As Salesforce accelerates its pivot toward AI, the organization is revisiting its approach to AI pricing, igniting a vital conversation in enterprise IT circles. The company previously experimented with usage-based and per-conversation charging models for AI solutions but now gravitates back to a seat-based licensing format. This evolution highlights a broader industry demand: financial predictability over experimental cost models.
The quest for predictability in AI pricing emerges amidst ambitious forecasts. By 2035, Gartner projects agentic AI will account for 30% of enterprise application software revenue, soaring to $450 billion. This enormous opportunity is the driving force for major software vendors, including SAP, Oracle, Workday, and Salesforce itself. Salesforce CEO Marc Benioff believes businesses can extract manifold value from AI products, multiplying monetization prospects.
However, the industry’s move toward AI-powered productivity raises intrinsic contradictions. Suppose AI indeed delivers the efficiency it promises. In that case, enterprises should logically require fewer resources, posing a challenge to per-user economic models long relied upon in enterprise software.
Salesforce initially favored a consumption-based pricing model with Agentforce, aiming to harmonize cost with perceived value ideally. In reality, unpredictable AI usage patterns across different organizational teams often created disproportionate costs unexpectedly for tech buyers. Consequently, seat-based pricing is being reintroduced. While not flawless, it offers enterprises much-needed certainty over fluctuating financials.
units, credits, and “fair use” metrics are now being built into today’s seat-based AI licenses. These concepts present a transitional bridge, allowing IT leaders wiggle room in exploring AI while maintaining budgetary control. This hybrid model grants vendors protection against unexpected spikes in computing expenses. Nevertheless, the balance of control tends to largely favor vendors.
Notably, the anticipated large-scale workforce reduction from AI is not materializing as expected. Reports indicate that laying off workers due to AI has left 55% of companies with regrets. Furthermore, 57% of leaders expect AI investments to lead to higher headcounts, whereas only 15% foresee reductions.
Enterprises navigate a delicate line between paying for capabilities versus tangible outcomes. AI pricing discussions, still in nascent stages, are expected to become more diverse. Copilots might retain a seat-based arrangement, but workflow automation agents may lean toward usage or outcome-driven pricing. These models require transparency, trust, and rigorous metrics-attributes historically neglected in software contracts.
AI agents promise transformative efficiencies, intelligence, and scale. But for businesses prioritizing financial stability, the most compelling feature may well be understanding their expenditure. This realization drives Salesforce’s recalibrated pricing approach, acknowledging that enterprises won’t fully embrace AI without a financial framework they can confidently navigate and understand.
