Full narration of the briefing. Click a section heading to jump the player to that moment.
For the last episode, one story instead of one topic. A composite of real engagements, running the whole method from the first export to signature. Names changed, numbers rounded, mechanics exactly as they happen. Because a method that only works on a slide is not a method.
Meet the estate: an expiring commitment and a migration. A manufacturer spending about eighteen million dollars a year on AWS, three years into a commitment expiring in September, with a data centre exit in flight that AWS had offered to fund, and an accelerator programme the AI team wanted committed capacity for. The account team's proposal arrived in April: a new five year commitment starting twenty three million in year one, rising steeply, with a headline discount two points better than the expiring deal and several million in migration credits attached.
The baseline moved the number before anyone met. The cost and usage report showed run rate at seventeen point four million, not the twenty one the proposal assumed. Commitment coverage sat at thirty one percent, so a large share of steady state was paying list. And the accelerator forecast came from a vendor slide rather than a metered pilot.
Three findings, and none of them required a negotiator. They required an export and the willingness to read it. The team spent six weeks right sizing and lifting coverage before saying a word to AWS, which lowered the honest floor to about sixteen million, and they metered the AI workload for a month instead of forecasting it.
The campaign ran on their machine, not ours. Starting nine months out, the alternative went to a competing hyperscaler as a genuine assessment in month three, with a named sponsor and two pilot workloads, not as a threat. That did two things. It gave the team a real price to compare against, and it made the deal competitive, which is one of the three triggers that moves an AWS file to the team with actual pricing authority.
The term sheet went in four months out, and the endgame was aimed at the December approval window with the paperwork submitted before the monthly cutoff.
At the table, four asks carried the deal. First, the measurement basis: commitment retiring on gross consumption rather than net of discount, which on this size of deal was worth more than the extra discount point they had been arguing over. Second, the ramp: year one set below the proven floor with growth pushed into years two and three, plus a shortfall cure period and a step down right tied to divestiture. Third, the credits taken as genuinely incremental migration funding, modelled in the same spreadsheet as the commitment so the shortfall interaction was visible rather than discovered later.
And fourth, the quiet one: support recalculated, the marketplace cap pushed above the standard quarter, and the third party software the business was buying anyway routed through marketplace, which retired commitment and stayed outside the support fee base at the same time. The signed commitment came in at seventeen million in year one against a proposal of twenty three, with the effective rate several points better than the headline once the marketplace and support treatment were counted, and the contract gained gross measurement, shortfall relief, a step down right and the thirty day services list protection.
What actually made the difference. Not eloquence in the room. The team won on three things that all happened before the first meeting: they measured their own estate, they optimised before committing rather than after, and they made the deal competitive early enough that it reached someone who could price it. And the discipline that made those possible was ordinary.
An export, a spreadsheet, a calendar, and the willingness to say that a number is a hypothesis until it is metered. That is the series. The instrument in episodes one to three, preparation in four to six, the negotiation in seven to nine, and the long game in ten to twelve. Watch them in order, or take the one you need this quarter.
One last point. At Redress Compliance this is the day job: we prepare and run AWS negotiations for large enterprises on pure contingency. Our fee is twenty five percent of what we save you. If we save you nothing, you pay nothing.
Thank you for watching. Your commitment is closer than it looks. Start the export this week.
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 AWS negotiator