Saas Pricing Auditor

Audit a SaaS pricing page and return a scorecard, specific rewrites, and an A/B test priority queue. Diagnoses tier psychology (3-tier vs 4-tier, where the anchor sits), price points (charm vs round, currency formatting, decoy effect), feature-gating logic (the right and wrong reasons to gate a feature), the freemium-vs-trial decision, value-metric selection (per-seat, per-usage, hybrid, outcome-based), expansion-revenue mechanisms (overages, add-ons, tier upgrades, seat sprawl), monthly/annual toggle defaults, comparison-table clarity, and social-proof placement. Outputs a 1-10 score per dimension with specific rewrites and a ranked test backlog. Use when asked to review a pricing page, redesign tiers, raise prices, fix poor conversion, choose a value metric, decide between freemium and trial, or prep for a pricing repackage.

Most SaaS pricing pages were drafted in a tab once and never revisited. They feel reasonable, convert below industry median, and quietly leak 20–40% of potential ARR through bad gating, weak anchoring, and a value metric that punishes power users. This skill audits the page on nine dimensions, scores each one, and ships specific rewrites plus a ranked test backlog.

Usage

Basic invocation:

Audit my pricing page: [URL or paste] Should I be 3-tier or 4-tier? Help me pick a value metric — I'm currently per-seat Free trial or freemium? Repackage from $19/$49/$99 to higher ACV

With context:

B2B SaaS, $40k ARR, single $49/mo tier, no annual PLG product, 8% free→paid, want to lift to 12% Mid-market sales-led, $25k ACV, comparing to Salesforce Vertical SaaS for dental clinics, 3 tiers, low expansion revenue

The auditor returns a scorecard, prioritized rewrites, and the next 3–5 tests to ship.

The Nine Audit Dimensions

Each dimension scored 1–10 with notes and a fix.

1. Tier architecture — number, names, anchor placement
2. Price points — charm pricing, round vs odd, currency display
3. Feature gating logic — what's gated and why
4. Free trial vs freemium — which model, parameters
5. Value metric — what you charge per
6. Expansion-revenue mechanisms — how revenue grows post-sale
7. Monthly/annual toggle — default, discount, framing
8. Comparison table clarity — readability, visual weight
9. Social proof placement — logos, quotes, numbers, near CTA

The order is the priority order. Architecture errors swallow everything below them — fixing button color while your value metric is wrong is rearranging deck chairs.

Dimension 1: Tier Architecture

3-tier vs 4-tier vs 5-tier

Count	When it works	When it fails
1 tier	Self-serve under $50/mo, dead-simple product	Anything multi-segment
2 tiers	Pure PLG with one upgrade path	Mid-market+
3 tiers	Default for SMB SaaS — Good/Better/Best	When segments diverge dramatically
4 tiers	When Enterprise needs custom pricing on top of 3	Otherwise; adds choice paralysis
5+ tiers	Almost never; only with strong segmentation	Most SaaS — analysis paralysis

Default recommendation: 3 tiers + custom Enterprise = 4 visible "options" without overwhelming.

Anchor tier (the most-read tier)

The center tier in a 3-tier display gets ~60% of visual attention and should:

• Be marked "Most popular" with a colored border
• Contain the features the median customer needs (so they don't have to upgrade for table stakes)
• Be priced at the sweet spot of your contribution-margin curve
• Hide enough power-user features to leave room for the top tier

Anchor errors to flag:

• "Most popular" badge on the cheapest tier (signals to prospects that paying more is rare)
• "Most popular" badge on the most expensive tier (read as upsell theater)
• No badge at all (you're letting the visitor pick blind)

Tier names

Names that work: Starter, Pro, Business, Enterprise / Free, Plus, Pro, Scale / Solo, Team, Company

Names that don't: Bronze, Silver, Gold (cliché, no signal); Basic (sounds like punishment); Premium (vague); Tier 1, 2, 3 (reads as internal ops).

The middle tier name should imply "this is the right answer for normal businesses" — that's why "Pro" and "Business" outperform "Standard" and "Plus."

Dimension 2: Price Points

Charm pricing (.99) vs round vs odd

Format	Where it works	Why
$19	Self-serve, low end	Reads cheap and confident
$19.99	Consumer / very low end	Squeezes loss-aversion bias
$29 / $49 / $99	SMB SaaS sweet spot	Round = serious; odd = approachable
$49	Common anchor SMB	The "tax floor" perception break
$50	Slightly worse than $49	Round numbers are processed as estimates
$499 / $999	Mid-market	Charm pricing in 3-digit range still moves the needle
$1,000+	Enterprise	Round numbers; charm pricing reads cheap
Custom / Talk to sales	Above ~$30k ACV	Forces qualification

The decoy effect: when tier B is meaningfully better than A and almost the same price as C, B captures most volume. Common pattern:

• Tier A: $19/mo — 5 seats, basic features
• Tier B: $49/mo — 25 seats, full feature set ← decoy-driven anchor
• Tier C: $99/mo — unlimited seats, integrations

Tier B is the trap. Make sure the delta from A to B is large enough that A looks bad-value, and the delta from B to C is small enough that C looks like overkill for most.

Currency display

• Always show currency symbol explicitly ($49, not 49)
• Show "/month" or "/mo" with every price; never assume the period is obvious
• For per-seat pricing, show "per user/mo" — never just "/mo" (creates sticker-shock at checkout)
• For annual: show monthly-equivalent ("$49/mo, billed annually") — annual lump sum scares small buyers

Dimension 3: Feature Gating Logic

The hardest pricing decision. Wrong gating either leaves money on the table (everyone gets the upgrade feature for free) or kills conversion (must-have features hidden behind enterprise tier).

Good reasons to gate

• Power-user-only features — API, SSO, audit logs, custom roles. Most users don't need them; those who do pay for them.
• High variable cost — features that cost you per-use (AI calls, transactional emails over a threshold)
• Compliance/security — SOC 2 reports, BAA, custom DLP. Buyer of these is enterprise; gate accordingly.
• Volume scaling — seats, projects, storage. Linear cost = linear gate.

Bad reasons to gate

• It feels premium — gating UI polish (dark mode, custom themes) annoys people without driving upgrades
• Competitor does it — they may be wrong; copying gates copies their conversion ceiling too
• You spent engineering time on it — sunk-cost fallacy; gate value not effort
• Table-stakes features — gating CSV export, basic search, or 2FA punishes everyone and signals greed

The gating decision tree

Is the feature's marginal cost > $0?  → Gate by usage (overage or tier)
Is it a power-user / admin feature?    → Gate to higher tier
Is it a compliance/security feature?    → Gate to enterprise
Is it consumed in a measurable unit?    → Use as the value metric (don't gate; meter)
Otherwise                                → Don't gate; ship to all tiers

Specific anti-patterns to flag in audits:

• SSO behind enterprise-only when the rest of the product is $99/mo (the "SSO tax" — well-known pricing sin)
• API access at the highest tier only (cuts off integrations that drive expansion)
• Hard caps on seats with no overage path (forces a contract negotiation for +1 user; high friction)
• Feature counts as the differentiator instead of usage (drives feature-comparison fatigue)

Dimension 4: Free Trial vs Freemium

Decision matrix

Factor	Freemium wins	Free trial wins
Time-to-value	Long (days/weeks)	Short (minutes)
Network effects	Strong	Weak
Marginal user cost	Near zero	Non-trivial
Buyer is end-user	Yes	Mixed
Sales motion	Pure PLG	PLG + sales-assist
Aha moment requires data	Yes (real workspace)	No (demo data fine)

Freemium parameters that work

• Free tier should solve a real, narrow problem completely — not a crippled version of the paid product
• Limits should hit at the moment users stop being hobbyists (number of projects, seats, integrations)
• Convert at the aha moment, not at month boundary
• Expect 2–5% free-to-paid conversion as median; 5–10% is excellent

Free trial parameters that work

• 14 days is the default; 7 days for simple products; 30 days only for complex products with onboarding
• Require credit card if conversion intent is high; don't if you're top-of-funnel
• Trial-to-paid conversion: 15–25% no-CC, 40–60% with-CC
• Reverse trial (full features for N days, then downgrade to free) increasingly common; works for engagement-led products

Anti-patterns

• Trial that requires CC + auto-charges silently (= chargeback magnet, bad reputation)
• Freemium with no path to paid (the "free forever" trap)
• 30-day free trial on a $19/mo product (you've given away ~10% of annual revenue per signup)

Dimension 5: Value Metric

The single highest-leverage decision in pricing. Get it right, expansion revenue is automatic; get it wrong, you cap your account growth at "reset the contract."

The value metric test

A good value metric:

1. Scales with customer value — when they get more out of the product, the metric grows
2. Is predictable to the buyer — they can estimate their bill before signing
3. Aligns incentives — your growth doesn't punish their adoption
4. Has expansion built in — accounts naturally grow on the metric

Common metrics, ranked

Metric	Pros	Cons	Best for
Per-seat	Predictable, SaaS-default	Caps at team size; punishes adoption	Collaboration, internal-tool SaaS
Per-usage (events, API calls, runs)	Scales with value, unbounded	Unpredictable bills, sticker shock	Infrastructure, AI, analytics
Per-record (contacts, customers)	Clear ROI mapping	Customers prune to save money	CRM, marketing
Per-revenue	Truly aligned	Hard to verify, requires trust	Payments, embedded fintech
Flat tiers	Predictable	Zero expansion mechanism	Very early or very low ACV
Hybrid (seats + usage)	Captures both	Complex to explain	PLG with team sprawl + power-use
Outcome-based	Maximally aligned	Hard to attribute, slow sales	Performance marketing, AI agents

How to choose

• If usage causes customer value (calls made, jobs run, data ingested) → meter it
• If a team is the unit of value → seats
• If a power-user does 10x what a casual user does → hybrid (seats + usage)
• If your customer is a small business and predictability matters → flat tier with usage caps

Repackaging from per-seat to hybrid is the most common 2025–2026 move; AI features have value-per-call that breaks pure-seat models.

Dimension 6: Expansion-Revenue Mechanisms

Net revenue retention >110% is the difference between a 4x and a 10x business. The pricing page is where you build the mechanism.

The four levers

1. Overages — soft caps with metered overages (e.g., "10k API calls included; $0.001/call after"). Best when usage is variable.
2. Add-ons — separately priced modules (extra storage, premium support, additional brands). Best when there are clear segments.
3. Tier upgrades — natural progression as the customer grows. Requires tier deltas to feel earned.
4. Seat/license sprawl — auto-add seats when invited, with admin approval. The most reliable expansion mechanism for collab tools.

Expansion audit checklist

• Does any tier have a hard cap with no overage path? (= contract friction)
• Are there add-ons visible on the pricing page? (= optional, non-disruptive expansion)
• Does the metric grow naturally with customer success?
• Is there a clear upgrade trigger (UI nudge when nearing a limit)?
• Does annual billing kill expansion? (Quarterly true-ups solve this for usage-based.)

Dimension 7: Monthly/Annual Toggle

Default state

Default the toggle to annual if your business depends on annual cash. Defaulting to monthly is leaving 20–30% of contract value on the table for buyers who would have taken annual if it were the default option.

Discount sizing

• 10% off annual: too small to motivate
• 15–20% off annual: industry standard, works
• 25%+ off annual: strong signal; use when cash flow matters
• 2 months free (≈17%): same as 17% but feels more concrete

Framing

• "$49/mo, billed annually" reads as $49/month
• "$588/year" reads as a big number
• Always show monthly-equivalent next to the annual price
• Show the savings explicitly: "Save $118/yr"

Dimension 8: Comparison Table

Common errors

• Every feature row checked for every tier (no actual differentiation visible)
• Cryptic "✓" / "—" / "Limited" with no explanation
• Long feature lists (>20 rows) that no visitor reads
• Hiding meaningful limits in tooltips
• Putting the cheapest tier on the left (visitors read left-to-right; anchor on the right)

What works

• 8–15 rows max; group into sections
• Use numbers, not just checks ("5 projects", "Unlimited", not "✓")
• Bold the differentiating features per tier
• Sticky headers on long tables
• Mobile collapse: each tier as its own card

Dimension 9: Social Proof Placement

Where it goes determines whether it converts.

Placement	Effect
Hero (above tiers)	Brand legitimacy; helps prospects believe pricing
Just below CTAs	Closes the deal; place the strongest quote here
Inside enterprise tier	Reinforces the "real companies pay this" cue
Footer-only	Wasted; nobody scrolls there before deciding

Types ranked

1. Logos of recognizable customers — fastest credibility
2. Numerical proof ("Used by 12,000 teams", "$2B processed", "ROI 4.2x")
3. Specific quotes with title and company — beats generic testimonials by 3–5x
4. Case study link near tier — for sales-led mid-market

Scoring Rubric

Score 1–10 per dimension:

• 1–3: actively harmful or missing
• 4–6: present but unoptimized
• 7–8: solid; minor improvements
• 9–10: best-in-class; nothing to fix

A typical SMB SaaS audit lands around 5.5 average. Above 7.5 average = strong page. Below 4.5 average = repackaging project, not a CRO project.

A/B Test Priority Queue

Tests ranked by typical lift × ease of implementation:

1. Default annual toggle — 5–15% ARR lift; one CSS change
2. Add "Most popular" badge to anchor tier — 5–10% mix shift to anchor; trivial
3. Reorder tiers (anchor in middle, expensive on right) — 3–8% conversion; layout change
4. Move SSO out of enterprise-only — 5–15% conversion lift on mid-tier; pricing change
5. Add usage overage instead of hard cap — 10–20% NRR lift; product+billing work
6. Replace "Talk to sales" on enterprise with starting price — increases qualified inbound; one-line change
7. Replace generic testimonial with named-company quote near CTAs — 3–7% conversion
8. Switch to charm pricing on the anchor tier — 1–4% conversion; trivial

Output Format

The auditor returns:

1. Scorecard — 1–10 on all nine dimensions with one-line rationale per
2. Top three architecture issues — the highest-leverage problems first
3. Specific rewrites — exact tier names, prices, headlines, feature names to change
4. Value-metric recommendation — current metric, recommended metric, migration path
5. Test backlog — ranked queue of 5–10 A/B tests with hypothesis and lift estimate
6. Repackaging plan (if needed) — when fixes can't be done via tests; phased rollout
7. Quick-wins list — changes that can ship today without a meeting