were seen, and on X that figure is counted per tweet rather than modelled. Unique accounts is how many people took part.

Picture a campaign hashtag against a fan hashtag. The fan one wins on volume because a few hundred people tweet it all day. The campaign one wins on impressions because three big accounts posted it once. Which is more popular? It depends on what you wanted from it, and that is worth writing down before you look at the numbers.

Look at who carried each one

Two hashtags with the same impressions can have completely different shapes. In one, the visibility is spread across hundreds of accounts. In the other, a single post did most of the work.

Concentration tells you which is which. A hashtag that depends on one account is fragile, and it will not repeat next year unless that account does. There is more on this in who carries a conversation.

Check the overlap

With the CSV of each hashtag you can cross the user IDs and see how many accounts used both. A large overlap means the two hashtags were talking to the same people, which matters if you were paying for the second one to reach somebody new.

That is one of those checks that takes ten minutes in a spreadsheet and changes the conclusion more often than you would expect.

Compare over time as well

Totals hide rhythm. Plot both by hour or by day and look at the shape. A hashtag that peaks and dies in an afternoon and one that holds steady for a week can end on the same total and mean very different things. We covered this in a peak or a plateau.

For comparisons between brands rather than hashtags, share of voice is the better frame.

Hash Tracker runs each hashtag as its own query, from the official API and within the Twitter compliance rules, and gives you the report plus the CSV for each. Reports are unlimited on every plan, so running two costs you tweets, not an extra fee.

Pick a plan and run both.

Pick a plan

Choosing a subscription

Every subscription carries the same product. The only decision is how many tweets you need to analyse.

See the subscriptions