For years, enterprise software vendors have fought to keep users inside their applications, but the rise of AI agents and automated workflows is changing that equation. Salesforce has been moving quickly to adapt, with last month’s launch of its new on Wednesday, Salesforce top executives positioned Headless 360 as both a major AI-era architecture shift and a fresh monetization opportunity, as enterprises increasingly move towards using AI agents, Slack bots and external copilots to access Salesforce data through servers, rather than its traditional application interfaces, for use in automated workflows .
“We’re going to bring our number one agentic , Salesforce chief revenue officer, told analysts during the call, explaining the company’s strategy around Headless 360. “And we’re going to work together with our customers and with our partners to find the right ways in a fair way to monetize those new interactions and those new users that are accessing our platform.”
Concerns over unpredictable costs
That evolving pricing strategy, analysts say, reflects customer enterprises’ hesitation around adoption of Headless 360, mainly driven by worries about unpredictable consumption costs.
, managing partner at KramerERP, sees the rise of automated workflows and agent-driven CRM interactions gradually pushing Salesforce away from the traditional subscription licensing models that were more predictable, and toward more consumption-based pricing when it comes to API and MCP usage.
Thus, according to Hinchcliffe, that metering of API and MCP use for Headless 360 will result in CIOs losing the budget predictability they traditionally associated with CRM platforms, replacing relatively fixed SaaS spending with cloud-style elastic consumption economics, where spend is driven by usage volume.
And predictability, said , advisory fellow at Info-Tech Research Group, added that API- and token-driven CRM consumption models could introduce further pricing volatility, because enterprise costs may increasingly fluctuate based on factors such as model routing, context caching, prompt efficiencies, and constantly shifting AI model pricing structures.
Understand costs before adoption
In fact, he sees automated CRM workflows almost certainly increasing enterprise costs.
“Automation based on consumption metrics increases transaction volumes. As the underlying modules of Salesforce are called upon for an integrated experience, each with its own billable layer, costs will explode higher,” Bickley said.
As a result, the analyst advised CIOs to carefully evaluate the business use case and check whether those higher costs can be tied to measurable productivity or revenue gains before broadly adopting automated CRM workflows at scale.
CIOs, Bickley warned further, should be hesitant about committing until the commercial model is formalized, communicated, and well understood.
“Buyers should be asking questions like: What counts as a billable call? Are internal agent calls priced differently from customer/external facing calls? Are there caps or alerts? Can consumption spike unexpectedly?” Bickley said.
Similarly, , AI development manager at IT consultancy firm Kanerika, noted that other operational changes that CIOs need to make would revolve mostly around governance, including agent approval systems, audit trails, and data leakage prevention.
The broader SaaS pricing reckoning
For , while keeping more deterministic workflows outside metered billing, he said, while Microsoft is experimenting with per-agent pricing models in Copilot Studio, and Workday, though still largely reliant on seat-based licensing, is also facing growing pressure to adapt its pricing strategy for AI-driven usage patterns.
Salesforce, Bickley pointed out, is ahead of many SaaS vendors both in terms of execution and marketing, having already “articulated a coherent enterprise architecture” around agentic AI, APIs, MCP servers and cross-platform workflow integration.
Increased pricing complexity
The caveat, according to Chaturvedi, is the increased complexity around licensing.
“Salesforce customers already navigate one of the most complex licensing structures in enterprise software. Layering consumption-based metering on top of per-seat pricing on top of flex credits on top of ELAs creates a commercial model that requires a spreadsheet to decode and a lawyer to negotiate. Every layer of pricing complexity becomes a reason for a CFO to slow down adoption,” Chaturvedi explained.
Salesforce declined to comment on how API and MCP interactions under Headless 360 are currently being billed. Analysts, however, agreed that most API and MCP calls are currently being billed through a mix of existing API consumption, Agentforce usage constructs, platform entitlements, and negotiated enterprise agreements rather than a single clean “per MCP call” pricing model.
Despite this uncertainty, during the earnings call Milano pointed to customers such as Adecco and Anthropic as examples of rising API- and Headless-driven Salesforce usage. “Anthropic is one of our biggest users of CRM, of Sales Cloud, and obviously Slack. Their usage through Q1 has exploded fivefold because now they’re using Sales Cloud from a Headless perspective,” Milano said.
SOCIAL SHARE CARD GENERATOR