The ladder is published and five percent have moved
Salesforce has replaced its Sales, Service and Industries editions with three tiers named Core, Advanced and Max. Core lists at $195 per user per month, Advanced at $395 and Max at $550, and every tier bundles Slack, Tableau Next, a Premier Success Plan, security and AI. On the latest earnings call, Miguel Milano, Salesforce President and Chief Operating Officer, is reported as saying that only five percent of knowledge workers using Sales Cloud and Service Cloud have upgraded to higher-end editions.
That five percent reaches us through reporting by Sasha Semjonova at Salesforce Ben on September 28. Salesforce Ben is a second-tier source, a community trade site rather than the vendor, and the figure arrives as reported speech rather than a verbatim quotation from the call. The price ladder and the credit entitlements are firmer, because our desk confirmed both against Salesforce's own editions announcement dated September 3.
A vendor restructuring its whole edition lineup around higher tiers, while its own chief operating officer reports that twenty out of every twenty-one users on the two largest clouds have stayed put, is describing a gap between what is being sold and what is being bought. For anyone sizing an estate ahead of Winter ’27, that gap is where the money sits.
The first consumption numbers anyone can model
Each tier carries a named Flex Credit allocation. Core includes 500,000 credits, Advanced includes 1 million and Max includes 2.75 million. Put those against $195, $395 and $550 and an architect has, for the first time this year, a concrete set of numbers on both sides of the consumption question rather than one side and a placeholder.
Say that plainly, because this desk has spent weeks reporting the opposite. We wrote about what the edition prices leave off the bill when the credit pool had no rate against it, and about how agent work gets counted before it gets priced while the unit itself was still moving. Named credit allocations attached to named seat prices are more information than we have had.
The allocation tells you the size of the pool and not what a unit of work draws from it. Salesforce has still not published a rate that converts an agent conversation, an automated case or an integration call into credits. A pool you can measure against consumption you cannot is half a model, which is further than most of this year's pricing coverage got.
Compare the credit pools before the seat prices
Do the arithmetic on the steps rather than the headline. Moving from Core to Advanced costs $200 more per user per month and doubles the allocation, from 500,000 credits to 1 million. Moving from Advanced to Max costs $155 more and raises the allocation to 2.75 million. The more expensive step buys the smaller increase in included capacity, and the cheaper step buys the larger one.
That makes included credits the number to compare against expected consumption, because a tier that exhausts its allocation in month eight can cost more across the year than the tier above it. Whichever way the overage is priced, the cheaper seat stops being the cheaper deal once the pool runs dry. Work out where your estate would reach that point under each allocation before anyone argues about $195 against $395.
Part of the consumption side still cannot be modelled at all. We reported on September 24 that a headless usage type arrived on the rate card with its multiplier blank, which leaves teams whose integrations and agents reach Service Cloud from outside the screens estimating against a figure Salesforce has not published. The ladder narrows the unknowns without closing them.
Five percent is no verdict on the product
Five percent uptake on a recently introduced ladder is weak evidence of failure. Edition structures take time to land, and renewals arrive on their own cycles. An organisation eighteen months into a three-year agreement will not reopen it because a new tier appeared in September. A good share of the other ninety-five percent sit nowhere near a decision point, and reading that as a rejection of the tiers would be reading it wrong.
The fair criticism is narrower. Renaming and repackaging faster than customers can evaluate imposes a real cost on the people who have to keep up. The Salesforce Ben report notes a headless offering repackaged under a new name inside roughly six months. Every rename sends an architect back through rate cards and internal approvals built on the previous names, and none of that work produces anything a sponsor can see.
Cecilia Chiderski, Salesforce Architect and Director of AI at CLD Partners, described the problem in the report. "There is a disconnect between the pace things are moving, and the pace organizations can adopt them." What she names sits on the customer's side of this relationship rather than the product's side. The evaluation work takes weeks of senior time, and it queues behind everything else the platform team already owes the business.
Take a consumption number into the renewal call
The useful preparation takes an afternoon. Pull twelve months of agent conversations, automated case handling and integration volume by workload, and write the assumption beside each figure so the sheet survives a challenge. Then set the total against 500,000, 1 million and 2.75 million, and see which allocation your current estate would already be sitting inside.
An estate consuming a third of the Core allocation has a different conversation ahead of it than one running at double. Ask the account team, in writing, what draws a Flex Credit and at what rate, workload by workload, and whether the unpriced headless multiplier will apply to the API volume your contract already covers. Our guide to negotiating a multi-year SaaS renewal covers the rest of the table.
Bring the five percent figure with you. A vendor whose own chief operating officer reports that almost everyone on Sales Cloud and Service Cloud remains on the tier below has a reason to make the step up worth taking, and asking what that is worth in credits or in price is a fair question before Winter ’27 planning starts.
