Where to draw the line between free and paid tiers
Five steps to split free and paid without gating on gut feel, with a worked example for a note-taking app.

Most free-versus-paid splits are set in a meeting where someone says that collaboration feels like a paid thing. Sometimes the guess is right. Often it gives away what people would pay for, or it locks the feature that carries a new user to their first good result. This guide gives you five steps to draw the line on purpose, and a worked example you can copy.
Product managers at SaaS or subscription products who must set a free versus paid split, or revise one that is not working.
A one-sentence job for the free tier, a tagged feature list, a value metric to gate on, a protected aha moment, and a test plan with a success metric.
Write the job of the free tier
Before you look at any feature, decide what the free tier is for. It can do one of three main jobs: acquisition (get people in the door), activation (get them to a first good result), or virality (get them to bring others). Pick the one that matters most right now.
Write it as one sentence. For example: the free tier exists to get an individual to their first useful result within one session. If you cannot write the sentence, you are not ready to gate anything.
This sentence is your test. When a feature is up for debate, ask whether keeping it free serves the job. If it does, it stays free. If it does not, it is a candidate for paid.

List every feature and tag it
Put every feature in a table. Add two tags to each one.
The first tag is who values it: an individual, a team, or a company. An individual wants speed and comfort. A team wants to work together. A company wants control, safety and proof that people are using the product.
The second tag is when it matters in the user journey: first session, first weeks, or after the habit has formed. A feature that matters in the first session is part of activation. A feature that matters later is part of growth in the account.
Do this with people from support, sales and engineering in the room. They know which features users ask about and which ones they ignore.
Gate on value metrics, not feature walls
A value metric is a number that grows when the customer succeeds: seats, projects, usage, or history kept. If a customer is getting more from your product, they hit the limit, and paying feels fair.
A feature wall works the other way. It blocks something for no reason the user can see. People feel the block, not the value, and they leave or look for a workaround.
| Approach | What the user feels | Risk |
|---|---|---|
| Value metric (seats, projects, usage, history) | I am growing, so I pay more | Limit set too low, too early |
| Arbitrary feature wall | Something useful is blocked | Frustration before value is shown |
Not every paid feature fits a metric. That is fine. Use a metric where you can, and use a clear feature boundary only where the feature serves a different buyer, such as a team or a company.
Keep the aha moment free
The aha moment is the point where a new user sees why the product is worth their time. Find the feature or action that leads to it, and do not gate it. If people cannot reach it, they never get a reason to pay.
What comes after is where payment makes sense: collaboration, control, scale and support. These are things people need once the product is part of their work. A person who has found value will pay for sharing it, protecting it, and getting help when something breaks.
Look at your tags from step 2. Features tagged individual and first session belong on the free side. Features tagged team or company and later in the journey are the natural paid ones.

Test the line with data
Your first draft of the line is a hypothesis. Check it against what users already do.
Look at conversion by feature use. Which features do paying customers use before they pay, and which do they never touch? Look at upgrade triggers. What was the user doing, or what limit did they hit, just before they upgraded?
Then run a small cohort experiment before a wide rollout. Apply the new line to a small group of new sign-ups, keep a comparison group on the old line, and watch two numbers together: free-to-paid conversion and retention. Decide the success metric before the test starts, so you do not bend it to fit the result.
Also plan the notice. Tell existing users what is changing and when, and decide how you will treat people who already rely on what you are moving.
Worked example: a note-taking app
A note-taking app has 50,000 free users. The team wants more paying customers but does not want to slow growth. The job of the free tier, in one sentence: help an individual capture and find their notes so they build the habit.
The team lists 10 features and tags each one. The feature list below is an illustration. The decisions on team sharing, version history and unlimited personal notes follow the plan above.
| Feature | Who values it | When it matters | Decision |
|---|---|---|---|
| Unlimited personal notes | Individual | First session | Keep free |
| Search | Individual | First weeks | Keep free |
| Tags and folders | Individual | First weeks | Keep free |
| Sync across devices | Individual | First session | Keep free |
| Templates | Individual | First weeks | Keep free |
| Import from other apps | Individual | First session | Keep free |
| Team sharing | Team | After the habit forms | Move to paid |
| Version history | Team | After the habit forms | Move to paid |
| Admin controls | Company | Rollout | Paid |
| Priority support | Company | Rollout | Paid |
The pattern is clear. Everything an individual needs to capture and find notes stays free, including unlimited personal notes, which is the heart of the aha moment. Team sharing and version history serve people who work together and need to recover work, so they move to paid.
The success metric: free-to-paid conversion above 4%, with no drop in week-4 retention. Both halves matter. Conversion alone could rise because the team squeezed people who then left. Retention alone could hold because nothing changed. Together they say the line is fair.
The team tests the new split on a small cohort of new sign-ups first, compares it with a group on the old split, and only then rolls it out to all users, with notice to the existing 50,000.

Mistakes to avoid
- Gating the core activation featureNew users never reach the aha moment, so they never pay.
- Copying a competitor's tiersTheir users, price and history are not yours. Copy the method, not the result.
- Too many tiersEach extra tier adds a choice. Two or three that are easy to tell apart are enough for most products.
- Changing limits with no noticeExisting users feel tricked. Tell them early and say plainly what changes for them.
- Judging only on short-term revenueA jump in revenue can hide falling retention. Watch both, over more than a few weeks.
Start with the one-sentence job. Tag the features. Gate on what grows with success, keep the aha moment free, and test before you ship. A line drawn this way is easier to defend, and easier to move when the data tells you to.


