MACC is a three year forward commitment. The fastest way to lose budget is to wait for the anniversary. The cleanest way to win it back is to renegotiate mid term, at the right trigger.
Microsoft Azure Consumption Commitment, or MACC, is the three year forward dollar commitment that sits inside most enterprise Azure agreements. The commitment is what wins the discount on the Azure price book. It is also what creates the budget risk if consumption misses the schedule.
Most enterprise buyers treat MACC as a fixed obligation and wait for the three year anniversary to renegotiate. That is the wrong moment. The right moment is at month 14 to 18, when 12 months of real consumption data is on the table and Microsoft's incentive to reshape is still high.
Azure AI workloads in 2026 burn MACC roughly four times faster than typical IaaS workloads on a dollar per workload basis. A MACC sized in 2023 without AI in scope is almost always under sized for 2026 consumption, which Microsoft will offer to reshape upward in exchange for a wider discount.
MACC is a forward dollar commitment. It is not a credit. The commitment dollars are consumed only when the customer consumes Azure meters that are eligible for MACC.
Several common spend lines do not draw from MACC and must be budgeted separately.
Microsoft will entertain a mid term reshape conversation when one of four triggers fires. The trigger is the customer's lever, not the vendor's.
| Trigger | What it looks like | Buyer side leverage |
|---|---|---|
| Burn under by 15 percent | Trailing 12 month burn rate is more than 15 percent below the original schedule | Downward reshape or scope swap |
| Burn over by 15 percent | Trailing 12 month burn rate is more than 15 percent above the original schedule | Upward reshape with deeper discount |
| AI scope expansion | Adding Azure OpenAI or AI Foundry workloads not in original scope | Reshape upward with AI specific discount |
| M&A scope change | Acquired entity adds new Azure tenancies inside the MACC | Reshape upward and consolidate enrollments |
The conversation goes nowhere without consumption data. The buyer side brings three reports and a single ask.
Azure OpenAI Service meters on input tokens, output tokens, and provisioned throughput units. Azure AI Foundry meters on model serving capacity and storage. GPU compute meters on dedicated instance hours.
A typical IaaS workload burns MACC at a roughly predictable monthly run rate tied to user activity. AI workloads burn unevenly. Training runs spike. Inference grows with adoption. Provisioned throughput units carry minimum commitments of their own.
Sizing AI workload MACC on year one inference volume guarantees an under size by year three. The fix is to size on a 36 month curve and to negotiate a mid term reshape right that allows scope adjustment at month 18 without penalty.
Customers carrying on premise SQL Server, Windows Server, or RHEL with Software Assurance can apply Azure Hybrid Benefit to reduce the per hour rate on Azure compute that runs those workloads. The reduction lowers MACC burn for the same workload.
Server and Cloud Enrollment renewal is the cleanest moment to reshape MACC. The SCE conversation already includes the Azure commitment. The buyer side ask is a single LOI that combines the SCE renewal, the MACC reshape, and the AI workload scope.
The levers below are the ones we use on a typical mid term MACC renegotiation. The order is roughly the order they tend to land at the negotiating table.
An enterprise customer signed a 12M USD three year MACC in 2023 with no AI in scope. At month 18 the trailing 12 month burn is 5.8M USD on track, but the AI pipeline is projected to add 3M USD of OpenAI consumption per year over years two and three of the next term.
| Lever | Outcome | Three year impact |
|---|---|---|
| Upward reshape to 21M USD MACC over three years | Discount widens from 12 percent to 18 percent | Saves 1.26M USD against list |
| AI scope addition with Azure OpenAI specific discount | Token rate discount adds 4 percent | Saves 360K USD on AI consumption |
| Hybrid Benefit applied to 800 SQL Server cores | Reduces SQL compute rate 55 percent | Saves 420K USD |
| Mid term reshape right at month 24 | Right negotiated into LOI | Avoids 600K USD downside risk |
| Total | 2.04M USD plus 600K USD avoided |
The seven step sequence below is the buyer side workflow on a mid term MACC renegotiation.
MACC is a forward dollar commitment across eligible Azure meters in aggregate. A reserved instance is a forward unit commitment to a specific VM SKU at a specific region for one or three years. Both reduce burn rate. They compose: a reserved instance consumes MACC dollars at the discounted reserved rate.
Yes. Microsoft will entertain a mid term reshape conversation when one of four triggers fires: burn under by 15 percent, burn over by 15 percent, AI scope expansion, or M&A scope change. The mid term reshape often nets a deeper discount than the original commitment, especially when scope is expanding.
Unused MACC dollars expire at the end of the three year term. They do not roll forward into the next term and they are not refundable. Burn under by more than 15 percent should trigger a mid term scope swap conversation rather than waiting for the anniversary.
Azure OpenAI Service meters on input tokens, output tokens, and provisioned throughput units. Each meter is MACC eligible and burns MACC dollars at the standard meter rate. Provisioned throughput units carry monthly minimums regardless of traffic, which can accelerate MACC burn unevenly.
Yes. Applying Azure Hybrid Benefit lowers the per hour rate on the eligible Azure compute, which reduces MACC dollars consumed for the same workload. Hybrid Benefit can be applied retroactively in some cases through a reshape conversation tied to an SCE renewal.
MACC dollars cannot be freely transferred between enrollments. A consolidation of multiple Azure enrollments under a single MACC is possible at renewal or at mid term reshape, particularly after an acquisition that adds new Azure tenancies.
Discount bands move with commitment size and term. A 5M USD three year MACC typically wins 8 to 12 percent. A 20M USD three year MACC typically wins 15 to 22 percent. AI scope added at mid term reshape often unlocks an additional 3 to 6 percent on the AI portion.
If the trailing 12 month burn rate or the forward 24 month projection diverges from the original schedule by more than 15 percent, renegotiating mid term is almost always the better choice. Waiting for renewal forfeits the leverage that real consumption data creates.
Buyer side reference on the Microsoft Enterprise Agreement renewal including MACC, M365, and Copilot commercials. Reshape triggers, AI metering, Hybrid Benefit gaps, and the levers procurement carries to the table.
Independent. Buyer side. Written for CIOs, CFOs, and procurement leaders carrying Microsoft Enterprise Agreements, MACC commitments, and Azure consumption. No Microsoft referral fee. No conflict on the table.
Open the white paper in your browser. Corporate email only.
Open the Paper →The leverage on a MACC is not at the three year anniversary. It is at month 18, when 12 months of real burn data sits on the table and Microsoft has an incentive to widen the discount in exchange for scope and commitment.
We have run 500+ enterprise engagements across 11 publishers. Every engagement starts with one conversation.
Monthly Microsoft intelligence on MACC reshape triggers, Azure AI burn patterns, Hybrid Benefit gap analyses, and renewal levers from every Microsoft engagement we run on the buyer side.
Once a month. Audit patterns, renewal benchmarks, vendor commercial signals across Oracle, Microsoft, SAP, Salesforce, IBM, Broadcom, AWS, Google Cloud, ServiceNow, Workday, Cisco, and the GenAI vendors. No follow up sales pressure.
Free providers (Gmail, Yahoo, Outlook) cannot subscribe. Work email only. Unsubscribe in one click.
