> For the complete documentation index, see [llms.txt](https://docs.archera.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.archera.ai/help-center/azure-faq/why-savings-plans-over-reservations-for-azure-compute.md). # Why Does Archera Recommend Azure Savings Plans by Default? Microsoft is [changing its reservation exchange policy on February 1, 2027](https://learn.microsoft.com/en-us/azure/cost-management-billing/reservations/reservation-exchange-policy-changes). Two questions follow from that, and this article answers both: whether the change affects Archera's money-back guarantee, and why Archera's default Azure plans lead with Savings Plans rather than Reservations. ## Archera's Money-Back Guarantee Has Never Depended on Exchanges Archera's money-back guarantee does not depend on, or rely on, the Microsoft Azure Exchange and Return system in any way. On Azure, the guarantee is honored through Archera's own rebate mechanism — not by exchanging, returning, or otherwise modifying your native Azure commitments. The mechanics are straightforward. Archera purchases a native commitment in your account and overlays an insurance-backed guarantee with a shorter Archera term — 30 days or 1 year — than the underlying native commitment. If that commitment goes underutilized once the Archera term is up, Archera automatically sends you a cash rebate for the net negative savings. That rebate is a payment from Archera to you, calculated from observed underutilization. It calls no Microsoft API, consumes none of your Azure exchange or cancellation eligibility, and leaves your native commitment untouched. The clearest proof is a product Archera already sells. **Azure Savings Plans cannot be exchanged at all — ever, by anyone.** That's a long-standing Microsoft platform limitation with nothing to do with the 2027 change, and Archera offers Guaranteed Savings Plans (GSPs) on Azure with the same money-back guarantee as Guaranteed Reserved Instances. A guarantee that already works on an instrument with zero exchange capability is, by definition, not built on exchanges. So February 1, 2027 changes nothing about your protection. There is no shared plumbing, no shared eligibility rule, and no shared cutoff date. Whether a commitment has zero, one, or unlimited native exchanges remaining has no bearing on whether it's protected. Archera was first to bring this model to AWS and remains the only provider offering it on Azure. One distinction is worth checking when you compare options: Archera's rebate is real money, returned to you, and the obligation is insurance-backed and carried in Archera's Terms of Service and MSA. Guarantees offered elsewhere are commonly settled as credits against the vendor's own fees, which reduces what you owe that vendor rather than returning your committed spend. ## Savings Plans Have Been Archera's Default Since 2024 Archera's default Azure plans have led with Savings Plans since 2024 — about two years before Microsoft announced this policy change. The reasoning hasn't changed, and it never had anything to do with exchange rights. A Reserved Instance is locked to a specific VM series, size group, and region. A Savings Plan commits to a dollar amount of hourly compute spend, so it applies automatically across VM families, regions, and other eligible services — App Service (Premium v3 / Isolated v2), Dedicated Host, Container Instances, and Azure Functions Premium — as your usage moves. One instrument tracks your infrastructure as it changes; the other assumes it won't. That difference had already stopped being theoretical. Microsoft has been progressively retiring older VM series, and as each retirement lands, **new Reserved Instances for those series can no longer be purchased or renewed at all** — including several widely used general purpose and compute optimized families whose purchase paths closed on July 1, 2026. The VMs keep running for years; the commitment instrument simply disappears. For customers on those series, a Savings Plan isn't the flexible alternative to a reservation — it's the only commitment available, and it keeps covering the workload if they eventually migrate. A SKU-locked instrument inherits every risk of the SKU it's locked to, including Microsoft's decision to stop selling commitments against it. That was true before this policy change, and it stays true after. ## What Changes on February 1, 2027 [Microsoft is restricting native exchange eligibility](https://learn.microsoft.com/en-us/azure/cost-management-billing/reservations/reservation-exchange-policy-changes) for Azure Reservations on any service also covered by a Savings Plan — including Virtual Machines, App Service, and SQL Database. * **Reservations purchased before February 1, 2027** keep one final exchange. Because an exchange is processed as a cancellation, refund, and new purchase, exchanging after that date produces a reservation that is no longer exchangeable. The right applies per quantity, not per reservation. * **Reservations purchased on or after February 1, 2027** can't be exchanged on these services at all. Several things are explicitly not changing. You can still **trade in** a reservation for a Savings Plan at any time, with no deadline. The cancellation policy is untouched, still capped at $50,000 per rolling 12 months per billing profile. [Instance size flexibility](/help-center/azure-faq/instance-size-flexibility.md) for VMs is unaffected. And reservations for services that Savings Plans don't cover — Azure VMware Solution, for example — are exempt entirely. For the full eligibility and cutoff mechanics, see [How Do Azure Reservation Exchanges Work, and What's Changing in February 2027?](/help-center/azure-faq/exchange-eligibility-and-2027-policy-change.md) The change removes the mechanism that made a mis-scoped reservation recoverable. Both instruments are now effectively fixed for their term, so the value of a reservation's deeper discount rests entirely on getting the forecast right the first time. Microsoft's own guidance now points the same direction: savings plans for "evolving or dynamic workloads," reservations for "predictable, stable workloads." ## Can You Still Choose Reservations? Yes. Reservations aren't going away and Archera fully supports them — but on Azure compute they're a deliberate choice rather than something the default plans propose. Archera's automatically generated plans (Recommended, Balanced, and High Savings) lead with Savings Plans. To commit to VM reservations instead, create a Custom Plan and narrow the contract types, deselecting Savings Plans and selecting the VM reservations you want. The reason to do that is the discount ceiling. Reservations still reach roughly 72% at three years for some series, against roughly 65% for a Savings Plan. That premium is worth capturing when a VM family, size group, and region are genuinely stable and predictable, and the series isn't approaching retirement. It also isn't strictly either/or — Azure applies reservation discounts first, then Savings Plan discounts to whatever eligible spend remains, so the two can be held together. See [How to Create a Custom Purchase or Renewal Plan](/help-center/user-guide/how-to-create-custom-plan.md) for the full walkthrough, and [Reservation vs. Compute Savings Plan](/help-center/cloud-service-intelligence/cloud-service-intelligence/azure/compute/virtual-machines.md#reservation-vs-compute-savings-plan) for the discount and eligibility breakdown. ## Related Resources * [Does Archera Rely on Azure Exchanges or Returns Policy?](/help-center/azure-faq/exchanges-returns-policy.md) * [How Do Azure Reservation Exchanges Work, and What's Changing in February 2027?](/help-center/azure-faq/exchange-eligibility-and-2027-policy-change.md) * [What does Azure RI 'Instance Size Flexibility' Mean?](/help-center/azure-faq/instance-size-flexibility.md) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.archera.ai/help-center/azure-faq/why-savings-plans-over-reservations-for-azure-compute.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.