Free as discount
- Promotional tactic
- Limited time / quantity
- Goal: convert to full price
- Margin = price minus cost
- Works in physical goods
Free
by Chris Anderson
Source book · ~4h read
“The first to free gets attention, and there are always ways to turn that into money.”
The argument
Central thesis
Chris Anderson argues that 'free' is no longer just a promotional tactic — it's a fundamental economic category in the digital age. When the marginal cost of distribution approaches zero (digital goods), pricing of zero becomes economically rational, not merely promotional. Free generates audience; audience converts to paid through specific business-model patterns: freemium, ad-supported, cross-subsidies, gift economy, and labor-as-payment. Founders who treat 'free' as a sales discount miss the strategic point: it's the new default, and the question is which paid model sits on top of it.
At a glance
Two ways to think about free
Free as default
- Economic category
- Permanent, scaled
- Goal: layer paid model on top
- Margin = paid tier serves all
- Works in digital goods
The hook
The founder problem this book solves
'Free with upsell' feels like the answer — until you discover why it's also the trap.
First-time founders building digital products are constantly negotiating the free question. Should the first tier be free? How free? Free for how long? What converts? Anderson's contribution is naming the underlying economics. When marginal cost approaches zero, free is the default; the strategic question is which of five paid models layers on top.
The freemium model (small % of users pay; their payment subsidizes the rest) works only if conversion rates clear a bar — typically 1–5% — and the paid feature genuinely matters to that segment. Ad-supported models work only if the audience is large and advertiser-attractive. Cross-subsidies (give product A free, sell complementary product B) work when the lock-in is real. Each model has its own economics, its own risks, and its own kind of customer.
For first-time founders, the value is clarity about which free-then-paid model fits the venture. Picking the wrong one — running ad-supported when freemium would've worked, or freemium when the audience is too small to pay — is one of the more expensive early-stage mistakes, and it usually shows up 18 months in, when you can't pivot.
0 takeaways
What to remember
Practice CardOne-screen exercise
The Free-Layer Choice
Take your current digital product (or planned one). Answer four questions in writing:
1. What's the marginal cost of serving one more user? If it's not near-zero, free isn't your default — Anderson's logic doesn't fully apply.
2. Which of the five 'free' models fits?
Freemium — works if the paid feature is genuinely important to a small but identifiable segment.
Ad-supported — works if you can build very large audience and advertisers want to reach it.
Cross-subsidy — works if free product locks users into a paid one.
Gift economy — works if user contributions are the value (think Wikipedia).
Labor-as-payment — works if user effort produces value you can monetize (think Reddit moderators).
3. What's the realistic conversion rate, not the optimistic one? If freemium, what percentage of free users will pay, and at what price? Be honest. Most founders inflate this 5–10×.
4. What expectation does my offering set in the category? If I price too low or too free, am I creating a market that can't sustain my business? Or am I joining one that's already free-by-default?
The answers should converge on one model. If they don't, you don't have clarity yet — and pricing without clarity is one of the more expensive mistakes you'll make.
Read
Get the book
Search Free by Chris Anderson on Amazon, your local bookshop, or your library system.
The loop closes here
Stories from founders who applied this
When a founder applies an idea from Free and something shifts, they post it as a Knack. Knacks tagged with this book surface here — practical, written by the people who lived it.
Knacks
Open invitation
Be the first to share a Knack about Free.
Did applying something from this book change a week, a decision, a meeting? Tell another founder. Even a small shift, written honestly, is the kind of Knack that gets marked “This worked” — and helps the next founder pick up the book and try it.
Pseudonymous by default. No humble-bragging — just here's what I tried, here's what shifted.
Share