← Back to Stories Illustration showing the hidden costs of free digital services and how user data becomes the real product
📖 Stories: Digital Economics

The Hidden Cost of 'Free': How Digital Services Turn Users Into Products

📅 February 10, 2026 ⏱️ 8 min read
🎧 Listen to the article “The story of free: why nothing is truly free” as a podcast
~12 min listen
0:00 0:00
Google: free. Gmail: free. Facebook: free. Instagram: free. YouTube: free. WhatsApp: free. You use 50+ services without paying a cent. But the companies behind them are worth trillions. Something doesn't add up. Unless you realize that “free” doesn't mean “no cost” — it means that you are the cost.
$600+ bn Digital advertising (2024)
$32.17 Your value per quarter (Meta ARPU)
0 Users who read the ToS
79% Worry about their data

📖 Read more: The History of Data: Why Our Personal Information Has Value

Before digital "free": an ancient idea

The idea of “free” didn't start with Silicon Valley. It's as old as strategy itself.

~2,500 BC

Roman “Panem et Circenses”

Bread and circuses — free. Juvenal (satirical poet, 1st century AD) denounced it: the Roman government distributed free grain and organized free gladiator games. The purpose? Social control. People who eat for free don't revolt. The earliest freemium model: free bread, paid for with obedience.

1903

King Gillette: give away the razor, sell the blades

Gillette sold razors at cost (or below) — and made money from the blades. The “razor” was free. The “consumables” were gold. The “razor and blades” model — copied by: printers (ink), consoles (games), Nespresso (capsules). Nothing is truly cheap — the payment just shifts elsewhere.

1920s-50s

Radio & television: “free” with ads

Television was “free” — but you paid with time. Every 10 minutes, 3 minutes of ads. Soap operas were called that because they were funded by Procter & Gamble. Entertainment was never “free” — it was funded by someone who wanted something.

"If you're not paying for the product, you are the product."

— Andrew Lewis (2010, MetaFilter)

The digital era: “free” as a business model

Chris Anderson (Wired) published the book "Free: The Future of a Radical Price" in 2009. His thesis: in the digital world, the cost of copying tends toward zero. An email costs 0.000001 cents. A Google search: ~0.3 cents of electricity. If the cost is zero, the price becomes zero.

But Anderson didn't fully explain the catch: in exchange, you give up data — which is worth far more than a subscription.

Google: the empire of “free”

🔍 Search (1998)

8.5 billion searches/day. User cost: $0. Google's cost: ~$0.003/search. Google's revenue (2023): $307 billion — 80% from advertising. Every search tells Google what you're looking for, what you need, what you fear. That information is sold to advertisers in milliseconds (real-time bidding).

📧 Gmail (2004)

"1 GB free" — 100x more than Hotmail. The trade-off: Google scanned your emails (initially for targeted ads, stopped in 2017). But the metadata remains: who emails you, when, how often. Doctor? Lawyer? Pharmacy? Google knows.

🗺️ Maps (2005)

Before: GPS maps cost €200+. Google Maps: free. The trade-off: Google knows where you go, every moment. Google Timeline: a detailed record of every trip. Once, only intelligence agencies had this kind of information. Now, you give it away voluntarily.

Facebook/Meta: social life as a product

Zuckerberg (2004): create a profile, connect with friends, share photos — free. But every like, every share, every comment becomes a data point. 3+ billion users = the largest map of human behavior in history.

💰 How much are you worth to Meta (ARPU per quarter, 2023)

US & Canada $68.44
Europe $19.04
Asia-Pacific $5.12
Rest of World $3.43

Your value depends on how wealthy you are — because advertisers pay more to reach Americans (purchasing power) than Indians. You're not a “user” — you're a commodity priced by weight (purchasing power).

Freemium: the psychology of “free”

Dan Ariely (behavioral economist, “Predictably Irrational”): the word "free" isn't just a price — it's an emotional trigger. Experiment: offer Lindt chocolate at $0.15 and a Hershey Kiss at $0.01 → 73% chose Lindt. Offer Lindt at $0.14 and a Hershey Kiss for free → 69% chose the Kiss. A 1-cent reduction “normally” wouldn't change anything. But the shift to “FREE” changes everything.

🎮 Free-to-Play Gaming

Fortnite: free. Revenue (2018): $5.4 billion — from skins, emotes, battle passes. Only 2.2% of players pay — but those who do pay a lot. The term "whale": players who spend $1,000-10,000+. The industry calls them “engaged users.” Psychologists call them “addicts.”

📱 Mobile apps: 96% free

Google Play: 96% of apps are free. Models: in-app purchases, subscriptions, ads. TikTok (free): $16 billion revenue/year. Candy Crush: $1.3 billion/year — solely from micro-transactions ($0.99-4.99 each).

📖 Read more: Social Media History: From Communication to Exposure

✉️ “Free” trial

"30 days free" = you enter your card, forget to cancel, and pay. Amazon estimates that 72% of Prime trial users become paying subscribers. Planet Fitness: "$1/month" → bait & switch, you pay a $49 annual fee. Dark patterns: design that makes it hard to leave.

Open Source: is anything truly free?

Not all “free” is a trap. There is a tradition of genuinely free creation — that began before the Internet.

1983

Richard Stallman: GNU & free software

Stallman (MIT) rejected the idea that programs should be proprietary. “Free as in freedom, not free as in beer” — the freedom to view, modify, and share the code. The GPL license: if you use my code, yours must also be free. Viral freedom.

1991

Linux: the free kernel that runs the world

Linus Torvalds (21 years old, Finnish): “I'm making a free operating system — just a hobby.” Today: 96.3% of web servers, Android (3+ billion devices), NASA, Tesla, SpaceX, all supercomputers. Value of the Linux ecosystem: $5+ trillion — but the kernel itself remains free.

2001

Wikipedia: free knowledge

Jimmy Wales: an encyclopedia without profit, without ads, crowdsourced. 60+ million articles. Funded by donations ($170 million/year). Truly free knowledge needs someone to pay — the difference is that they don't ask for your data in return.

🆓 “Free as in freedom” vs “Free as in beer”

Stallman explains the difference: Free speech (freedom) vs Free beer (gratis). Gmail is “free beer” — you don't pay, but you're not free (you don't control your data). Linux is “free speech” — you can see the code, modify it, share it. The real question isn't “how much does it cost?” — but "who's in control?"

Surveillance capitalism: the true price of “free”

Shoshana Zuboff (Harvard, “The Age of Surveillance Capitalism”, 2019) named the model: surveillance as a business model. It's not that they “sell your data” — it's that they use it to predict and modify your behavior. Google doesn't sell data — it sells prediction: “this user will buy this, with 87% probability.”

🔄 What you actually pay

Google Search Search history
Facebook Relationships, interests, politics
Google Maps Real-time location
YouTube What keeps you engaged
TikTok Attention patterns + preferences

The resistance: I pay, therefore I control?

The pushback started slowly — but it exists.

🔒 Privacy-first services

Signal (free, non-profit, encryption), DuckDuckGo (search without tracking), ProtonMail (encrypted email, Switzerland). Model: you pay (or rely on donations) instead of “paying” with data. Problem: very few people use them (DuckDuckGo ~3% market share).

💶 Subscription fatigue

The “solution” to “free” was subscriptions. But now: Spotify, Netflix, Disney+, iCloud, Dropbox, NYT, Patreon, YouTube Premium... Average American household: $219/month on subscriptions. The “subscription economy” (Zuora): $275 billion (2024). We wanted to stop being the “product” — we became subscribers to everything.

⚖️ GDPR & regulations

The EU (GDPR, 2018): every user can request what data you hold (Right of Access) and deletion (Right to Erasure). Fines: Amazon €746 million, Meta €1.2 billion. But the reality: how many people have ever requested their data? Very few.

AI: The New “Free”

ChatGPT: free. Claude: free. Gemini: free. The most powerful AI tools are free — but they were trained on billions of data points without consent. Each conversation improves the model. Each prompt becomes training data. OpenAI is worth $80+ billion — built on texts, images, and code from people it never paid.

History repeats itself: new technology → “free” → mass adoption → monopoly → “actually, it costs.” ChatGPT Plus: $20/month. Copilot Pro: $30/month. The pattern is always the same: free until you're dependent — then, you pay.

"In Silicon Valley, the word 'free' has become a political tool. It means: 'I give you something so attractive that you'll never ask what the true cost is.'"

— Jaron Lanier, «Who Owns the Future?» (2013)

The history of “free” is the history of an illusion. Nothing was ever truly free — from Roman grain to Gmail. But the form of payment changes: money → time (ads) → data → attention → behavior. Each generation of “free” takes away something less visible — and therefore more valuable. The real question isn't “how much does it cost” — but what are you giving away without knowing it.

● ● ●
Free Services Freemium Google Facebook Digital Privacy Data Mining Surveillance Capitalism Internet Economy