Zendesk's Outcome-Based AI Pricing Puts CRM Vendors on the Hook for ROI

Vendors Now Share Execution Risk

Zendesk's outcome-based pricing for AI agents, introduced in August 2024, charges customers only for interactions the AI autonomously resolves. The company states customers "will only incur costs for issues that are resolved autonomously by AI." This pricing model directly pressures Salesforce, ServiceNow, Microsoft, and other CRM vendors still using seat-based or consumption-based pricing for AI modules. For enterprise buyers, this changes budget planning: vendors must prove ROI before collecting payment, but buyers need precise definitions of what counts as "resolved" to avoid billing disputes.

The shift matters because it moves CRM evaluation from feature comparison to measurable automation outcomes. Buyers can now demand that vendors share execution risk rather than paying for capability that may or may not deliver value. The challenge is measurement: without clear agreement on resolution criteria, outcome-based pricing creates new contract negotiation friction around what qualifies as a successful interaction versus an escalation or partial answer.

Salesforce and ServiceNow Respond With Configurable Agents

Salesforce released Agent Builder, which lets users customize pre-built agents or create new agents for specific roles, industries, or processes. This positions Salesforce against ServiceNow Studio and composable CRM vendors like Creatio by reducing time-to-deploy for workflow automation. ServiceNow announced ServiceNow Studio as a centralized environment for no-code and low-code developers to build agentic applications, blending workflow automation with customer operations.

Both moves shift spend from custom integration work to platform-native configuration, but they also increase vendor lock-in risk. Agent logic tightly coupled to a single CRM stack makes switching costs higher. For enterprises standardizing on workflow platforms, ServiceNow's approach can reduce integration sprawl, but it also shifts CRM buying decisions toward broader enterprise workflow platforms rather than standalone CRM tools.

Governance becomes the critical buying criterion. Low-code agent creation expands who can deploy customer-facing automation, which requires stronger role-based controls, auditability, and guardrails for regulated processes. Buyers must evaluate whether their compliance teams can maintain oversight when business users build agents without central IT involvement.

Composable CRM Challenges Monolithic Suites

Creatio is leading the composable CRM movement, which promotes modular functionality buyers can license selectively rather than committing to full suites. This challenges monolithic offerings from Salesforce, Microsoft, and ServiceNow by lowering implementation risk and supporting phased procurement. Buyers can trial specific AI or automation components before committing to a full platform, which preserves optionality while AI use cases evolve.

Composable CRM supports more granular spend allocation, which appeals to CIOs managing budgets during uncertain AI adoption curves. The trade-off is integration complexity: selecting best-of-breed modules across multiple vendors increases the burden on internal teams to maintain data flows, security policies, and user experience consistency.

Veeva's industry cloud for life sciences extends the vertical CRM battle against Salesforce Industry Cloud offerings. Verticalized clouds can reduce customization cost and time-to-value in regulated sectors, but they narrow future flexibility if the vendor's industry model does not match internal processes. Buyers in life sciences may see lower validation and compliance effort versus horizontal CRM, but they must assess whether the vendor's industry model is open enough for adjacent systems and data flows.

What to Watch

CRM budgets are shifting from "innovation" line items to measurable automation payback. Procurement teams now require proof of lift in resolved cases, productivity, or conversion before approving AI modules, which tightens ROI scrutiny and extends pilot-to-production timelines. Vendors that cannot quantify impact face longer sales cycles.

The competitive question is no longer which CRM has the most features, but which platform can prove measurable automation outcomes with acceptable governance. Buyers should focus contract negotiations on three areas: precise definitions of what counts as a resolved interaction for outcome-based pricing, governance controls for low-code agent creation, and exit rights that limit switching costs if agent logic becomes tightly coupled to a single vendor stack. Vendors willing to share execution risk through outcome-based pricing will have an advantage, but only if measurement disputes do not consume the savings.