{"id":24700,"date":"2022-12-20T12:24:20","date_gmt":"2022-12-20T12:24:20","guid":{"rendered":"https:\/\/mailinvest.blog\/index.php\/2022\/12\/20\/400000-articles-what-we-learned-about-content-engagement\/"},"modified":"2022-12-20T12:24:20","modified_gmt":"2022-12-20T12:24:20","slug":"400000-articles-what-we-learned-about-content-engagement","status":"publish","type":"post","link":"https:\/\/mailinvest.blog\/index.php\/2022\/12\/20\/400000-articles-what-we-learned-about-content-engagement\/","title":{"rendered":"400,000 Articles: What We Learned About Content Engagement"},"content":{"rendered":"

<\/a>\n
<\/p>\n

\n

If Twitter is your main platform, carefully consider amplification strategies<\/h3>\n

Twitter does have the advantage of being far more open than Facebook, in the sense that you can better understand what people are talking about, and more easily\u00a0find influencers<\/a>\u00a0who can help amplify your content.<\/p>\n

Using these strategies can help you overcome some of the challenges of working with a platform that has a smaller user base<\/p>\n

Tweaking content can turn a dial, not a switch<\/h4>\n

This following section is based on the behavior of machine learning models, which predicted whether or not a piece of content would be in the top 20% of most highly engaged posts (Compared to the expected number of engagements based on the prominence of that site).<\/p>\n

We provided these models with information about the words and phrases in the title, as well as some metadata about the content (for example, the number of words and the readability score). These became our \u201cFeatures\u201d; things that the model can identify within content, and that a person creating content is able to change.<\/p>\n

Across all models, we didn\u2019t see any features making more of a 6% contribution to the probability of creating high content engagement, which reinforces that there\u2019s no magic formula to uncover when creating high performing content. Some tricks may help you, but the actual content itself matters, and we have every reason to believe there\u2019s still a strong element of chance.<\/p>\n

How to write engaging content<\/h2>\n

To understand what businesses and individuals can do to create more engaging content, we recreated the model with only blogs included. Here, our dataset is mostly made up of three types of blog:\u00a0business<\/a>,\u00a0hobbyist<\/a>\u00a0and\u00a0commentary<\/a>\u00a0on social and current affairs.<\/p>\n

See below for our top-level findings on the most engaging \u2018Content types\u2019 across Twitter and Facebook, and read on for more insight into each category.<\/p>\n<\/p><\/div>\n