05 — Phase 2A: Pricing & Packaging
Goal: choose the pricing model that best supports adoption, retention, and revenue across markets with different willingness-to-pay — and package it onto the three tiers from doc 04.
The verdict: a hybrid model (subscription anchor + usage + seats)
2025 benchmarks are decisive: pure per-seat is in decline (57% and falling; ~$45/seat/mo median), usage-based is now mainstream (61% of B2B SaaS), and hybrid pricing posts the highest median growth (~21%). For a data/insights platform whose cost and value both scale with consumption (markets analyzed, concepts generated), hybrid is the natural fit.
The three components and what each is for
| Component | What it meters | Role | Why |
|---|---|---|---|
| Platform subscription (per tier) | Access to a tier | Revenue anchor & predictability | Enterprises want budgetable, committed spend |
| Usage / credits | Concept Genie runs, # markets, # categories, refresh frequency | Value-metric & expansion | Aligns price to value; powers NRR; lets small buyers start cheap |
| Seats | Named users | Secondary expansion lever | Land small, grow with adoption — not the primary meter |
Value metric = markets × categories × concept-generation usage. This is the single most important pricing decision: we charge for breadth of intelligence consumed, which is what actually drives Ai Palette's cost-to-serve and the customer's value.
Why not the alternatives (the trade-offs)
- Pure per-seat: caps revenue (insights are consumed by few, valued by many), penalizes adoption, and is structurally declining. ✗ as primary.
- Pure usage: unpredictable for enterprise budgeting and for our revenue; creates bill-shock and discourages exploration. ✗ as primary, ✓ as a layer.
- Pure flat enterprise subscription: simple but leaves expansion money on the table and can't flex to low-WTP markets. ✓ only at the very top, wrapped with usage upside.
Hybrid captures the best of each: predictable base + value-aligned expansion + adoption-friendly entry.
Packaging onto tiers (illustrative list prices — see disclaimer)
Illustrative, benchmark-anchored [AB], not Ai Palette's actual prices. Purpose: show the structure and ratios, which is what matters. Anchored to enterprise insights-SaaS ACVs ($30K–$120K, Assumption #1) and PLG/SLG ACV thresholds (doc 06).
| Tier | Subscription (annual, US benchmark) | Included usage | Overage / expansion | Target ACV |
|---|---|---|---|---|
| Essentials | ~$6–10K | 1 market, ~3 categories, low Concept Genie credits | buy more markets/credits | < $10K |
| Growth | ~$25–45K | several markets/categories, metered Concept Genie | markets, categories, credits, seats | $25–60K |
| Enterprise | ~$80–150K+ | all markets/categories, high usage, integrations, residency | custom data, services, volume | $80K+ |
Ratios matter more than absolutes: ~1 : 4 : 12 across tiers, with usage providing 15–30% incremental within each tier.
Regional willingness-to-pay (the same product, repriced — not rebuilt)
WTP ordering (Assumption #3): US > W. Europe > Japan/Korea > SEA > India. We do not change the product or tier structure by region — we apply regional price multipliers and shift the tier mix we lead with.
| Region | Price index (US=1.00) | Lead motion / tier emphasis |
|---|---|---|
| US | 1.00 | Enterprise-led; highest ACV |
| W. Europe | 0.85–0.95 | Enterprise + Growth; compliance is a selling point |
| Japan/Korea | 0.75–0.90 | Enterprise via partners; high service expectation |
| SEA | 0.45–0.65 | Growth-led; Essentials to widen funnel |
| India | 0.35–0.55 | Essentials/Growth-led; volume play, PLG-assisted |
This is price localization, not product localization — it protects margins in high-WTP markets and protects adoption in price-sensitive ones, using one codebase.
Retention & expansion design
- Target blended NRR 110%+ (Assumption #6) via four upsell levers: markets, categories, usage, seats.
- Annual contracts at Growth/Enterprise for predictability; monthly option at Essentials to lower entry friction.
- Guardrail: keep CAC payback ≤ 12–18 months (median is ~20mo; we beat it via PLG-assisted Essentials and land-and-expand). See doc 06.
Machine-readable version: frameworks/pricing-model.csv.