“Should I offer a free tier?”

I did not believe in free. I believed in value. The question of whether to offer a free tier is really a question about whether you believe deeply enough in your product to ask people to pay for it — and whether you have built something good enough to justify that ask.

The companies that built free tiers as their primary growth strategy often built free tiers as a substitute for having a product people would pay for. If your product is genuinely valuable — if it solves a real problem in a way that meaningfully improves someone's life or work — then the right ask is a fair price for that value. You are not doing your users a favor by giving your product away. You are treating your own creation as though it is not worth anything.

Look at what we built at Apple. We did not offer free versions of the Macintosh or free versions of the iPhone. We built products that were good enough that people were willing to pay for them. The premium created a standard. It told the world what we believed about the product before anyone bought it.

The freemium model is built on a theory that is often wrong: that users who experience the product for free will eventually pay for the full version. In practice, the majority of free users never convert. They are extracting value from your product without contributing to the revenue that allows you to keep building it. You end up with a large number of users who cost you money to support and a small number who pay you. The economics are difficult, and the incentive is perverse — you are spending your development resources on features that serve people who will not pay you, rather than features that make the product excellent for people who will.

There are narrow exceptions. If your product gets better with more users — if network effects are real and structural — then a free tier can be a deliberate growth investment. But most founders invoke network effects as a justification for what is really pricing insecurity. Be honest about which situation you are in.

I have always believed that business is a means of creating value for society, and that value created should be compensated fairly. But I have also built businesses large enough to understand the difference between a pricing philosophy and a pricing strategy, and the difference between a competitive moat and a revenue model.

The question of free tiers is, at its core, a question about how you acquire customers and at what cost. Paid acquisition — advertising, partnerships, sales teams — costs money upfront and creates revenue faster. Free tiers cost money in product and support, delay revenue, and create the possibility of a much larger installed base that eventually converts to paying customers or creates network effects that raise the cost of switching.

Which approach is right depends entirely on your business. If you are building a tool that individuals use daily in their professional lives — a design application, a development environment, a writing tool — a free tier can be the most efficient customer acquisition mechanism available. The person who uses your free product every day will eventually encounter the ceiling of what the free product allows, and at that moment the cost of upgrading is low and the incentive is real. The product has already proved its value. The conversion from free to paid is not a sale; it is a natural extension.

But if your business requires large account relationships — if your revenue comes from enterprises, from multi-seat contracts, from annual deals — then a free tier may be the wrong mechanism entirely. The buyer in that context is not the individual user who has been using the free product; it is a procurement process, a budget cycle, a decision involving multiple stakeholders. Free trials and product-led growth work differently for enterprise sales than they do for individual tools.

My counsel: be clear about what your free tier is doing for you. Is it acquiring individual users who will convert at high rates? Is it building the network that makes the paid product more valuable? Or is it a form of price insecurity — an unwillingness to ask for what the product is worth? Each of these has a different implication for how you should design the tier and whether it is worth offering at all.

The question of whether to offer a free tier is a question about discipline — specifically, whether you have the discipline to design a tier that serves your users honestly without compromising the integrity of your business.

In my experience governing Rome, the worst decisions were made by officials who tried to be everything to everyone. The attempt to satisfy every constituent simultaneously is not generosity — it is a failure of principle. The official who gives away what he does not have to give, who makes promises he cannot keep, who provides benefits he cannot sustain, is not a generous leader. He is a reckless one. The short-term applause he receives does not change the long-term consequences of what he has done.

The same applies here. A free tier that is designed as a genuine service to a segment of users — a permanent gift to individuals or nonprofits who cannot afford the full product, or a limited experience that gives real value without destroying the economics of the paid version — is an act of discipline. It has defined constraints, defined beneficiaries, and defined limits. It is honest about what it is.

A free tier that is designed to maximize the appearance of traction — that attracts users who will never pay, who strain your support capacity and your infrastructure, who provide a vanity metric of user count without the substance of a viable business — is not generosity. It is a kind of self-deception. You are telling yourself and your investors a story about growth that does not reflect the health of the business.

My counsel is to apply the Stoic discipline of clear thinking. Ask what your free tier is for, who it serves, and what it costs. If you can answer those questions honestly and the numbers work, offer it. If the honest answer is that you are not sure — that you are hoping the numbers will work out — then the discipline is to charge first, prove the value, and revisit the question when you have evidence.

Carnegie's point about the type of buyer is important and I want to sharpen it. The free tier works when the person using the product is the same person making the purchasing decision. Individual tools, developer tools, small team tools — these work on free-to-paid conversion because the user is the buyer. The friction between experience and payment is low.

The free tier fails, or at least works very differently, when there is distance between the user and the buyer. If your users love your free product but your customers are their employers — if the company holds the budget and the individual holds the preference — then a large free user base does not convert to revenue without a sales motion layered on top. You have not eliminated the cost of acquisition; you have moved it downstream, and you have also created a dynamic where the free users have no incentive to advocate for a purchase they will not benefit from personally.

Know the distance between your user and your buyer. That distance is the single most important factor in whether your free tier will convert.

Aurelius has described the discipline question and Jobs has described the buyer distance question. I want to add the moat question. A free tier, properly designed, can create a competitive moat that is very difficult to replicate. If your free product reaches enough users and creates enough habits — if it becomes the default tool in a category, the thing people use because everyone uses it, the product whose export format defines the standard — then the free tier has built a defensible position that no competitor can easily unseat.

This is not guaranteed, and it requires a specific type of product: one that benefits from being widely used, one that creates data or network effects that make it more valuable at scale, one whose free users generate something — content, connections, standards, training data — that benefits the paying users. If your product has this structure, a free tier is not just a customer acquisition tactic. It is a strategic investment in the moat.

If your product does not have this structure — if each user's experience is independent, if there are no network effects, if the free users generate nothing that benefits the paid users — then the free tier is not building a moat. It is just delayed revenue.

The two principles my colleagues have named — distance between user and buyer, and whether the free users build something that benefits paying users — are both specific applications of the general principle I offered. Know what your free tier is for before you build it.

I will add one more discipline question that neither colleague has raised: the support cost. Every free user who contacts your support team, files a bug report, or requires onboarding costs you money that they are not paying for. This is not an argument against free tiers — it is an argument for designing them with explicit constraints on what support you will provide.

The companies that have built successful free tiers have generally been honest about this boundary. They provide documentation, community forums, and self-service resources for free users. They do not provide direct human support. They design the free tier so that the users it attracts are capable of self-service. The founders who ignore this boundary will find that their free tier is subsidized not just by their paying customers but by their own time and their team's capacity — a cost that does not show up in unit economics but is paid in founders' hours.

The synthesis is this: a free tier is a tool, and like all tools it is appropriate in some situations and inappropriate in others. The founders who use it well have a specific thesis about what the free tier will do — which users it will acquire, how it will convert them, what it will build in the product that benefits the paid version. The founders who use it poorly are using it as a substitute for conviction about their product's value.

If you are asking whether to offer a free tier because you are not sure anyone will pay for your product, the free tier will not answer that question. It will defer it. You will accumulate free users and discover, eventually, that most of them will not pay — not because the price was too high, but because they did not have the problem you thought they had in the way you thought they had it. The paid version would have told you that earlier.

Build the best product you can. Ask people to pay for it. If they pay, your thesis is right. If they don't, the free tier will not save you — it will only slow the feedback.

Practical guidance: if you offer a free tier, design it with the conversion rate in mind from day one. Know what percentage of free users you need to convert to make the economics work. Build the product so that the boundary between free and paid is placed at exactly the moment where the value of the paid upgrade is obvious to the user who has been using the free product. That boundary should not be arbitrary — it should be the natural point at which a user has proven to themselves that the product solves their problem and is ready to commit.

If you cannot articulate where that boundary is, you do not yet have a free tier strategy. You have a free product with a paid option that may or may not be visible to the users who need it.

My final position is this: discipline in design. A free tier that is designed with honest answers to three questions — who is it for, what does it cost, and what does it build toward — is a legitimate business strategy. A free tier that is designed to defer hard questions about pricing, conversion, and value is a form of self-deception that will eventually require a reckoning.

The Stoic practice is to ask the hard question now rather than later. What does your free tier actually do for your business? If the answer is clear, proceed with it. If the answer is not clear, charge first, prove the value, and revisit the question when you have real data about what your users will and will not pay for. The data will be worth more than the free users.