Good Better Best Packaging for SaaS
How to split features across three tiers, which capabilities to gate, how to price the middle tier to win it, and when a fourth enterprise tier earns its place.
On this page 8 sections
- Sort every feature into leaders, fillers and killers
- The four gate types, and which one each feature belongs to
- Engineering the middle tier to win
- Price ratios: why 1 to 3 to 9 beats even spacing
- Add-ons, and when to unbundle instead of upgrade
- Tier counts by motion, and whether a fourth tier earns its place
- What this costs you, honestly
- What to do next
- Frequently asked questions
The short answer
Good better best packaging splits a product into three tiers where the middle tier is engineered to win 60 to 70 percent of self serve volume. Sort features into leaders that drive the buying decision, fillers that add perceived value, and killers that must not be gated. Use price ratios closer to 1 to 3 to 9 than even spacing, gate on value, volume, roles and compliance, and never gate integrations or seat invites.
Key points before you start
Teams spend a week arguing about whether the middle tier is called Pro or Business, and twenty minutes deciding what goes in it. That ratio is backwards. Packaging is feature allocation. The names are the last decision and the least consequential one.
The work is deciding which capabilities sit where, which gates are defensible, and how far apart the prices should be so the tier you want people to buy is the one they buy.
Sort every feature into leaders, fillers and killers
Start with a complete feature inventory and put each item in one of three buckets. This takes a couple of hours with product and sales in the room and it makes every subsequent decision mechanical.
Leaders drive the purchase decision. A buyer would switch vendors for them. Leaders belong in the tier you want that buyer to land in, and they should appear in the first three rows of your pricing table.
Fillers add perceived value without driving the decision. They make a tier feel complete. Fillers are how you make the middle tier look generous without giving away a leader.
Killers suppress adoption when gated. Integrations, seat invites, basic data export, API read access. Gating a killer produces a short-term upgrade in a small number of accounts and a long-term reduction in usage across everyone else.
The seat invite gate
Charging for, or restricting, the ability to invite a teammate is the most self-defeating gate in SaaS. Slack, Figma and Notion all grew on the opposite choice: make inviting free and easy, then charge for what those people do once they arrive. Blocking collaboration blocks the expansion loop that pays for the tier.
The four gate types, and which one each feature belongs to
Once features are sorted, decide the dimension each tier gate runs on. There are four that work and they can be combined.
| Gate type | Examples | Why it works | Risk |
|---|---|---|---|
| Value | Advanced reporting, automation rules, AI features | Tracks what the customer gets, so it feels fair | Hard to communicate on a pricing table |
| Volume | Records, API calls, seats, projects | Scales with customer size naturally | Punishes growth if the steps are too steep |
| Roles and permissions | Granular access, approval workflows, guest roles | Only larger teams need it, so it self-segments | Frustrating for small teams with a real need |
| Support and compliance | SSO, SAML, audit logs, DPA, SLA, dedicated CSM | Real cost to serve, requested by enterprise only | SSO gating attracts criticism if the step is extreme |
Most good three-tier structures use volume between the first and second tier and compliance between the second and third. That is the pattern Atlassian, Monday and dozens of others converged on, and it works because it maps to how customers actually grow: they add people first, then they add a security review.
4
Gate dimensions that hold up in practice: value, volume, roles, and compliance
saas-marketing.net model, method shown on the page
Engineering the middle tier to win
The middle tier should capture 60 to 70 percent of self serve purchases. That is a design target, not an observation, and you get there by removing one thing buyers need from the entry tier and adding one thing they want to the middle.
Pick the removal carefully. It should be a capability that becomes necessary at a predictable point of growth, such as a second workspace, a volume ceiling, or a permission level. It should not be something the buyer needs on day one, because then the entry tier stops being credible and readers assume the whole page is a trap.
The addition should be a leader, not three fillers. One capability that a buyer would name as the reason they upgraded.
Then check the result. Pull the last 200 self serve purchases and count the distribution. If the entry tier is winning 60 percent, you gave away a leader. If the top tier is winning 40 percent of self serve volume, your middle tier is missing something obvious and you are probably leaving contract value on the table in the other direction.
Editable CSV worksheet
Save your marketing measurement plan
Keep a worksheet for your inputs, assumptions and next actions. You can also print the calculation directly from your browser.
Price ratios: why 1 to 3 to 9 beats even spacing
Even spacing flattens the comparison. At 10, 20 and 30 a month, nothing about the numbers signals which tier is intended for whom, so buyers default to the cheapest and treat the others as upsells.
Wider ratios do more work. At 10, 30 and 90, the middle tier reads as the considered choice and the top tier anchors the range so that 30 looks reasonable rather than expensive. The specific multiples matter less than the principle that the gaps should widen, because the value gaps widen too.
| Structure | Entry | Middle | Top | Typical middle tier share |
|---|---|---|---|---|
| Even spacing | $10 | $20 | $30 | Low, entry tier dominates |
| Doubling | $12 | $24 | $48 | Moderate |
| 1 to 3 to 9 | $10 | $30 | $90 | High, middle tier default |
| Wide anchor | $15 | $45 | Contact sales | High, with enterprise separated |
Ahrefs runs a wide structure where the gaps between tiers are substantial and the differences are volume-based, which makes the upgrade trigger legible: you hit a limit, you move up. Notion runs a narrower per-seat ladder because the upgrade trigger there is team size and features rather than consumption. Both are internally consistent, which is the actual requirement.
Test the sentence
For each tier, finish this sentence in under ten words: “You move to this tier when you…”. If you cannot, the gate is not legible to buyers either, and your sales team will spend every call explaining it.
Add-ons, and when to unbundle instead of upgrade
Add-ons solve a specific problem: a capability wanted by a minority of customers across every tier, with a real cost to serve. Extended data retention, premium support, extra API volume, an additional workspace. Putting these in a tier forces customers to buy eight things they do not want to get the one they do.
The rule I use: if the capability correlates with company size, put it in a tier. If it correlates with a use case that appears at every size, make it an add-on.
The failure mode is add-on sprawl. Once you have six, your pricing page becomes a configuration exercise, self serve conversion falls, and every deal needs a human to explain the total. Cap yourself at three visible add-ons and fold the rest into tiers or retire them.
Self-paced learning
Build your SaaS marketing study plan
Choose a free course and work through its published lessons at your own pace. Save the course index for later.
Tier counts by motion, and whether a fourth tier earns its place
Three visible tiers plus contact sales is the default for good reasons. Buyers compare three options comfortably and five slowly. Pages with five or more tiers push more visitors into the contact-sales path or out entirely, because the comparison work exceeds what a self serve buyer will do unaided.
Motion changes the shape at the edges.
- Pure self serve, low ACV: three tiers, no contact sales option, published prices throughout
- Hybrid self serve and sales assisted: three tiers plus enterprise, prices published on the first three
- Sales led, high ACV: two or three tiers used mainly as a framing device, with real pricing negotiated
- Usage based products: fewer tiers, more emphasis on the consumption dimension and its included allowances
A fourth visible tier earns its place when a real buyer segment genuinely fails to fit the existing three, and you can name that segment and point to lost deals. It does not earn its place because the pricing page looks unbalanced, or because a competitor has four, or because someone wants a place to put a feature that has not found a home.
Building or rebuilding a three tier structure
- Inventory every feature and sort into leaders, fillers, killers
Two hours with product and sales. Done when every item has exactly one label and nobody is still arguing.
- Choose the gate dimension between each pair of tiers
Usually volume between one and two, compliance between two and three. Done when you can finish the 'you move up when' sentence for both.
- Place leaders to make the middle tier the intended default
One removal from entry, one addition to middle. Do not move killers anywhere.
- Set prices with widening gaps, roughly 1 to 3 to 9
Check the middle tier does not exceed the budget a manager can approve without procurement in your market.
- Write the pricing page spec before the design
Rows, order, tooltips, the comparison table, the FAQ. The [pricing page spec template](/templates/saas-pricing-page-spec/) covers the structure.
- Grandfather existing customers explicitly and say so
Announce the policy before the change. Done when support volume about the change stays flat in week one.
- Measure tier distribution at 60 and 90 days
Middle tier at 60 to 70 percent of self serve volume means it worked. Outside that band, one of the gates is in the wrong place.
What this costs you, honestly
Repackaging is not free. Every gate you add creates a support conversation, and every gate you move breaks an assumption for existing customers. Expect a spike in tickets for two weeks and a small number of angry cancellations from customers who feel a feature was taken away, even when you grandfather them.
The harder cost is engineering. Feature gating needs entitlement infrastructure, and if your product currently has a single boolean for paid status, adding four gate dimensions is real backend work that competes with roadmap. Teams that skip this end up gating in the UI only, which power users route around, and then the pricing model quietly stops meaning anything.
There is also a genuine argument against the middle tier optimisation described here. If your product serves two clearly distinct buyers rather than one buyer at different sizes, engineering a default tier is the wrong move and you should build two packages instead of three tiers. That case is rarer than teams believe, but it exists.
What to do next
Run the feature sort this week. Leaders, fillers, killers, with sales in the room. It usually reveals at least one killer currently sitting behind a gate, and removing that gate is the highest-return packaging change available to most teams.
Then check your tier distribution against the 60 to 70 percent target before you change any prices. The full method is in SaaS packaging and tiering, and the guided version is lesson 2: package and price the tiers. If you are running a full refresh, sequence it with the 90 day pricing and packaging refresh. The psychology behind the tier gaps is covered in price anchoring and the mechanics in feature gating. For self serve products specifically, pricing for product led growth covers the free plan decision that sits underneath all of this, and the billing systems that enforce your gates are in SaaS pricing and billing tools. The wider cluster is at SaaS pricing strategy.
Editable CSV worksheet
SaaS Pricing planning worksheet
A practical pricing planning worksheet: decisions, owners, evidence and next actions.
Frequently asked questions
How many pricing tiers should a SaaS product have?
Three visible tiers plus an enterprise contact-sales option is the standard that works for most self serve and hybrid motions. Three gives buyers a comparison without paralysis. Five tiers measurably slow the decision and push more people into contacting sales or leaving. If you think you need five, you probably have two products or a missing usage-based dimension.
What is good better best pricing?
A packaging structure with three ascending tiers where the entry tier serves the simplest need, the middle tier is designed to be the default choice, and the top tier anchors the price and serves larger buyers. The middle tier is where the design effort goes, because it should capture the majority of self serve volume while making the top tier look reasonable.
Which features should you gate behind higher tiers?
Gate on four dimensions: value delivered such as advanced analytics, volume such as records or API calls, roles and permissions such as granular access control, and support and compliance such as SSO, audit logs and a signed DPA. Do not gate integrations, seat invites or basic data export, because each of those suppresses the usage that drives expansion.
How should you price the gap between tiers?
Ratios closer to 1 to 3 to 9 than to even spacing. A 10, 30, 90 structure makes the middle tier read as the sensible compromise while the top tier anchors the range. Even spacing like 10, 20, 30 flattens the comparison and pushes more buyers to the cheapest option because nothing signals which one is intended for them.
Should we use add-ons or more tiers?
Use an add-on when the capability serves a minority of customers across all tiers and has its own cost to serve, such as extra data retention or a premium support SLA. Add a tier when the capability defines a different buyer with a different budget. Add-ons keep the pricing page readable; tiers proliferate and make it unreadable.
Should SSO be gated behind the enterprise tier?
Yes, and this is one of the few gates that is genuinely defensible. SSO, SAML, audit logs and compliance documentation carry real implementation and maintenance cost, and they are requested almost exclusively by buyers with security review processes and enterprise budgets. The criticism of SSO pricing is fair when the gap is extreme, so keep the step reasonable.
The saas-marketing.net editorial team Research and editorial
We research, write and maintain every page on this site. The library explains marketing decisions through practical frameworks, explicit assumptions and references. Corrections can be requested through the contact page.
Published September 11, 2026. Last updated .