The Impact of Caption Length on Instagram Engagement
The Impact of
Caption Length
on Instagram Engagement
6 min read
Today we’ll take a look at how your Instagram audience or that of your competitors responds to posts with different caption lengths.
Does caption length impact audience engagement? It does affect an important metric – the engagement rate – but only if long-form captions are well-written. Producing catchy and captivating copy for a caption requires strong communication skills and understanding of your audience. Deciding on an image or video is also just as important as your captions. Today, you might go with a video accompanied by a short and snappy caption, and tomorrow you may be in the mood for a photograph with longer text. All components are equally important if you plan to craft winning posts.
LiveDune breaks down all posts into 5 categories by number of characters:
1. 0-150
2. 151-500
3. 501-1,000
4. 1,001-2,000
5. 2,000+
To retrieve the data for each category, we have analysed the caption length of 19 million Instagram posts and constructed the following graph. The horizontal line (x axis) indicates the number of characters and the vertical line (y axis) stands for the number of posts in an excerpt.
The results are as follows:
- The majority of posts consist of up to 150 characters. Beyond this caption length, the curve becomes more consistent.
- When the curve reaches 2,000 characters, it starts going up. Some of the posts from this group include captions not only in a caption box, but two-part captions that continue in the first comment section.
We distinguished two differing groups of posts: those with very small captions and those with very large captions. All remaining posts that have from 151 to 2,000 characters, we subdivided into: a) short posts – 500 characters, b) medium posts – 1,000 characters, and c) long posts – up to 2,000 characters. The new graph is available under the ‘Posts’ tab for both your private Instagram account and that of your competitors. In addition to likes and comments, a saves metric is also available for individual accounts.
Examples of workflow
Let’s consider an example who you can work with the “Length of the post’s text” graph:
The number of posts is displayed under the categories to which the posts belong. However, there are only ten publications in the last three categories, meaning we don’t have sufficient data to draw conclusions to make conclusions. Let’s try to analyse the information for all of 2018:
The situation is better here – short posts and long reads accrue the most interactions. Click below on ‘Likes’ to exclude them from the chart and to view comments only:
The longer the text, the more users comment on posts. Yet one of the reasons that statistics for this user’s account are so positive is that his content has variety, including posters, videos of performances, and long reads all in one account.
Another example:
The result we get is strange: the longer the captions, the lower the engagement they create, and the majority of posts consist of more than 2,000 characters. This chart indicates that there is a snagging point with this user’s caption length – something to think about and address closely.
@thedadlab
Our next account, thedadlab, will help us understand how the length of posts affects their saves.
Although 95% of the content here ranges between 150 and 1,000 characters, and there are hardly any lengthy captions for the whole of 2018, certain conclusions can still be made. We can see that more extended captions provoke people to leave comments. The audience wants to engage in conversation:
The situation with content saves is quite the opposite – shorter posts are saved much more often. Most likely, the reason for that is the majority of the content on the account is video tutorials on various indoor activities for children. Trying not to lose or forget useful material, followers of the account save the posts to refer to them later.
On the contrary, long posts are often a simple image with a long description. They can be commented on, but there is little sense in saving these types of posts.
The bottom line is that caption length analysis is a rather complex method of monitoring, requiring between 5 and 20 categories.
