Full narration of the briefing. Click a section heading to jump the player to that moment.
Salesforce counts go up easily and come down rarely. Adding seats takes a phone call and lands mid term without ceremony. Removing them, in most contracts, is simply not a right you hold, because the order form commits you to quantities for the term and the renewal quote arrives pre-filled with what you bought rather than what you use. I am Tom, Daniel is with me, and this session is about the harder direction: how to actually shrink an estate.
Not the wishful version where you ask nicely, but what genuinely moves a count down, what to trade when a straight cut is refused, and how to handle the moments when shrinking is not optional.
Start with timing, because reductions have exactly one natural window. Mid term, your quantities are contractual and Salesforce has no commercial reason to release you. At renewal, the whole quantity is back in play, because the contract is being written again. That single fact should reorganise your calendar.
If you know a business unit is closing, a project failed, or headcount is falling, the work starts a year out, not when the quote arrives. And mind the notice window: in an auto renewing contract the reduction conversation must happen before the deadline, because after it you are committed to another term at the old count, and no amount of goodwill unwinds that.
A reduction succeeds on evidence, not on how firmly you ask. Two numbers do the work. First, licences assigned against licences actually used, from your own admin data, ninety day login and activity windows, per product and per business unit. Second, the money that waste represents at your net price, stated annually.
Eight hundred dormant Sales Cloud seats at a net rate near a hundred and thirty dollars a month is more than a million dollars a year, and a number like that changes who attends the meeting on your side. What does not work is the budget argument alone. Every account team hears budget pressure from every customer. Nobody argues with their own login data.
Now the practical part. A pure cut is the hardest ask, because it reduces the account, and the account is what the team is compensated on. The trade that lands is the one that keeps the account whole while the count comes down. Retire dormant Sales Cloud seats and move that value into the product they are pushing this year, Agentforce or Data Cloud, so the number holds while the waste stops.
Or convert full CRM licences down to Platform licences for the read only population, which is the same people at a fraction of the rate. Or shorten the term in exchange for the reduction. Every one of those gives them a story to take to their deal desk, and a deal desk needs a story more than it needs your sympathy.
Corporate events are the sharpest version of this problem. When you sell a business unit, its Salesforce users leave, but your commitment usually does not, so you can find yourself paying for an organisation you no longer own. Nothing transfers unless the paper says so. The rights that help are assignment and divestiture language: the right to split an org, to assign a portion of the entitlement to the buyer, or to reduce your commitment by the divested population at the next anniversary.
Those are first purchase terms, per session five, and if a transaction is on your horizon, raise them at the next negotiation you have rather than after the announcement, because after the announcement your leverage is public and gone.
Then the hardest case: the estate must shrink, materially, and the account team knows it. Do not open with a percentage demand, because a demand without a plan invites them to wait you out. Open with the smaller estate you have actually designed: which users keep full licences, which drop to Platform, which clouds you retire, and what the resulting configuration costs at fair rates. That is a proposal they can price against, and it is credible precisely because it is specific.
Then be honest about the alternative you are prepared to accept, which is usually a reduced Salesforce footprint rather than none at all. A partial exit you would genuinely execute carries more weight than a total exit nobody believes.
The move: build the gap ledger now, this quarter, not at renewal. Entitlements against active users, product by product, refreshed quarterly, with the annual money attached. It is the same file that powers your leverage inventory from session seven, and it is the only thing that makes a reduction credible when you finally ask. Next session is the one everybody wants: Agentforce, the three pricing models, and how to estimate what it will actually cost before you sign.
See you there.
Redress Compliance works on contingency: our fee is 25 percent of what we save you. Nothing saved, nothing paid. Independent, buyer side only, never vendor funded. Want Redress to contact you? Reach out and we respond the same day.
Talk to a Salesforce negotiator