How Many Membership Tiers, and Where to Draw the Lines
A tier is not a price point. It is a claim that two kinds of member want two different things — and if that claim is false, the tiers will not work no matter how you price them.
The most common tier structure in the wild is Bronze, Silver, Gold, where Silver is Bronze with more videos and Gold is Silver with a call. It underperforms reliably, because it asks members to buy quantity when what differentiates them is need.
Start with two
Two tiers force you to name the single most important difference between your members. That constraint is useful. Three tiers let you avoid the question by scattering features across a spectrum.
The two that work most often are the thing and the thing plus access to you. Content and community scale; your attention does not, which makes it the most naturally defensible boundary available to a small operation.
What makes a boundary defensible
A boundary works when a member can tell, without asking you, which side they are on. Test each proposed line against three questions:
- Can a prospect self-identify in under ten seconds from the pricing page alone?
- Does crossing the line correspond to something real changing in their situation — team size, revenue, stage, ambition?
- Would you be comfortable defending the difference out loud to someone on the lower tier?
If the honest answer to the third question is "the higher tier just gets more", you have a discount, not a tier.
Boundaries that tend to hold
- Access to you — group calls, office hours, direct messaging, review of their work.
- Depth versus breadth — the library for everyone, the implementation programme for the people doing it seriously.
- Team versus individual — seats, shared workspaces, admin controls.
- Done-with-you versus do-it-yourself — templates for one tier, feedback on their use for the other.
- Recognition and status — visible standing in the community, directory listing, early access. Weaker on its own; strong as a supplement.
Boundaries that tend to fail
- Volume of content. Members cannot evaluate quantity before joining, so it does not drive the decision.
- Archive access. Gating the back catalogue punishes the exact behaviour you want, which is deep use.
- Support speed. Slow support for paying members is a churn mechanism you have chosen to install.
- Arbitrary limits with no cost basis. Members can feel the difference between a limit that reflects something real and one invented to force an upgrade.
When to add a third tier
Add it when members tell you, unprompted, that they want something you do not sell. Not before. The signal is people asking to pay you more for a specific thing — that request is the tier, already specified by the customer.
The third tier is usually at the top, not the bottom. A cheaper entry tier attracts members who churn faster and support more, and it drags your average revenue down in exchange for a member count that looks good and behaves badly.
If you are considering a cheap entry tier to lower the barrier, consider a trial or an annual discount instead. Free trial versus freemium covers the trade-offs.
Presenting the tiers
Three practical things measurably affect which tier people choose:
- Name tiers after who they are for, not after metals. "Solo" and "Team" outperform "Pro" and "Premium" because they let people self-select.
- Mark one as recommended. Most people want to be told. The recommended tier should be the one you actually want most members on, not the most expensive.
- Lead the higher tier with what it adds, not with a repeat of everything below. A feature list that restates the lower tier makes the upgrade look like padding.
Moving between tiers
Upgrades should be instant and obvious; a member who has decided to pay you more should not have to wait or email you. Downgrades should be available and slightly less prominent — a downgrade path saves members who would otherwise cancel entirely, and a cancelled member is much harder to recover than a downgraded one.
Check both flows in the product before you commit to a platform, and check what happens to the member mid-cycle. Proration behaviour is where this gets ugly, and it is rarely mentioned in a demo unless you ask.