Microsoft 365 E7: What the Frontier Suite Actually Means for Your Organisation
The Challenge
Every enterprise IT leader I speak to is having the same conversation with their board: "We're spending on Copilot. Where's the ROI?" And right behind that question sits a harder one: "We have agents everywhere now. Who's governing them?"
Microsoft has been shipping AI capabilities into Microsoft 365 at a relentless pace. Copilot landed. Agents followed. But the buying experience has been fragmented — E5 here, Copilot licence there, agent governance as yet another add-on. For organisations trying to plan budgets and architecture, the à la carte model created friction, not clarity.
Meanwhile, the agent problem is real. IDC projects 1.3 billion AI agents in circulation by 2028. Microsoft's own internal deployment surfaced over 500,000 agents across the company. Without a control plane, that's not adoption — it's sprawl.
What's Changed
On March 9th, Microsoft announced what amounts to a strategic reset for how enterprises consume AI. Four things happened at once.
Wave 3 of Microsoft 365 Copilot brings next-generation agentic experiences directly into Word, Excel, PowerPoint and Outlook. This isn't just better autocomplete. Users can now create and modify artefacts within Copilot chat and build their own agents within the canvas they work in every day. The underlying system — what Microsoft calls "Work IQ" — draws on organisational context: how you work, with whom, and on what content. This is the differentiation layer that separates Copilot from a generic ChatGPT wrapper.
Model diversity arrives in production. Claude from Anthropic is now available in mainline Copilot chat via the Frontier programme, alongside the latest OpenAI models. This is significant. Microsoft isn't betting on a single model provider — they're building a routing layer that picks the right model for the task. For customers, this means better results without having to think about which model sits underneath.
Agent 365 goes GA on May 1st at $15 per user per month. This is the enterprise control plane for AI agents. One place to observe, govern, manage and secure agents across your organisation — using the same identity, compliance and security infrastructure you already run for people. In preview, tens of millions of agents appeared in the Agent 365 Registry. The signal is clear: agents are proliferating whether you govern them or not. Agent 365 gives you the dashboard.
Microsoft 365 E7 — the Frontier Suite — launches May 1st at $99 per user per month. It bundles E5, Copilot, and Agent 365 into a single SKU, with Microsoft Entra Suite plus advanced Defender, Intune and Purview security capabilities included. Priced below purchasing these components separately, it's designed to make the adoption decision simpler and the budget conversation shorter.
Getting Started
If you're already on E5 with Copilot, the path to E7 is straightforward — it consolidates what you're already paying for and adds Agent 365 and enhanced security at a net saving. Talk to your Microsoft account team or licensing partner about upgrade paths before May 1st.
If you're on E5 without Copilot, E7 is essentially Microsoft's answer to "should we add Copilot?" The bundled price removes the incremental decision — you get Copilot, agent governance, and security hardening in one move.
For Agent 365 specifically, start here: 1. Audit your current agent space. How many agents exist across your tenant? Who built them? What data do they access? 2. Review the Agent 365 documentation for the registry, governance policies, and security controls available at GA. 3. Engage your security team now. Agent governance touches identity (Entra), data protection (Purview), endpoint management (Intune), and threat detection (Defender). This isn't an IT-only project.
Timeline consideration: GA is May 1st — roughly seven weeks away. That's tight for procurement, security review, and change management. If E7 is on your radar, start the internal conversation this week.
What This Means
This announcement is less about new features and more about how Microsoft sees the enterprise AI market maturing. The experimentation phase is over. The question isn't "should we use AI?" — it's "how do we govern AI at scale while actually getting value from it?"
The E7 bundle is a strong commercial signal. By pricing the full stack below the sum of its parts, Microsoft is incentivising organisations to go all-in rather than cherry-pick. That's good for adoption rates, but it also means organisations need to be ready — culturally, technically, and operationally — for AI that's embedded everywhere, not siloed in a pilot.
The model diversity play deserves attention too. Having Claude available alongside OpenAI models in production Copilot isn't just a feature — it's architecture. It means Microsoft is building the orchestration layer, not just the model layer. For customers, that reduces lock-in risk. For competitors, it raises the bar significantly.
There are legitimate questions. $99 per user per month is serious money at scale — a 10,000-seat deployment is nearly $12 million annually. Organisations that haven't yet demonstrated Copilot ROI at current pricing will need strong evidence before committing. And the "tens of millions of agents in the registry" headline needs scrutiny: how many of those agents are actively governed versus simply registered?
But the direction is unmistakable. AI copilots and agents are becoming core infrastructure, not optional add-ons. And the organisations that treat governance as a first-class concern — not an afterthought — will be the ones that scale successfully.
Leon Godwin, Principal Cloud Evangelist at Cloud Direct