Good-Better-Best Pricing
Packaging beats pricing. A well-designed three-tier structure generates more revenue than a price increase because buyers self-select into their willingness-to-pay band instead of your sales team negotiating every deal. GBB is the most commonly deployed SaaS packaging architecture and the most commonly broken one, and the failure mode is almost always the same: tiers built around features, not buyer jobs.
What Is Good-Better-Best Pricing?
Good-better-best (GBB) pricing is a packaging architecture that presents buyers with three distinct product tiers at ascending price points, designed to capture different willingness-to-pay segments while anchoring buyers toward the middle or top tier.
It relies on three mechanisms, covered below: anchoring from the Best tier, the compromise effect on the Better tier, and self-selection through fences, so the page sorts the market without a negotiation.
Three tiers is the default, not a law. Over-packaging confuses a homogeneous market: if every buyer hires the product for the same job, extra columns add discovery time and no revenue. The SaaS packaging framework covers the wider set of models this page sits inside.
Entry buyers. Completes one whole job for a well-defined segment at a price that segment can approve. Overbuilding it is the most common GBB mistake; underbuilding it creates a trial with an invoice.
Core buyers. Completes the core job at more depth, plus the adjacent job that creates urgency to leave Good. The tier the compromise effect should land on.
Executive buyers. Outcomes at organizational scale: compliance, governance, multi-team coordination, strategic reporting. Differentiated by risk and complexity, not feature count.
FintastIQ's working target. A large deviation is a packaging finding, not a sales finding.
How GBB Pricing Works on Buyers
Anchoring: the Best tier sets the reference frame
The first price a buyer sees shapes how they judge every price after it. When Best is $60,000 a year and Better is $20,000, the buyer weighs $20,000 against the $60,000 anchor. The Best tier earns its place even at low conversion: 5 percent into Best and 55 percent into Better beats a two-tier company charging $20,000 with 55 percent conversion.
The compromise effect: the Better tier wins from the middle
B2B buyers need defensible decisions. The cheapest option looks like under-investment, the most expensive like overspend, and the middle is the safest choice to defend in a committee. That holds only while Better is clearly superior to Good on what the core buyer cares about and clearly below the enterprise commitment of Best. Tiers that differ only by feature count collapse the effect.
Fences: self-selection only works when the boundary is honest
A fence is the condition that separates one tier from the next: a record count, a seat threshold, an audit requirement. Buyers self-select honestly when the fence is something they already know about themselves. A team of 40 knows it is not a team of 5. When the fence is a feature the buyer cannot evaluate before purchase, your reps negotiate the boundary deal by deal, the outcome the structure exists to prevent. Anchoring and compromise get the buyer to the page. The fence decides the column.
Why Most B2B SaaS GBB Implementations Fail
All three trace back to tiers designed around the product, not the buyer.
Features assigned to tiers by engineering cost, not buyer value
Features land in tiers by build cost or ship order. Good gets most of the useful functionality because it was built first; Best gets expensive capabilities that solve problems 5 percent of buyers have. Nobody sees a reason to upgrade, the mix flattens toward Good, and the Better discount rate rises as sales fixes packaging with price.
No upgrade trigger from Good to Better
Without a condition that creates urgency to leave Good, buyers stay indefinitely. "Good handles up to 500 records; Better handles unlimited records with automated processing" is a trigger. "Better includes advanced analytics" is not, because buyers will look at the analytics later, and later never comes.
Best tier not differentiated by executive-level value
Executives evaluate organizational risk, compliance, and strategic outcomes, not features. A Best tier that differs from Better by extra features fails to convert them; one defined by audit logging, multi-team governance, SLA guarantees, and executive reporting converts. Ask of every Best feature: does this protect the sponsor from organizational risk?
The tier-mix diagnostic: 60 to 70 percent in one tier
Pull revenue by tier for the last twelve months. If 60 to 70 percent sits in one tier, the structure is not segmenting, whichever tier it is. Heavy Good means the entry tier does most of the job and nothing pushes buyers out. Heavy Better is subtler: the founder reads a 72 percent middle tier as product-market fit, when it usually means Best buyers cannot see enough difference to justify the step and have landed one tier too low. On $30M ARR, ten points of mix moved from the middle to the top is roughly $3M at zero acquisition cost. A packaging problem, not a sales problem.
Fathom Data, $12M ARR, had 72 percent of ARR in Better and a CEO who took it as proof. An audit found 41 percent of Better accounts using capabilities that only existed in Best, switched on through custom terms and a CS team preventing churn: Best buyers paying $720 a month who would have paid $1,200 to $1,400 for a tier built for them.
The Good tier that is a trial in disguise
The opposite mistake is an entry tier the product cannot serve. Harborline, a PE-backed vertical SaaS, grew 40 percent a year to $18M ARR with NRR at 103 percent; growth hid the problem. At $32M growth slowed to 18 percent and the board funded a sales overhaul: nine months, $800K, growth back to 22. The packaging audit that should have come first found a $399 Good tier churning at 34 percent a year. The product solved that segment's problems at Better depth, not Good depth. Good was a trial with an invoice. Retiring it into a 60-day onboarding cohort that fed Better cut churn by 14 points and moved NRR to 111 percent within two quarters. The four failure patterns of a packaging redesign covers a fix sequenced badly.
How Do You Design a GBB Structure That Works?
Start from buyer jobs, not features or costs. Six decisions, in order.
Write the three jobs, in the buyer’s words
A job is an outcome in the buyer’s words: "reduce time-to-close for my sales team," "hold compliance across twelve jurisdictions." Different jobs carry different willingness to pay, and those differences are your tiers. Write one sentence per current tier; if you cannot do it in ten minutes, your tiers are placeholders.
Assign each capability to the first tier where it is essential
For each job, find the smallest set of capabilities that completes it; it is almost always smaller than your current tier. A capability one large customer requested in year two is an add-on, not structure. Brightpath, $22M ARR marketing automation, found 67 percent of Good accounts never used the capability that justified the Good price. Moving it to Better cut Good churn from 31 to 18 percent in 90 days.
Choose one fence per transition, on the value metric where you can
Put the Good-to-Better fence on the value metric: records, seats, monitored assets, whatever scales with the job, so growth in the customer’s business is the trigger and nobody has to sell it. Put the Better-to-Best fence on governance, compliance, or organizational scale. The test for each: will the buyer meet it through usage, or does it take a sales call to explain?
Price each tier from its segment’s willingness to pay
Set each tier at roughly the 70th percentile of the willingness-to-pay distribution for its segment. The 1:3:8 ratio is a check on the result, not a method. Flatter than 1:2:4 means the tiers differ by feature count. In a new category the benchmark does not exist, so price from the alternative the buyer already pays for.
Give the Best tier a job beyond anchoring
A Best tier that only anchors is still worth having; one with a real enterprise job holds 25 to 35 percent of revenue. The difference is whether it answers the sponsor’s question: what happens to me if this goes wrong? Audit trails, cross-team permissions, uptime commitments, a named support path, board-ready reporting. A longer feature list is not an answer.
Decide how many tiers the market can hold
Three tiers when there are three distinguishable jobs, buyers can recognize theirs from the page, and Best has an executive job of its own. Two tiers and an add-on when the market has two jobs plus a compliance or scale need only some buyers share. One edition and add-ons when you launch a new module into a single segment; add tiers when usage shows where buyers split.
Related: the willingness-to-pay concept for step 4, the new-product guide for the one-edition launch, killing features for the pruning in step 2, product and pricing integration for keeping the roadmap and the tiers in one conversation, and value-based pricing for pricing against the buyer's economics.
One change before a full redesign
Your tier structure is a business claim: buyers in this segment will pay this much for this job. Before you relaunch the pricing page, move one capability across one fence and measure conversion for 30 days, or offer one add-on to a cohort of Good accounts and track attach rate. One controlled change beats any amount of competitive benchmarking and costs nothing.
Enterprise GBB vs. Consumer GBB
Consumer GBB (streaming, storage, music) differentiates on quantity and convenience: more screens, more storage, fewer ads. One person decides in under a minute. The gate has to appear after the moment of value, once the habit is forming, or the user leaves before converting.
Enterprise GBB serves the practitioner who evaluates functionality, the security team that evaluates compliance, and the executive who evaluates risk. Differentiate only on features and you address the practitioner and nobody else. Enterprise buyers tolerate gates earlier, because they are testing fit against a defined workflow; what they cannot tolerate is a locked admin console, because without it they cannot build the internal case.
Consumer GBB
- + Differentiated by quantity and convenience
- + Single decision-maker, fast purchase cycle
- + Gate after the moment of value, not before it
- + Feature ladders work because buyers self-evaluate
Enterprise GBB
- + Differentiated by organizational outcome and risk
- + Multiple stakeholders, longer purchase cycle
- + Compliance, governance, and scale drive the Best tier
- + Outcome-based differentiation required; feature ladders fail
The B2B2C layer
B2B2C adds a third requirement: the buyer and the user are different people. An employer buying a financial wellness platform decides on compliance, reporting, and administrative control; the employee decides whether the interface was built for them or bought for them. Package only for the buyer and you get high purchase rates, low adoption, and a non-renewal explained as "nobody used it." One packaging for the buyer and one experience for the user can coexist in a product, not on a pricing page. Bundling and packaging across business models covers the gating logic by segment.
Packaging Before You Scale the Sales Team
If the tiers are not working when the growth round closes, scale makes them worse faster. Discounting debt: at four reps, closing around weak differentiation with price is a habit; at twenty it is structure. Margin compression: high-cost-to-serve capabilities in a low-price tier are invisible while CS handles exceptions by hand, and catastrophic once 400 Good accounts consume enterprise support at $299. Competitive fragility: a buyer who cannot tell your tiers apart finds a competitor whose tiers make sense.
Meridian Pay, a fintech SaaS, closed a Series B at $12M ARR and hired twelve account executives without revisiting packaging. Eighteen months later it was at $28M with NRR down from 114 to 97 percent. The Better tier had been priced at $599 a month from what one customer would pay at $4M ARR; at scale that segment churned at 24 percent while reps discounted Best down to Better pricing to make quota.
Three prerequisites before the hiring push. Gross margin by tier, fully loaded with CS hours: if Good runs at 60 percent and Best at 78, those are different businesses; close or inverted, the pricing is wrong. The five capabilities most used by churned accounts in each tier, because heavy use by churners means the capability attracts the wrong buyer. A written deal desk policy before the sixth rep: approval levels, and which capabilities are add-ons versus hard fences.
How Do You Diagnose a Broken GBB Structure?
Tier mix distribution
Target: 15 to 25 percent Good, 45 to 55 Better, 25 to 35 Best. More than 60 to 70 percent in any one tier means the packaging is not segmenting, in either of the two shapes above.
Discount rate on the Better tier
If the average discount on Better runs more than five points above Good, buyers do not see the step-up value at list and reps are compensating with price. The fix is packaging redesign, not a discount ceiling. Measure it at pocket price with the price waterfall; list-price discounts understate the leak.
Organic upgrade rate from Good to Better
What share of Good customers moved to Better in the last twelve months without CS outreach? A working structure sees 15 to 25 percent a year. Below 5 percent, the fence is weak or absent.
Four baselines and the packaging ROI model
Pull four numbers before you change anything: contract value, discount rate, twelve-month NRR, and time to close, each by tier. If you cannot get clean versions in two hours, that is a finding. Then model three levers, top-tier mix, discount rate, and retention, at conservative, base, and upside cases: nine cells. If the base case is not three times the project cost, either the packaging is fine or the assumptions are timid, and the second is more common. Lock the baseline and a 30-day control period before a launch date exists.
Where the cost hides in your P&L
Bad packaging never appears as a line item. It is booked as discounts, as CS hours when a manager who could carry 80 well-matched accounts carries 50, as acquisition spend replacing accounts that churn out of a misaligned Good tier, and as reps who stop believing the list price. Northwind Projects, $28M ARR, ran 18 CS managers for 620 accounts because 38 percent sat in a Good tier built for teams with workflow infrastructure they did not have. Moving two onboarding capabilities into Good cut CS load per account by 40 percent, and the company reached $36M ARR without a CS hire: roughly $1.1M in avoided headcount.
Three wrong instincts
The most confident beliefs founders hold about their tiers tend to be the expensive ones. That more features make a better tier, when buyers pay for outcomes and not a dollar more for an integration that does not change theirs. That raising the Best price loses customers: the sample is biased, because you hear the pushback and never hear from buyers who would have paid more, so ask your last ten Best deals at what price it would have been an easy yes. That the middle tier should be the most popular: if Best holds under 20 percent of new ARR, you are collecting Better revenue from Best buyers, and the question is why they land there.
Moving Customers When You Re-Tier
A redesign fails where the new structure meets the installed base. New tiers launch on the website and in new contracts, existing accounts are grandfathered indefinitely, and sales is running two pricing systems: the deal desk does not know which policy applies, and the discount rate rises because reps discount across both.
Set the migration timeline before the launch date. Accounts move at their next renewal, and every legacy plan gets a written rule for sales: what it maps to, what it costs, when it ends. Communicate at least 90 days ahead with a value narrative for why the new structure is better for that customer. Grandfathering is a decision with an end date, not a default. The price increase playbook covers the communication sequence, and most of it transfers: a re-tier is a price change to the customer even when list price did not move.
Run a recertification pass on accounts using capabilities above their tier. Fathom Data took a business case to 82 Better accounts using Best features; within 90 days 31 had upgraded and average ARR per account moved from $8,640 to $14,400. Change the sales motion with the packaging: live deal practice before launch, the objection guide and deal desk policy updated the same week, one set of tier names in the CRM. And change one variable at a time; names, prices, fences, and policy launched together produce a number nobody can attribute.
Three Places This Framework Breaks
Three places good-better-best breaks, what it cost, and what to do instead.
Three tiers in a market with two jobs
The framework is three tiers benchmarked against the category. It breaks when buyers only have two jobs. Ridgeline Deploy, $15M ARR DevOps, built its tiers from four competitor pages: Better at $899 a month with 31 capabilities, Best at $1,499 with 47. Win rate was 42 percent in Better and 29 in Best, and nobody could say why anyone would choose Best. The buyers had two jobs: automate deployments for one team, and manage them across an organization with audit requirements. The correction: two tiers and a $400 a month compliance add-on. Win rate on the top tier moved to 41 percent and contract value rose 28 percent.
A competitor-led repackage launched without a test
The framework is the outcome test: write the claim, test one change. It breaks when competitive anxiety skips the test. Alder Field Systems, $18M ARR vertical SaaS, decided its tiers looked cheap and shipped new names, new prices, and a third enterprise tier: three months, roughly $180K of internal time. Within 60 days trial-to-paid conversion was down 14 percent, enterprise prospects picked the middle anyway, and reps defaulted to discounting. Not one assumption had been checked. The correction: one tier change, measured for 30 days, at zero incremental cost.
A redesign without the data audit
The framework is the 90-day sequence below. It breaks when the first 30 days are skipped. Corvid People, $9M ARR HR tech, took a new pricing page and feature matrix from a consultant in six weeks, built on a competitive analysis and three customer interviews. Within 90 days reps were back to the old tier names because the new ones confused prospects, the CRM and deal desk policy were untouched, and the average discount rose from 17 to 23 percent. With no gross margin by tier, capabilities had moved on competitive positioning, not unit economics. The correction: the audit first, the design only from what it shows.
The 90-Day Packaging Sequence
The first two steps need a spreadsheet and two hours with finance and product analytics, not a consultant.
- Day 1 to 30: pull the pocket price waterfall by tier
- Day 1 to 30: map capability usage by tier for the top and bottom 20 percent of engaged accounts
- Day 1 to 30: list the ten largest churned accounts with their tier and stated reason
- Day 1 to 30: calculate gross margin by tier, fully loaded, including CS hours per account
- Day 1 to 30: find the three largest expansions, what triggered each, and which tier they moved to
- Day 31 to 60: write one job sentence per tier and test whether reps can recite it without notes
- Day 31 to 60: move two or three misplaced capabilities in staging and test the pricing page message
- Day 31 to 60: write the deal desk policy, including what triggers an upgrade conversation instead of a discount
- Day 31 to 60: build the objection map and a one-page champion brief for each tier
- Day 61 to 90: run a 30-day test on one change, measuring conversion, contract value, and time to close
- Day 61 to 90: train the sales team with live deal practice, then update the pricing page, collateral, and CRM
- Day 61 to 90: set the 90-day review against the four baselines
Company names and figures in worked examples are illustrative.
Good-Better-Best Pricing: Common Questions
What is good-better-best pricing?
How many pricing tiers should a B2B SaaS have?
What goes in the base tier?
What is the right price ratio between Good, Better, and Best tiers?
How do you decide which features to fence between tiers?
How do you know your good-better-best structure is broken?
How do you move customers to a new tier structure?
Does the compromise effect work in enterprise B2B deals?
Find out if your tier structure is working
The FintastIQ Packaging Assessment scores your tier structure against the three diagnostic metrics above. Or bring the four baselines to a working session and leave with the tier boundary to test first.