In SaaS pricing and packaging, packaging decides what customers buy and how their bill grows; pricing decides the number on it. Most pricing problems are packaging problems in disguise, so fix that first: one value metric customers accept, tiers built around segments rather than features, and a tested migration plan for the customers you already have.
Pricing vs packaging: the distinction that matters
Pricing is the price point: $49, $199, “contact us.” Packaging is everything around it: the unit you charge per, what each plan includes, where the limits sit and what pushes a customer to upgrade.
Teams debate the number because it’s easy to change, but no price point fixes a wrong unit or tiers that don’t match how customers grow. Symptoms I look for:
| Symptom | Likely packaging problem |
|---|---|
| Most customers stay on the entry plan for years | Entry plan too generous, or the upgrade trigger doesn’t match how customers grow |
| Large deals need heavy discounts to close | The metric scales faster than the value customers get |
| Customers share logins or remove seats | The seat metric penalizes adoption |
| Sales can’t explain which plan fits | Tiers built around features, not segments |
| Churn or complaints right after an invoice | Usage pricing with bills customers can’t predict |
Layout and copy for the page itself are covered in SaaS pricing page best practices.
Choosing a value metric customers accept
The value metric is the unit your price scales with: users, contacts, projects, transactions, locations. It matters more than any single price point, because it determines whether revenue grows with each customer or stalls at the first contract.
Five tests for a value metric
- It tracks value. When the customer gets more out of the product, the metric goes up.
- It’s predictable. A buyer can estimate next quarter’s bill before signing.
- It grows with the customer’s business, so expansion revenue arrives without a renegotiation.
- It doesn’t punish adoption. Customers shouldn’t avoid using the product to save money.
- You can meter it. Product and billing count it the same way, with no disputes.
A worked example
Take a hypothetical email marketing tool weighing three candidates:
| Candidate | Tracks value | Predictable | Grows with customer | Doesn’t punish adoption |
|---|---|---|---|---|
| Seats | Weak: value comes from the list, not logins | Yes | Slowly | No: teams share one login |
| Emails sent | Partly | No: campaign volume swings | Yes | No: customers send less |
| Contacts stored | Strong: list size drives revenue | Yes | Yes | Mostly |
Contacts win, which is why many email platforms price on list size.
To find yours, interview your best customers about what “getting more value” looks like, then check which candidate metric rises in accounts that retain and expand. If a metric scores well on paper but customers can’t explain it back to you, it will fail in sales calls.
Seat, usage, flat and hybrid models compared
Once you have a metric, choose the model that charges for it.
| Model | How it works | Fits when | Main risk |
|---|---|---|---|
| Per seat | Price per user | Value comes from each person using it: CRM, design, collaboration | Discourages rollout; revenue shrinks when customers cut headcount |
| Active user | Charge only for users active in the period | Adoption varies widely across accounts | Harder to forecast; needs a clear definition of “active” |
| Usage-based | Price per unit consumed | Value scales with volume: APIs, messaging, infrastructure | Unpredictable bills; lumpy revenue |
| Flat rate | One price per plan | Simple product, narrow range of customer sizes | Undercharges your largest customers |
| Hybrid | Platform or seat fee with included usage, then overages | B2B products whose customers range widely in size and usage | Complexity if you add too many meters |
Your go-to-market motion constrains the choice. Self-serve buyers need a bill they can predict from the pricing page, so keep them to one metric with generous included amounts. Sales-led deals can carry committed usage and custom terms because a person explains them. If you haven’t settled that motion yet, start with product-led vs sales-led growth.
My rule: one primary metric per plan, at most one secondary meter. Each extra meter is one more thing a buyer must model before signing.
Drawing tier boundaries: good, better, best
Tiers should map to customer segments, not to the order features shipped. Build them in four steps:
- Name three segments by need. For example: small teams doing the core job, growing teams that need collaboration and reporting, and larger organizations with security and admin requirements.
- Put the core job in every tier. If the entry plan can’t do the main thing your product is for, it becomes a trial with a price tag.
- Place upgrade triggers at the boundaries. An upgrade trigger is something the next segment needs and the current one doesn’t: approval workflows, multiple workspaces, SSO.
- Space the prices. As a starting point, not a rule, I set each step at roughly two to three times the tier below, which keeps the middle tier the obvious choice for your core buyer.
| Feature type | Where it usually goes |
|---|---|
| Core workflow | Every tier |
| Collaboration, permissions, reporting | Middle tier |
| SSO, audit logs, SLAs, custom contracts | Top tier or enterprise |
| Specialized modules a minority needs | Add-on |
| Higher limits on the value metric | Scales across tiers |
The most common mistake I see is fencing by build cost: putting a feature in the top tier because it was expensive to build, though mid-size customers consider it standard. Lost-deal notes usually reveal it.
Add-ons, feature fencing and free plans
Fence with features, limits and service
There are three ways to separate tiers: features (what the plan can do), limits (how much of the value metric it includes) and service (support level, onboarding, response times). Good fences are things the higher segment genuinely needs. Bad fences cripple the core job and make the entry plan feel broken.
Use add-ons for narrow needs
If most customers in one segment want a feature, put it in that segment’s tier. If a minority of customers in every tier want it, sell it as an add-on: extra storage, a specialized module, premium support. Keep add-ons few. When a buyer needs a calculator to price your plan, packaging has become a negotiation.
Decide deliberately on a free plan
A free plan works when individuals can reach real value alone, the cost of serving free users is low, and free users bring in paying ones through sharing or team invites. Set the free limit where an individual gets value but a team hits the wall.
Choose a time-limited trial instead when value depends on setup, data import or a team adopting together. Offer neither when deals are large and sales-led; a guided pilot does the job better.
Validating changes before you ship them
Packaging changes are hard to reverse, so test them first. In my growth strategy work this step gets the most time, because packaging touches acquisition, expansion and retention at once.
- Re-price your current base. Run every existing account through the new packaging in a spreadsheet. Sort into pays less, pays the same, pays somewhat more and pays a lot more. That table tells you the revenue at stake and who needs a conversation.
- Talk to buyers. Run willingness-to-pay interviews with recent customers and lost prospects. Survey methods like the Van Westendorp price sensitivity questions give a rough acceptable range; use them as input, not as the answer.
- Test in live deals. In sales-led motions, have reps present the new packaging to a set of new prospects and log objections word for word.
- Roll out to new customers first. Apply the change to new signups only and compare against the prior cohort. Split-testing public prices creates fairness and trust problems, so prefer sequential cohorts.
- Set success metrics before launch and record today’s baseline for each:
- Paid conversion rate for new signups
- Average revenue per account for the new cohort
- Discount rate and win rate on sales-led deals
- Expansion revenue at 90 and 180 days
- Share of new customers choosing each tier
- Support tickets about billing and plan limits
Migrating existing customers without churn
New customers are the easy part. Moving your installed base is where packaging changes cause churn.
| Approach | How it works | Best for | Risk |
|---|---|---|---|
| Grandfather indefinitely | Existing customers keep old plans | Small base, minor changes | Legacy plans pile up and complicate billing |
| Grandfather for a window | Old terms until a set date or next renewal | Most changes | Needs clear notice and follow-through |
| Migrate with a credit | Move everyone now, offset increases for a period | Old plans you must retire quickly | Upfront revenue cost |
| Opt-in migration | Customers switch when the new plan suits them | Changes that mostly lower bills | Slow; some never move |
A migration plan that holds up:
- Segment by impact using the re-pricing table. Customers who pay less or the same can move quickly. Those facing large increases need the most care.
- Give real notice. Thirty to ninety days is a common range; annual contracts move at renewal, so check what each agreement allows.
- Explain what they gain. Lead with new capabilities or limits, then the price.
- Offer a lock-in. Let customers facing an increase commit annually at current terms for a period, which trades short-term revenue for retention.
- Call the largest affected accounts before the email goes out. Big accounts shouldn’t learn about an increase from an email.
- Track the migrated cohort’s churn and downgrades against your baseline for two renewal cycles, and pause the rollout if it moves.
Get it built
If your plans were set up around features shipped years ago and nobody has revisited the value metric since, I’ll rework the packaging, model the impact on your current base and run the migration. Start with a Growth Audit, $1,500 fixed and credited if we continue. See pricing or get in touch.