Microsoft Copilot Licensing 2026:
Five Products, Five Price Tags

Microsoft Copilot Licensing in 2026: The Complete Price Map

Microsoft's AI licensing portfolio has grown significantly since the initial M365 Copilot launch. In 2026, enterprise procurement teams must navigate five distinct Copilot products, each with separate licensing models. Microsoft 365 Copilot Enterprise costs $30 per user per month and requires a qualifying M365 base licence (E3, E5, Business Standard, or Business Premium). A promotional Business plan at $18 per user per month applies through June 2026, rising to $21 thereafter; enterprise negotiations should focus on the $30 enterprise tier, where discount headroom of 10–40% is available depending on deal size and strategic importance.

Copilot Studio — the platform for building custom AI agents and automations — operates on a credit-based model. Capacity packs cost $200 per month for 25,000 Copilot Credits, with up to 20% savings available through upfront Credit Commit Units. GitHub Copilot Business costs $19 per user per month; GitHub Copilot Enterprise costs $39 per user per month, adding knowledge base integration and organisation-wide code customisation. Copilot for Security is priced at $4 per Security Compute Unit (SCU) per hour, with a minimum practical deployment of one SCU equating to approximately $35,000 annually. Critically, E5 licences now include Copilot for Security at 400 SCUs per month per 1,000 paid licences — a significant cost offset that changes the E5 upgrade economics for security-focused organisations. Our full analysis of the Microsoft 365 E3 vs E5 comparison covers how this inclusion affects the broader tier decision.

Microsoft 365 Copilot Prerequisites: What You Must Have Before You Buy

Microsoft 365 Copilot does not work on F1 or F3 licences, nor on legacy Office 365 plans. It requires a qualifying base: E3, E5, Microsoft 365 Business Standard, Business Premium, Business Basic, Apps for Business, or Apps for Enterprise. Within that base, three technical prerequisites must be met before Copilot will function effectively: Microsoft Entra ID authentication, OneDrive enabled for file-based features, and new Outlook or Outlook for Microsoft 365 (Classic Outlook is not supported).

Beyond the base licence, data governance readiness is the most commonly underestimated prerequisite. Microsoft Copilot operates within your M365 data boundary and honours your existing sensitivity labels and access controls — but if those controls do not exist or are poorly configured, Copilot will surface data that users should not see. Microsoft Purview data classification, sensitivity labels published to SharePoint and OneDrive, and Data Security Posture Management for AI are all required to deploy Copilot without creating compliance exposure. Organisations without mature Purview deployments face a 6–12 month preparation runway before Copilot can be deployed at scale. This preparation cost is rarely factored into initial Copilot business cases. Use our Microsoft Copilot ROI assessment to baseline your readiness before committing to any licence volume.

Microsoft Copilot ROI Reality: What the Numbers Actually Show

Microsoft and Forrester cite impressive ROI figures — $19.7 million NPV and 116% ROI for enterprise deployments. The independent evidence tells a more qualified story. Only 5% of organisations that started a Copilot pilot have moved to a larger-scale deployment (Gartner, 2025). The conversion rate from Copilot licence access to active use sits at just 35.8%. Among users who have stopped using Copilot, 44% cite distrust of AI-generated answers as the primary reason. When users are given a choice between Copilot, ChatGPT, and Gemini, only 18% choose Copilot — a figure that falls to 8% when all three options are available simultaneously.

The features that do drive consistent value are meeting summarisation in Teams, email drafting in Outlook, document creation in Word, and data analysis in Excel. These are the use cases where Copilot has the most structured data to work with and where users have the clearest quality bar for output. The use cases that routinely underdeliver are complex document synthesis across multiple SharePoint sources, code generation in mixed-language environments, and customer-facing communications — all areas where hallucination risk is highest and quality review overhead erodes the time savings. Download our Copilot procurement strategy guide for the full framework on selecting pilot use cases that maximise your probability of a successful business case.

Microsoft Copilot Licensing Negotiation Tactics in 2026

Microsoft removed volume licensing discounts for Online Services from November 2025, which in theory means all enterprises pay the same list price. In practice, enterprise negotiations still generate meaningful discounts through deal structure, commitment mechanics, and strategic leverage. The most effective approach is a structured pilot-first commitment: negotiate a 90-day pilot covering 100 users in two defined workflows, with explicitly stated KPIs (meeting summary adoption rate, report creation time reduction, weekly invocation counts). Tie any expansion commitment to KPI achievement, and lock in the volume pricing terms for Phase 2 expansion during the Phase 1 negotiation — before Microsoft knows whether the pilot will succeed.

Microsoft's Enterprise Commercial Investment Fund (ECIF) provides credits for large organisations that Microsoft classifies as strategically important. If your overall Microsoft spend exceeds $1 million annually, request ECIF consideration as part of your Copilot negotiation. For the EA mechanics of structuring a mixed Copilot commitment — including how NCE annual commitments interact with an existing EA — see our guides on Microsoft EA vs MCA-E and Microsoft NCE price lock strategy. The July 2026 price increase on the underlying E3/E5 base creates an additional negotiation lever: your willingness to expand Copilot deployment is contingent on acceptable pricing for the base licences that underpin it.

For organisations evaluating GitHub Copilot separately from M365 Copilot, treat it as a distinct negotiation. GitHub Copilot Enterprise at $39 per user per month is typically justified only for organisations with 50+ active software developers and a clear policy for reviewing and accepting AI-suggested code. For smaller developer populations, the Business tier at $19 per user per month and a structured code review policy delivers comparable value at lower cost. To understand how all Microsoft AI licensing commitments should be structured within your broader EA renewal, book a confidential call with our Microsoft advisory team — we have supported over 500 enterprise Microsoft negotiations and can benchmark your current Copilot pricing against market norms.

Copilot vs Alternatives: The Vendor-Neutral Question Microsoft Won't Answer

Microsoft's commercial narrative positions M365 Copilot as the default choice for enterprises already on the Microsoft stack — the logic being that native integration with Teams, Outlook, and SharePoint justifies the premium over standalone AI tools. The adoption data complicates this. When enterprise users are given free choice between Copilot, ChatGPT, and Gemini, only 18% choose Copilot as their primary tool — a figure that falls to 8% when all three are simultaneously available. ChatGPT held 55.2% of paid AI subscriber market share in January 2026, compared to Copilot at 11.5%.

This does not mean Copilot is the wrong choice — it means the choice should be made on evidence, not on vendor convenience. The strongest Copilot use cases are those that require deep M365 data integration: meeting summaries that pull from Teams transcripts, email drafts that reference calendar context, and document creation that cites SharePoint documents with appropriate sensitivity controls. For general knowledge work, research, and code generation, the competitive landscape is material. Our Copilot vs Gemini vs Amazon Q comparison framework gives enterprise procurement teams the vendor-neutral analysis to make this decision based on your specific use case portfolio, not Microsoft's sales pitch. The right AI strategy for most enterprises involves multiple providers — the procurement discipline is ensuring you have not contractually locked yourself into a single platform before the market has settled.