Salesforce shifts to outcome-based AI pricing

Your software bill is about to stop being predictable. Salesforce, one of the world's largest business software vendors, is publicly admitting its pricing model is undergoing an 'anxiety-filled' evolution. That phrase did not come from a critic. It came from Salesforce itself. For Finance Managers, Operations Managers, and Managing Directors at UK SMEs, this is not a story about one CRM vendor. It is a signal that the comfortable era of counting user seats and signing annual contracts is ending.

What Salesforce is actually changing

For years, business software pricing has followed a simple formula. You pay a fixed amount per user, per month, and costs are easy to forecast. Salesforce is now moving away from that model for its AI tools. The company has introduced 'Flex Credits', a consumption-based pricing approach where you pay for what the AI actually does rather than how many licences you hold. It is also testing outcome-based pricing, which means charging per resolved customer issue rather than per seat.

The reason for the shift is the rise of autonomous AI agents. When software performs tasks independently rather than serving as a tool a human uses, the old per-user logic breaks down. Salesforce's own executives have described this pricing evolution as 'anxiety-filled', a candid admission that even the vendor has not yet settled on what fair value looks like. Salesforce discussed this direction publicly in September 2026, and the final structure remains in flux.

Before this shift, a Salesforce contract was largely a Finance Manager's domain. You knew the headcount, you multiplied by the per-seat cost, and you set the budget line. Consumption and outcome-based pricing changes that calculation entirely. Costs now depend on activity levels, task volumes, and how well the AI actually performs.

Why this matters beyond Salesforce

Analysts are clear that Salesforce is a bellwether, not an outlier. When a company of this size and market influence changes its commercial model, other major SaaS vendors tend to follow. Microsoft, HubSpot, ServiceNow, and similar platforms are all investing heavily in autonomous AI agents. Each of them faces the same structural question Salesforce is wrestling with: how do you price software that works on its own?

For UK SMEs, the implications reach further than your CRM contract. If outcome-based and consumption pricing becomes the SaaS industry norm, your entire approach to software procurement changes. The question shifts from 'how many people need access?' to 'how many tasks do we expect AI to complete, and at what success rate?'. Most finance teams are not yet equipped to answer the second question. Most operations teams have not yet defined what a successful AI outcome even looks like within their workflows.

That gap between the two questions is where the real risk sits. Vendors are moving faster than buyers. Contracts will become more complex before they become clearer.

The budget and process risks UK SMEs need to price in

Outcome-based pricing sounds attractive in principle. Paying only for successful results seems fairer than paying for a seat whether or not the software delivers value. But the practical reality for a Finance Manager is that it introduces a new category of financial unpredictability.

Consider a scenario familiar to many UK SMEs. You currently pay a fixed monthly fee for your CRM. You know that cost will not change unless you hire more staff or add modules. Under a consumption model, a busy trading period, a new marketing campaign, or an increase in customer queries could all drive your AI usage costs up without any deliberate decision on your part. You would need monitoring systems and internal alerts before the invoice arrives.

Operations Managers face a parallel challenge. For outcome-based pricing to work, someone needs to define what counts as a resolved issue, a completed task, or a successful agent interaction. That definition has to be precise enough to appear in a contract. It also has to be tracked continuously. Without that discipline, disputes with vendors become inevitable and budget overruns become hard to explain to the board.

Membership bodies are a case in point. Many run member records, renewals and events on CRM platforms such as Salesforce, so a change in how AI features are priced lands directly on their budgets.

The deeper issue is that most SMEs have not yet aligned their AI adoption to business outcomes at all. They have adopted tools because tools became available, not because they mapped each tool to a measurable result. Outcome-based pricing forces that conversation into the open, and not always at a time of the buyer's choosing.

Three ways to prepare before the contract lands on your desk

1. Audit your current AI usage before your next renewal. Ask your Operations Manager to document which AI features your team actually uses, how frequently, and what business task each one supports. This takes one focused meeting and a shared spreadsheet. It creates the baseline you need to evaluate any consumption-based proposal against real activity data rather than vendor estimates.

2. Define what 'success' means for each AI-assisted process. Before your next vendor conversation, agree internally on what a successful outcome looks like for each automated task. A resolved customer query has a definition. A completed data entry task has a definition. Write them down. If you cannot define success yourself, a vendor will define it for you, and their definition will favour their margin.

3. Ask vendors for a 12-month cost model under different usage scenarios. When a vendor presents consumption or outcome-based pricing, request a worked example at three usage levels: your current volume, a 25% increase, and a 50% increase. If the vendor cannot provide that model, or if the numbers are significantly higher than your current spend, you have a negotiating position. Do not sign until you understand the ceiling.

What this pricing shift will not solve

It is worth being direct about one thing. Outcome-based pricing does not automatically align vendor incentives with your business goals. A vendor charging per resolved query has an incentive to define 'resolved' broadly. A vendor charging per completed task has an incentive to count tasks generously. Contracts written in the vendor's favour will still need legal scrutiny.

For many UK SMEs, the honest limitation here is internal capacity. Monitoring AI consumption, defining outcome metrics, and tracking performance against contracted terms requires someone to own that process. In a business with a lean finance function and a Managing Director who is already stretched, adding a new category of vendor management is not trivial. This is one reason the shift to outcome-based pricing will not be painless even when it is, in principle, fairer.

The technology is moving. The pricing models are catching up. The internal processes at most SMEs have not started the journey yet.

Getting ahead of a contract you have not seen yet

If your organisation uses any major SaaS platform with an AI component, the likelihood is that your next renewal conversation will introduce some form of consumption or outcome-based pricing. That conversation will go better if your Finance Manager and Operations Manager have already aligned on how your team uses AI, what results you expect, and what monitoring is in place.

gecco's Training and consultancy programme is designed for exactly this moment. If you want to build vendor-neutral AI capability in your team before the contract arrives, our Essentials and Complete training tiers are built for that.

The AI Readiness survey is a practical starting point for any Managing Director or finance lead who wants to understand where their organisation sits before the next renewal cycle. Taking the survey gives you access to 65+ free resources and a custom AI Readiness report, followed by a free 30-minute showcase to walk through the results with a member of the gecco team.

Website · LinkedIn · Case Studies · Newsletter