SOLID STATE PRESS
← Back to catalog
How Social Media Algorithms Work cover
Coming soon
Coming soon to Amazon
This title is in our publishing queue.
Browse available titles
Artificial Intelligence

How Social Media Algorithms Work

Ranking Signals, Engagement Loops, and the Recommender Systems Behind Your Feed — A TLDR Primer

Your feed feels like it knows you better than your friends do. This TLDR primer explains exactly why — no computer science degree required.

If you've ever wondered how TikTok decides what you see, why Instagram keeps showing you the same three accounts, or why one YouTube video turns into three hours of rabbit-holing, this guide walks you through it in plain English. You'll learn what a ranking algorithm actually is (spoiler: it's just a scoring system for millions of posts), the concrete signals platforms track about you and your content, and how collaborative filtering and neural networks combine to guess what you'll tap next.

It also covers the parts platforms don't put in their help center: why optimizing for engagement naturally produces filter bubbles and outrage spirals, what shadowbanning and content moderation really do behind the scenes, and how to read your own feed more critically once you understand the machinery behind it.

Written for high school and early college students — and just as useful for a parent, teacher, or curious adult who wants the real explanation instead of a hot take — this book is concise and to the point. No padding, no jargon left undefined, no chapter you have to skim through to find the one sentence you needed. Just the ideas, explained clearly, with real examples from real platforms.

If you want to actually understand the algorithm instead of just complaining about it, start here.

What you'll learn
  • Explain what a social media 'algorithm' actually is and what problem it solves
  • Identify the key ranking signals platforms use to score posts
  • Describe how collaborative filtering and content-based recommenders work at a high level
  • Understand engagement-optimization feedback loops and their consequences (filter bubbles, virality, misinformation)
  • Reason critically about your own feed and the tradeoffs platforms make
What's inside
  1. 1. What a Feed Algorithm Actually Is
    Frames the algorithm as a ranking problem: given millions of candidate posts, score and order them for one user.
  2. 2. Ranking Signals: What the Algorithm Sees About You and a Post
    Breaks down the concrete features platforms use — user signals, content signals, context signals, and interaction signals — with real examples from TikTok, Instagram, and YouTube.
  3. 3. How Recommender Systems Learn: Collaborative Filtering and Neural Models
    Explains the two main families of recommender algorithms — collaborative filtering and content-based — and how modern platforms use neural networks and embeddings to combine them.
  4. 4. The Feedback Loop: Engagement Optimization and Its Side Effects
    Shows how optimizing for engagement creates filter bubbles, rabbit holes, virality spikes, and amplification of outrage or misinformation — and why this is a structural feature, not a bug.
  5. 5. Moderation, Shadowbans, and the Rules Layered on Top
    Covers the non-ML rules layer: content moderation, demotion, shadowbanning, and how platforms adjust the algorithm for safety, ads, and legal compliance.
  6. 6. Reading Your Feed Critically: What This Means for You
    Practical takeaways: how to interpret what your feed shows you, how to influence it, and what open questions researchers and regulators are debating.
Published by Solid State Press
How Social Media Algorithms Work cover
TLDR STUDY GUIDES

How Social Media Algorithms Work

Ranking Signals, Engagement Loops, and the Recommender Systems Behind Your Feed — A TLDR Primer
Solid State Press

Contents

  1. 1 What a Feed Algorithm Actually Is
  2. 2 Ranking Signals: What the Algorithm Sees About You and a Post
  3. 3 How Recommender Systems Learn: Collaborative Filtering and Neural Models
  4. 4 The Feedback Loop: Engagement Optimization and Its Side Effects
  5. 5 Moderation, Shadowbans, and the Rules Layered on Top
  6. 6 Reading Your Feed Critically: What This Means for You
Chapter 1

What a Feed Algorithm Actually Is

Open Instagram or TikTok and you're not seeing everything posted since you last checked — you're seeing a tiny, hand-picked slice of it, in an order someone else chose. That someone is not a person. It's a piece of software solving a very specific problem: out of millions (sometimes billions) of possible posts, pick a few dozen and put them in an order, for you, right now.

This is fundamentally a ranking problem — the task of ordering a set of items from "most likely to matter to this person" to "least likely." It is not fundamentally different from what a search engine does when you type in a query and get ten blue links back. The difference is that on a search engine, you tell it what you want. On a feed, nobody asked a question — the platform has to guess what you want based on everything it knows about you.

It helps to contrast this with the alternative: a reverse-chronological feed, which simply shows you every post from accounts you follow, newest first, no scoring involved. This is how Twitter and Facebook feeds worked in their early years, and it's still an option you can sometimes toggle back to. A reverse-chronological feed is easy to understand and hard to game, but it has an obvious problem: if you follow 800 accounts and each posts a few times a day, you'd need to scroll through thousands of posts to find the handful you actually care about. Most of what you'd see would be noise.

Ranking algorithms exist to solve that noise problem — but solving it requires two separate steps, and it's easy to conflate them.

About This Book

If you've ever wondered how does the TikTok algorithm work, or asked Google for the Instagram algorithm explained simply, this book is for you. It's also for the media studies or intro-to-AI student who needs a fast, clear breakdown for class, and for the parent or teacher looking for a real media literacy guide for teens instead of a scare-tactic lecture.

Inside, you'll get a straight answer to how social media algorithms decide feed content, a youtube recommendation system guide that covers watch time and suggested videos, and a plain-English answer to what is a recommender system explained through the same collaborative filtering and neural ranking models Netflix and Spotify use. Think of it as an AI algorithm study guide for students who want the real mechanics — ranking signals, feedback loops, moderation — without a semester-long course. A concise overview with no filler.

Read it straight through first, then revisit the worked examples on ranking signals and engagement loops before testing yourself with the questions at the end.

Keep reading

You've read the first half of Chapter 1. The complete book covers 6 chapters — readable in one sitting.

Coming soon to Amazon