stigg-pricing-expert

Stigg Pricing Expert — Pick the Right Monetization Model

This skill helps the user choose the right monetization model before they start modeling it in Stigg. It's advisory — it asks diagnostic questions, maps answers to Stigg's supported pricing models, and then hands off to stigg-pricing-modeling for execution.

What this skill does NOT do: model plans, create features, configure charges, or write code. It surfaces the right shape of the pricing model and points the user at the implementation skill.

Before You Start

Per the umbrella stigg skill: search first. Stigg's supported pricing models evolve (recent additions include credit-based monetization with auto-recharge, custom credit consumption formulas, seat-based credit pools). Confirm what's currently supported before recommending.

The Process

1 — Clarify what's being priced

Ask if not stated:

• What does the product do for the customer? (Outcome they pay for.)
• Who is the buyer? (End user, team admin, procurement?)
• What's the granularity of value? (Per request, per seat, per project, flat access?)
• Is the product already in market, or pre-launch? (Pricing-changes velocity differs.)

2 — Identify the value metric

The value metric is the thing that, when more of it happens, the customer gets more value. Picking the right value metric is the highest-leverage decision in this whole process. Discovery questions, common metric patterns by domain, and how to handle multiple competing metrics: references/value-metric-discovery.md.

Common value metrics by domain:

Domain	Likely value metric
Collaboration tool	Active users / seats
Storage / cloud infra	GB stored, GB transferred, compute-hours
AI / LLM apps	Tokens, generations, model calls
Marketing / outreach	Emails sent, contacts, conversions
Analytics	Events ingested, dashboards, queries
API products	API calls, requests, throughput
Vertical SaaS	Domain-specific (transactions processed, employees managed, properties listed)

Ask:

• Where does usage scale with value to the customer? That's a candidate metric.
• What would the customer pay more / less for if it grew? Validates the metric.
• Is the metric measurable in real time? (If not, switch to a proxy or aggregate-after-the-fact model.)

If multiple metrics surface, you can model a hybrid (see below). Don't try to monetize on every dimension — pick one or two primary, the rest as soft caps.

3 — Map onto Stigg's supported models

Once the value metric is clear, map onto a Stigg pricing model:

If…	Use…	Implementation skill
Predictable monthly fee, one user paying	Flat fee	stigg-pricing-modeling (Plan + flat charge)
Pay per resource selected (seats, environments, projects)	Per-unit	stigg-pricing-modeling (Per-unit charge)
Usage scales unpredictably; bill what was used	Pay-as-you-go	stigg-pricing-modeling (Usage-based charge)
Customer commits to N units up front, true-up after	In-advance commitment	stigg-pricing-modeling
Floor on revenue regardless of usage	Minimum spend	stigg-pricing-modeling
Usage-based with included quota	Quota + overages	stigg-pricing-modeling (overages section)
Variable-cost workloads (multi-model AI, mixed inputs)	Credits with custom formula	stigg-credits (formulas)
Prepaid / pay-before-you-use (OpenAI-style)	Credits with auto-recharge	stigg-credits (auto-recharge)
Team product where credits should grow with seats	Seat-based credit pools	stigg-credits (seat pools)
Multiple revenue streams in one offering (e.g., flat base + usage)	Hybrid — combine flat / per-unit / usage charges on one plan	stigg-pricing-modeling (combining charges)
Free tier to drive top-of-funnel	Freemium plan	stigg-pricing-modeling (Free plan)
Time-bound preview of paid features	Free trial	stigg-pricing-modeling + stigg-subscriptions (trials)
Negotiated, sales-led pricing per customer	Custom plan	stigg-pricing-modeling (Custom plan)
Multi-region pricing	Layer price localization on top of any model	stigg-pricing-modeling/references/price-localization.md

The decision tree with edge cases lives in references/pricing-model-decision-tree.md.

4 — Sanity-check, then hand off

Sketch revenue at median and top-decile usage; confirm the bill is predictable enough for the buyer's segment (self-serve customers can't reason about graduated overages on top of credits); confirm an expansion path exists (Free → Pro → Enterprise should feel inevitable); flag Finance pitfalls (non-zero cost basis on promotional grants, hand-rolled discounts outside Stigg coupons, per-customer pricing without custom plans). The "Done" checklist below names the seven things the user should know before you route to implementation.

5 — Hand off to implementation

Once a model is chosen, route the user explicitly:

• Catalog modeling (features, plans, addons, charges) → stigg-pricing-modeling.
• Credits-based monetization → stigg-credits.
• Subscription provisioning / lifecycle → stigg-subscriptions.
• Drop-in UI → stigg-widgets.
• End-to-end recipes (freemium / hybrid / credits) → stigg-recipes.

Do not start authoring catalog config in this skill — that's the implementation skill's job. The full hand-off checklist (what counts as a "done" recommendation, which target skill to route to per concern, how to phrase the hand-off message): references/handoff-to-modeling.md.

Heuristics — When to Pick What

Per-seat vs usage-based

• Per-seat when value scales linearly with users; admin buys for the team; predictable budgeting matters.
• Usage-based when value scales with consumption regardless of who's consuming; customers expect to pay only for what they use.
• Both (hybrid) when both axes matter — e.g., per-seat base + per-call usage on top.

Credits vs raw usage charges

• Credits when:
  • Costs vary by workload (LLM models, file types, regions).
  • Customers benefit from flexibility (one balance for many features).
  • You want prepaid / commitment / promotional mechanics.
• Raw usage when:
  • Pricing is uniform per unit.
  • Customers think of the metric directly (API calls, GB).
  • Simplicity wins over flexibility.

Freemium vs free trial

• Freemium — perpetual free tier with limits. Drives top-of-funnel; customers convert when they hit limits.
• Free trial — time-bound preview of paid features. Drives conversion through urgency and exposure to value.
• Both — freemium for low-end discovery; trial for upgrading freemium users to paid.

Self-serve vs sales-led

• Self-serve — published plans, paywall, checkout. Best for SMB / individual users.
• Sales-led — custom plans, sales reps configuring pricing per deal. Best for enterprise.
• Both — single product, multiple GTM motions. Stigg supports this — public plans + a custom plan tier.

What Counts as a "Done" Recommendation

Before handing off, the user should know:

1. The value metric they'll meter on.
2. The plan structure — free / paid / custom; how many tiers.
3. The charge model per plan (flat / per-unit / usage / credits).
4. Whether add-ons extend plans, and which plans they're compatible with.
5. Whether credits are involved and which currencies.
6. Whether trials apply and for how long.
7. Whether price localization applies and which markets.

If any of these are still ambiguous, ask one more question before handing off.

When NOT to Use This Skill

• The user has already decided on a model and wants to implement it → go directly to stigg-pricing-modeling (or stigg-credits for credits).
• The user is asking about subscription lifecycle (provisioning / cancellation) → stigg-subscriptions.
• The user is asking about runtime entitlement enforcement → stigg-entitlements.
• The user wants a market-specific pricing study (competitive analysis, willingness-to-pay surveys) → that's outside Stigg; this skill is about mapping intent onto Stigg's supported models.

Common Mistakes

Mistake	Fix
Recommending a model without surfacing the value metric	Every tier becomes arbitrary. Surface it first.
Mapping competitor pricing 1:1	Anchor on the customer's value metric, not benchmarks.
Authoring catalog config in this skill	Hand off — modeling lives in stigg-pricing-modeling.
Recommending credits because "the user mentioned AI"	Credits fit when costs vary by workload. Uniform-cost AI works fine on raw usage charges.
Recommending hybrid by default	Hybrid adds complexity. Default to one primary model unless two metrics genuinely matter.
Skipping the median + top-decile sanity check	Pricing that breaks at scale gets discovered by your highest-value customers. Sketch the math.
Treating Stigg as the source of pricing strategy	Stigg implements pricing models; it doesn't pick the right one. That's this skill's job.