# greenwald-sierra-outcome-based-pricing-ai-agents-2024-12-10

## Veille

Sierra blog post (December 10, 2024, Elliot Greenwald) laying out the **founding text of *outcome-based pricing*** for AI agents. **Pivot thesis**: AI agents that execute processes autonomously make possible an **entirely new pricing model** — ***"you pay only when the software achieves specific, valuable outcomes: outcome-based pricing."*** The article traces a **four-age genealogy of software pricing**: (1) **shrink-wrapped software** (1980s-90s, the floppy-disk/CD-ROM box at Fry's Electronics — *"Whether you actually used it or not, you paid for it"*) → (2) **SaaS / seat-based** (pioneered by **Salesforce**, followed by Google/Microsoft/Adobe — the Internet makes it possible to sell software *as a service*) → (3) **consumption-based** (**Amazon/AWS** and **Snowflake** — *"charged only for what you used"*) → (4) **outcome-based** (AI agents). **Canonical definition**: ***"outcome-based pricing is tied to tangible business impacts—such as a resolved support conversation, a saved cancellation, an upsell, a cross-sell, or any number of valuable outcomes. If the conversation is unresolved, in most cases, there's no charge."*** **Aligned-incentives principle**: ***"With outcome-based pricing, Sierra gets paid only when we complete a task for you. Our incentives are aligned."*** **Critique of seat-based pricing & the concept of *shelfware***: *"Unused seats sit idly on a proverbial store shelf, hence the derisive moniker 'shelfware'"* — thousands of dollars per year are paid per license, whether used or not. **Structural conflict for Fournisseurs CX legacy**: their revenue depends on seat-based pricing, yet *"the more effective their AI becomes, the fewer contact center seats their clients need—undermining the provider's own revenue model"* — an effective AI agent **cannibalizes** the revenue model of a vendor whose pricing rests on seats. **Granularity of the outcome**: a distinction between **simple resolutions** (answering a question) and **complex resolutions** (handling a case that requires a 20-minute L2 call); **escalations generally incur no charge**; **blended pricing** is possible (e.g., consumption-based for routing/greeting interactions). **Continuous-optimization commitment** on the vendor side: *"we continue to deploy concerted, directed optimizations to refine the agent's performance over time"* — the vendor stays aligned to improve performance since it is paid only for the outcome. Significance: posed in **late 2024**, this post **precedes and grounds** the entire 2026 debate on the agentic economy — it supplies the **vocabulary of the billing unit** (the completed *outcome* rather than the seat, usage, or token) that will later be taken up by Gupta (*cost of a completed outcome*, *token-to-outcome attribution*), Bain (*outcome-based pricing shifts revenue from fixed seats to labor/operations economics*), Ng (*pricing power anchored on the salary of the replaced employee*). With Sierra as the **reference example** cited by Bain (*autonomous customer issue resolution*), this text gives the **vendor-side view** of the mechanics that others analyze from the buyer side. Directly relevant to the firm's positioning on **agentic-delivery / value-based pricing** and to the **Cost Optimization** slot (the vendor-side counterpart of *cost per outcome*).

## Titre Article

Outcome-based pricing for AI Agents

## Date

2024-12-10

## URL

https://sierra.ai/blog/outcome-based-pricing-for-ai-agents

## Keywords

outcome-based pricing, results-based pricing, AI agents, autonomous AI agents, Sierra, seat-based pricing, shelfware, consumption-based pricing, SaaS, Software as a Service, shrink-wrapped software, Fry's Electronics, Salesforce, AWS, Snowflake, aligned incentives, incentive alignment, resolved support conversation, saved cancellation, upsell, cross-sell, unresolved no charge, Fournisseurs CX legacy conflict, contact center seat cannibalization, simple vs. complex resolution, unbilled escalation, blended pricing, mixed pricing, routing greeting interactions, continuous optimization, cost of completed outcome, billing unit, value-based pricing, customer experience, CX automation, software pricing genealogy, Elliot Greenwald

## Authors

**Elliot Greenwald** — Sierra (entreprise fondée par Bret Taylor & Clay Bavor, plateforme d'agents IA conversationnels pour l'expérience client). Billet publié sur le blog Sierra le **10 décembre 2024**. Sierra est l'**exemple-référence** cité par Bain (*The $100-Billion SaaS Opportunity*) pour l'*autonomous customer issue resolution*, et fait l'objet de plusieurs fiches du dossier (recrutement AI-native, interview Plan/Build/Review).

## Ton

**Profile**: A corporate-pedagogical blog post aimed at CX buyers / decision-makers (VP Customer Experience, COO, CFO) and the product community. An assumed "we" (Sierra) perspective, a **clear and didactic** register, low technical depth — the argument is **economic**, not technical. Short format (a pricing manifesto, ~5 min read).

**Style**: Storytelling through **historical genealogy** (the floppy-disk box at Fry's → SaaS → consumption → outcome) — a classic device that makes *outcome-based pricing* appear as the **natural culmination** of an evolution rather than a self-interested commercial argument. Polished marketing prose: each pricing age is summed up in an image-phrase (*"shelfware,"* the usage-era version of *"the meter is running"*). **Prescriptive, self-positioning** tone: the article implicitly sells the Sierra model while presenting it as the client's aligned interest (*"Our incentives are aligned"*). Tactical honesty: it acknowledges granularity (simple vs. complex resolutions, unbilled escalations, blended pricing) to lend credibility to the model's fairness.

**Key aphorisms**:
- ***"you pay only when the software achieves specific, valuable outcomes: outcome-based pricing."*** (pivot definition).
- ***"With outcome-based pricing, Sierra gets paid only when we complete a task for you. Our incentives are aligned."*** (alignment principle).
- ***"If the conversation is unresolved, in most cases, there's no charge."*** (the shared-risk promise).
- ***"Unused seats sit idly on a proverbial store shelf, hence the derisive moniker 'shelfware.'"*** (critique of seat-based pricing).
- ***"the more effective their AI becomes, the fewer contact center seats their clients need—undermining the provider's own revenue model."*** (the structural conflict facing legacy vendors).

**Crafted metaphors**:
- ***Shelfware*** — unused seats *"sitting idly on a proverbial store shelf"*: an image that turns unused fixed cost into visible waste, set against the outcome-based model.
- ***The floppy-disk box at Fry's*** — a nostalgic anchor that makes tangible the idea of *"paying whether you use it or not,"* the starting point of a trajectory toward strictly value-based payment.
- ***Incentives aligned*** — the rhetorical pivot: outcome-based pricing is presented not as a pricing grid but as a **structural alignment** between vendor and client (the vendor only wins if the client gets the outcome).

**Epistemic stance**: a **business-model** manifesto that posits a pricing category (outcome-based) as the culmination of a genealogy. A **normative** argument (this is the right model) served by a **historical** narrative (this is the inevitable model).

**Authority**: built through (a) **Sierra's position** as a pure-player CX AI-agent company (Bret Taylor), (b) the **genealogical narrative** that places the proposition in continuity with Salesforce/AWS/Snowflake, (c) the **consistency of "skin in the game"** (*we only get paid when you get the outcome*), (d) the **granularity** (simple/complex resolutions, escalations, blended pricing) that defuses the fairness objection.

## Pense-betes

- **Date / source**: **December 10, 2024**, Sierra blog, Elliot Greenwald. **Founding** text of outcome-based pricing — roughly 18 months ahead of the 2026 debate (Gupta, Bain).
- **Pivot thesis**: autonomous AI agents make possible a 4th era of pricing — ***"you pay only when the software achieves specific, valuable outcomes."*** ### The genealogy across 4 ages | Age | Model | Unit | Example | Waste | |-----|--------|-------|---------|-----------| | 1980s-90s | **Shrink-wrapped** | the box | Fry's Electronics (floppy disks/CD-ROMs) | total (paid whether used or not) | | Internet | **SaaS / seat-based** | the seat (license) | Salesforce, Google, Microsoft, Adobe | high (**shelfware**) | | Cloud | **Consumption-based** | usage | AWS, Snowflake | moderate | | AI agents | **Outcome-based** | the **completed outcome** | Sierra | low | ### The definition to remember > ***"outcome-based pricing is tied to tangible business impacts—such as a resolved support conversation, a saved cancellation, an upsell, a cross-sell, or any number of valuable outcomes. If the conversation is unresolved, in most cases, there's no charge."*** ### The structural conflict facing legacy CX vendors
- Legacy revenue = **seat-based** (thousands of $/year per license).
- Yet *"the more effective their AI becomes, the fewer contact center seats their clients need"* → **a good AI agent cannibalizes the vendor's revenue model**.
- Sierra (a pure-player) has no such conflict: paid on outcomes, it benefits from **reducing** the need for seats. ### The model's granularity (fairness)
- **Simple resolution** (answering a question) vs **complex resolution** (case type: a 20-min L2 call) → priced differently.
- **Escalations**: *"typically incur no charges."*
- **Blended pricing** possible: e.g., consumption-based for **routing/greeting** interactions, outcome-based for resolutions.
- **Continuous optimization**: *"we continue to deploy concerted, directed optimizations to refine the agent's performance over time"* — durable alignment. ### ⚠️ What the article does NOT contain
- **No figures**: no ROI, no client metrics, no named clients (not even Klarna), no precise pricing. It is a **conceptual manifesto**, not a case study. ### To be drawn on for
- **"Cost Optimization" slot (Claude Code morning session)**: the **vendor-side view** of *cost per completed outcome* — a perfect complement to Gupta's *cost of a completed outcome* (buyer-side view). Use it to explain why the KPI becomes the outcome, not the token/seat.
- **Agentic-delivery pricing positioning (firm)**: canonical reference for proposing **value-based / outcome-based** models on engagements, vs. staff-augmentation/day-rate billing.
- **CX decision-maker messaging**: the **shelfware** concept + incentive alignment (*"we get paid only when we complete a task"*) = simple, memorable arguments. ### Connections within the watch file
- **Gupta — Token Budget Wars** (2026-05-28): the exact **buyer-side counterpart** — *"what is the cost of a completed outcome?"*, *token-to-outcome attribution*. Sierra (2024) lays out the **billing model**, Gupta (2026) lays out the **measurement problem** that makes it operable. Gupta also cites *"which replace BPO"* — Sierra is this type of vendor.
- **Bain — cross-system labor** (2026-05) & **Rule of 40** (2026-04): Bain cites **Sierra** as the reference example and describes **outcome-based pricing** as shifting revenue *from fixed seats to labor/operations economics* — Greenwald is the **primary source** of this shift.
- **Ng — No AI jobpocalypse** (2026-05-08): *pricing power* (vendors anchor pricing on the **salary of the replaced employee**) — Sierra's outcome-based pricing is precisely this anchoring (a resolution = a unit comparable to the cost of a human agent/BPO).
- **VoxComm / MediaPost — billable hours are dead** (2026-03): the same *value-based / performance-based* shift in knowledge services (agencies) — a cross-sector convergence.
- **Hezarkhani — Paying Engineers like Salespeople** (2025-11): compensation-to-outcome alignment, another variant of *skin in the game*.
- **Sierra (file)**: Bret Taylor's *AI-native interview* (2026-04-20) + Iyengar/Asemanfar/Wang (2026-04-22) — this post completes the Sierra portrait on the **business-model** side.

## RésuméDe400mots

On December 10, 2024, **Elliot Greenwald** (Sierra) published the **founding text of *outcome-based pricing*** for AI agents. **Pivot thesis**: AI agents executing processes autonomously make possible an unprecedented pricing model — ***"you pay only when the software achieves specific, valuable outcomes: outcome-based pricing."***

The article unfolds a **four-age genealogy**. (1) **Shrink-wrapped software** (1980s-90s): the floppy-disk/CD-ROM box at Fry's Electronics — *"Whether you actually used it or not, you paid for it."* (2) **SaaS / seat-based**: Salesforce pioneering, followed by Google/Microsoft/Adobe; major flaw = ***shelfware*** (*"Unused seats sit idly on a proverbial store shelf"*). (3) **Consumption-based**: AWS and Snowflake — *"charged only for what you used."* (4) **Outcome-based**: AI agents.

**Canonical definition**: *"outcome-based pricing is tied to tangible business impacts—such as a resolved support conversation, a saved cancellation, an upsell, a cross-sell, or any number of valuable outcomes. If the conversation is unresolved, in most cases, there's no charge."*

**Alignment principle**: *"With outcome-based pricing, Sierra gets paid only when we complete a task for you. Our incentives are aligned."* Greenwald highlights the **structural conflict facing legacy CX vendors**: their revenue depends on seat-based pricing, yet *"the more effective their AI becomes, the fewer contact center seats their clients need—undermining the provider's own revenue model."* An effective AI agent **cannibalizes** the seat-based revenue model; a pure-player paid on outcomes has no such conflict.

The model is **granular**: it distinguishes **simple resolutions** (a single question) from **complex** ones (case type: a 20-min L2 call); **escalations** generally incur no charge; **blended pricing** is possible (consumption-based for routing/greeting). On the vendor side, a commitment to **continuous optimization**: *"we continue to deploy concerted, directed optimizations to refine the agent's performance over time."*

**Scope**: posed in late 2024, this post **precedes and grounds** the 2026 debate on the agentic economy. It supplies the vocabulary of the **billing unit** (the completed outcome, not the seat/usage/token) that will later be taken up by **Gupta** (*cost of a completed outcome*, buyer-side view), **Bain** (outcome-based pricing shifts revenue *from fixed seats to labor/operations economics*, Sierra cited as an example), and **Ng** (pricing anchored on the replaced salary). ⚠️ The article contains **no figures** (no ROI, no named client): it is a conceptual manifesto. To be drawn on for the **Cost Optimization** slot (vendor-side view of *cost per outcome*) and the **value-based** positioning of agentic delivery.

## GrapheDeConnaissance

- Elliot Greenwald —publie→ Outcome-based pricing for AI Agents (DOCUMENT, 0.97)
- Sierra —recommande→ outcome-based pricing (CONCEPT, 0.97)
- Agents IA —permet→ outcome-based pricing (CONCEPT, 0.95)
- Outcome-based pricing —s_applique_à→ business outcomes tangibles (resolved conversation, saved cancellation, upsell, cross-sell) (CONCEPT, 0.96)
- Sierra —affirme_que→ « If the conversation is unresolved, in most cases, there's no charge » (CITATION, 0.94)
- Outcome-based pricing —permet→ Alignement des incitations (CONCEPT, 0.95)
- Salesforce —a_créé→ modèle SaaS (seat-based) (CONCEPT, 0.95)
- Seat-based pricing —permet→ Shelfware (CONCEPT, 0.94)
- AWS et Snowflake —a_créé→ consumption-based pricing (CONCEPT, 0.94)
- Fournisseurs CX legacy —s_oppose_à→ efficacité de leur propre IA (réduit les sièges) (CONCEPT, 0.93)
- Résolution complexe —est_variante_de→ résolution simple (tarification différenciée) (CONCEPT, 0.9)
- Sierra —affirme_que→ les escalades n'entraînent en général aucune facturation (AFFIRMATION, 0.9)
- Sierra —utilise→ optimisation continue de la performance de l'agent (METHODOLOGIE, 0.92)
- Cost of a completed outcome (Gupta) —est_basé_sur→ outcome-based pricing (CONCEPT, 0.88)

---
Canonical: https://www.thekb.eu/en/fiches/greenwald-sierra-outcome-based-pricing-ai-agents-2024-12-10/