Instagram is a social network for photos and short videos that has belonged to the Facebook group Meta since 2012. What users get to see is largely decided by software that ranks content according to expected attention.
Instagram is a smartphone app in which people publish photos and short videos. You follow other accounts and see their posts in an endlessly scrollable list. The service was founded in 2010 by two developers from San Francisco. Two years later, Facebook bought it for around one billion dollars, a sum much discussed at the time. Today Instagram belongs to the Meta group and has over two billion active users worldwide per month. The company earns its money almost exclusively through advertising, which is mixed in between the posts.
The advertising engine behind the photos
Instagram is not a side project for Meta but one of its most important sources of revenue. Analysts estimate the app’s share of the group’s revenue at well over a third. Advertisers pay to have their ad shown exactly to the people they want to reach. That is the actual value of the service: not the photos, but the knowledge of who is interested in what.
For many small businesses, Instagram has also become the most important sales channel. A café, a fashion label, or a musician can reach a large audience there without their own website. This dependency has a downside. If Meta changes the rules by which posts are distributed, an account’s reach can collapse overnight.
Added to this is the political dimension. Authorities in the EU and the US have been examining for years whether Instagram keeps young people hooked for too long and how it handles their data. The EU’s Digital Services Act obliges large platforms to disclose risks to minors. Such proceedings therefore regularly appear in business news.
How the app decides what you see
At first, Instagram showed all posts in chronological order, the newest first. Since 2016, this has been taken over by ranking software, often called an algorithm. It estimates for each possible post how likely you are to like it, comment on it, or watch it for a long time. These estimated values are calculated together, and the post with the highest score ends up at the top.
The estimate comes from machine learning models. These are programs that derive patterns from millions of past user reactions instead of following fixed rules. They evaluate who you follow, what you have viewed so far, and for how long. Characteristics of the post also count, such as the length of a video or how quickly others reacted to it.
This is especially important for Reels, the short videos in portrait format. There, you mostly get content from accounts you don’t even follow. The software therefore has to select from a huge pool, not just sort it. A common misconception is that Instagram listens in on conversations via the microphone. Meta denies this, and as a rule, the evaluated click behavior is entirely sufficient to make accurate suggestions.
Instagram in the news and in everyday life
In everyday life, you mainly encounter Instagram as the feed, meaning the list of posts, and as the Story, a post that disappears after 24 hours. In addition, there are Reels and direct messages. Many users spend more than half an hour there every day, which makes the app one of the most-used services overall.
In tech news, Instagram usually appears in three contexts: new features with artificial intelligence, disputes over child protection, and competition with TikTok. Reels were introduced in 2020 because TikTok was taking usage time away from Instagram. Meanwhile, Meta is building in tools that can automatically generate or alter images.
It is important to distinguish this from related terms. Instagram is a product, Meta is the group behind it, which also includes Facebook and WhatsApp. And the algorithm is not a single program, but a bundle of many models that work differently depending on the area of the app.