Connect with us


Instagram Shares New Insights into How it Selects Recommended Posts to Highlight in User Home Feeds



Instagram Shares New Insights into How it Selects Recommended Posts to Highlight in User Home Feeds

Looking to get a better handle on how Instagram’s feed algorithms work, and how you can optimize your content approach accordingly?

You’re in luck – today, Instagram has published a new overview of how it ranks content for its ‘Suggested Posts’, or the posts that you see in your Home feed from accounts that you don’t follow in the app.

This element became a key focus recently, after Instagram began inserting a lot more AI-based content recommendations into user feeds, which prompted widespread user backlash, and has since seen IG scale it back, as it works to refine its algorithms. But even with that shift, Instagram does see AI recommendations as a key element of its future, and in maximizing user engagement.

In other words, even if you’re not seeing as many recommendations in your home feed right now, they will be ramping up again sometime soon.

So how does Instagram select which additional content to show you in your Home feed? Here are some insights:

First off, Instagram’s engineering team outlines the focus of its recommendation system, and underlines the key aims of its approach:

  • Users spend a lot of time crafting the perfect home feed for themselves. How can we do some of that work for them and make it feel like they crafted these recommendations themselves?
  • Anecdotally speaking, users who stay engaged keep finding newer sources of interests to follow. Can we help in this act of progressive personalization a bit?

Whether people actually want an automated system to do this work for them is another question entirely, but the intended aim is to replicate human discovery with AI features, in order to enhance user engagement.

That then sees Instagram’s post recommendations fall into two categories – ‘Connected’ and ‘Unconnected’, with the latter being the posts that Instagram’s systems find and highlight, based on your interests.


The process, as you would expect, is largely based on implicit signals – i.e. actions you’ve directly taken in the app, like following and liking posts. But it can also extend to the people you follow, and what they like, as a proxy for direct engagement, while some popular posts are also highlighted based on overall engagement.

But these elements are more related to its Explore surface – in the Home feed, the aim is to replicate the feel of the posts and profiles that you’ve chosen to follow, in order to make it increasingly familiar.

Scrolling through the End of Feed Recommendations should feel like scrolling down an extension of Instagram Home Feed.

Instagram algorithm overview

That’s important to note – the recommendations that Instagram wants you to see in your main feed should closely replicate the accounts that you follow, down to the types of posts they share. At the same time, Instagram’s also trying to insert more and more video – specifically Reels – into user feeds, which is another factor in its more recent experiments.

But the aim, as noted, is to build more directly on your stated interests, as opposed to simply adding in the latest trending content.

So how does Instagram do that?

“In order to ensure that our recommendations feel similar to posts in Home Feed we prioritize accounts that are similar to accounts a user encounters in Home.

  • In the candidate selection step while training and evaluating our ranking models we ensure that the overall distribution is not skewed away from Home-based sources.
  • We follow the same freshness and time sensitivity heuristics as Home Feed to ensure that suggested posts provide a similar kind of fresh feeling as the rest of Home Feed.
  • We also ensure that the mixture of media types (like photos/videos/albums etc.) are relatively similar in Home and suggested posts.
  • For users whose immediate engagement graph is relatively sparse, we generate candidates for them by evaluating their one-hop and two-hop connections. Example: If user A hasn’t liked a lot of other accounts, we can probably evaluate the accounts followed by the accounts A has liked and consider using them as seeds. A → Account Liked by A → Accounts followed by the accounts A likes (Seed Accounts). The diagram below visualizes this line of thinking.

Key points for marketers:

  • Instagram tries to recommend content which is similar to the accounts that people have chosen to follow, so it may be worth conducting more research into what other brands in your industry, particularly those that are doing well on IG, are posting, in order to better align with the specific elements that could then see your content highlighted to your target consumers
  • ‘Freshness’ is important, which means that you need to be posting regularly to ensure that you’re maximizing your opportunities in this respect
  • Worth also noting that Reels is becoming a bigger focus over time, so while it’s not explicitly stated here, as more users engage with Reels, more Reels will, in turn, be recommended in Home feeds

There’s not a heap of nuggets to latch onto here, but the key point is that Instagram wants its Home feed recommendations to feel familiar to each user, so it’s less about highlighting the latest viral hits from across the app, and more about aligning with each users’ explicit interests.

That, in itself, could be very valuable insight for your IG approach.

You can read the full research post here.


Source link


UK teen died after ‘negative effects of online content’: coroner



Molly Russell was exposed to online material 'that may have influenced her in a negative way'

Molly Russell was exposed to online material ‘that may have influenced her in a negative way’ – Copyright POOL/AFP/File Philip FONG

A 14-year-old British girl died from an act of self harm while suffering from the “negative effects of online content”, a coroner said Friday in a case that shone a spotlight on social media companies.

Molly Russell was “exposed to material that may have influenced her in a negative way and, in addition, what had started as a depression had become a more serious depressive illness,” Andrew Walker ruled at North London Coroner’s Court.

The teenager “died from an act of self-harm while suffering depression”, he said, but added it would not be “safe” to conclude it was suicide.

Some of the content she viewed was “particularly graphic” and “normalised her condition,” said Walker.

Russell, from Harrow in northwest London, died in November 2017, leading her family to set up a campaign highlighting the dangers of social media.

“There are too many others similarly affected right now,” her father Ian Russell said after the ruling.


“At this point, I just want to say however dark it seems, there is always hope.

“I hope that this will be an important step in bringing about much needed change,” he added.

The week-long hearing became heated when the family’s lawyer, Oliver Sanders, took an Instagram executive to task.

A visibly angry Sanders asked Elizabeth Lagone, the head of health and wellbeing at Meta, Instagram’s parent company, why the platform allowed children to use it when it was “allowing people to put potentially harmful content on it”.

“You are not a parent, you are just a business in America. You have no right to do that. The children who are opening these accounts don’t have the capacity to consent to this,” he said.

Lagone apologised after being shown footage, viewed by Russell, that “violated our policies”.

Of the 16,300 posts Russell saved, shared or liked on Instagram in the six-month period before her death, 2,100 related to depression, self-harm or suicide, the inquest heard.

Children’s charity NSPCC said the ruling “must be a turning point”.


“Tech companies must be held accountable when they don’t make children’s safety a priority,” tweeted the charity.

“This must be a turning point,” it added, stressing that any delay to a government bill dealing with online safety “would be inconceivable to parents”.

Source link

Continue Reading

Subscribe To our Newsletter
We promise not to spam you. Unsubscribe at any time.
Invalid email address