
Microsoft's Dynamics 365 2026 release wave 1, rolling out from April through September this year, is not a typical feature update. It marks a fundamental shift in how ERP software works: from tools that help people do tasks faster, to agents that do certain tasks autonomously — stepping in, taking action, and only escalating when human judgement is genuinely needed.
For UK mid-market businesses running Dynamics 365 Finance or Supply Chain Management, this matters. The new capabilities aren't theoretical or limited to enterprise-scale deployments. They're arriving inside the software you already use, and many are included in your existing licence. The question is whether your business is ready to take advantage of them.
If you've been following Microsoft's AI journey, you'll know Copilot has been part of Dynamics 365 for a while now. Until this release, Copilot was essentially an intelligent assistant — it could suggest, summarise, and draft, but a human still had to initiate every action and approve every step.
Wave 1 changes that model. Microsoft is introducing what it calls an "agentic operating layer" — AI agents embedded directly into core business processes that can act autonomously within defined rules. Think of it as the difference between a colleague who gives you advice and a colleague who handles the task and comes back to you when there's a problem.
This isn't a bolt-on. These agents are woven into the standard Dynamics 365 workflows, which means they work with your existing data, your existing security model, and your existing business rules.
The key shift: Wave 1 moves Dynamics 365 from "AI that helps you work" to "AI that works for you." The agents don't just flag issues — they resolve routine ones and surface the exceptions that need your attention.
The finance updates in Wave 1 are some of the most consequential for mid-market businesses. Here's what's arriving and why it matters.
The new Account Reconciliation Agent automatically matches subledger balances to the general ledger and flags exceptions for human review. If your finance team currently spends days at month-end manually reconciling accounts, this agent handles the matching and brings you only the items that don't line up.
For a finance team of five to fifteen people, this could reclaim significant hours every close cycle — hours that can be redirected to analysis, forecasting, and the kind of work that actually moves the business forward.
The enhanced invoice capture AI learns from your team's corrections over time. When it matches a supplier invoice to a purchase order and your AP team adjusts the coding, the system remembers. Over weeks and months, the matching accuracy improves without anyone having to retrain a model or configure rules manually.
For businesses processing hundreds of supplier invoices a month, this is the kind of compounding improvement that transforms accounts payable from a bottleneck into a streamlined process.
This is a new one, and it addresses a pain point that every procurement team knows well. The Supplier Communications Agent reads incoming vendor emails and takes action based on rules your business defines — confirming delivery dates, acknowledging receipt of invoices, or flagging pricing discrepancies for human review.
The manual back-and-forth with suppliers that typically slows down procurement cycles and extends delivery confirmations? This agent handles the routine exchanges so your team can focus on the exceptions and negotiations that require human judgement.
The Finance Agent, which was introduced in preview, now supports reconciliation, variance analysis, and data preparation directly in Excel and Outlook. This is significant because it brings financial intelligence into the productivity tools your finance team already lives in, rather than requiring them to switch between ERP screens and spreadsheets.
Your FD can ask a question about variance trends in Outlook and get a data-backed answer without opening Dynamics 365. Your management accountant can prepare month-end schedules in Excel with the agent pulling and validating live ERP data in the background.
The supply chain updates in Wave 1 are equally practical, particularly for businesses managing warehousing, distribution, or manufacturing operations.
Wave 1 introduces AI-powered warehouse picking and inventory rebalancing. The system analyses order patterns, stock levels, and warehouse layout to optimise pick paths and suggest inventory movements — reducing the time and cost of fulfilment without requiring a warehouse management system overhaul.
Hands-free scanning capabilities complement this, allowing warehouse staff to work more efficiently with voice-directed and vision-assisted picking processes.
The demand planning improvements in Wave 1 go beyond simple historical forecasting. The new capabilities incorporate price-demand correlation — understanding how pricing changes affect demand volumes — and capacity-to-promise date protection, which ensures your sales team doesn't commit to delivery dates your operations can't meet.
For mid-market manufacturers and distributors, this closes a gap that often exists between what sales promises and what operations can deliver. Instead of discovering a supply risk after committing to a customer, your planners receive context-aware signals before commitments are made.
Practical impact: The shift here is from reactive exception management to proactive, AI-surfaced decision-making. Your teams spend less time firefighting and more time planning.
It's easy to read about enterprise AI features and assume they're designed for large organisations with dedicated IT teams and seven-figure technology budgets. Wave 1 is different for three reasons.
First, the agents are built in, not bolted on. You don't need a separate AI platform, a data science team, or a complex integration project. If you're already on Dynamics 365 Finance or Supply Chain Management, the agents are part of your application — they just need enabling and configuring.
Second, mid-market teams benefit disproportionately. A finance team of eight people can't absorb the same volume of manual reconciliation, invoice processing, and supplier chasing as a team of forty. When agents take over the routine work, the productivity gain per person is much higher in a smaller team. The same applies to supply chain — a warehouse team of twenty gains more from AI-optimised picking than a mega-warehouse that already has a dedicated WMS with sophisticated automation.
Third, the governance model is sensible. Microsoft has designed these agents with clear human-in-the-loop checkpoints. Agents act within the rules your business defines. They escalate exceptions. They don't make decisions outside their scope. This matters for mid-market businesses where the CFO or operations director needs to trust the system before stepping back.
If you're running Dynamics 365 Finance or Supply Chain Management today, there are practical steps you can take now to be ready for the Wave 1 features as they roll out.
Make sure you're on the latest Dynamics 365 version and that your Copilot entitlements are active. Some Wave 1 features require specific licence tiers or preview opt-ins. Your Microsoft partner can confirm exactly what's available to you.
Agents are only as good as the data they work with. Before enabling the Account Reconciliation Agent, ensure your chart of accounts is clean and your subledger-to-GL mappings are consistent. Before turning on the Supplier Communications Agent, review your vendor master data — are email addresses current? Are purchase order terms standardised?
This isn't about achieving perfection. It's about removing the obvious blockers that would cause an agent to produce poor results or flag too many false exceptions.
Don't try to enable everything at once. Identify the one or two areas where your team spends the most manual time on repetitive, rule-based work. For most mid-market finance teams, that's month-end reconciliation or accounts payable. For supply chain teams, it's often demand planning or warehouse pick optimisation.
Start there. Measure the time saved. Build confidence. Then expand.
Wave 1 also introduces natural language interfaces for custom agent design — meaning your functional users, not just developers, can create and configure agents. Invest time in familiarising your finance and operations leads with Copilot Studio and the agent design tools. The businesses that will get the most from Wave 1 are those where the people closest to the process can shape how the agents work.
Wave 1 isn't just about individual features. It signals that agentic AI is now the architectural standard for ERP. Microsoft is betting that the future of enterprise software is not just cloud-hosted — it's agent-executed. The platform is evolving from a system of record into a system that actively manages work.
For UK mid-market businesses, this is an opportunity. The tooling is mature enough to be practical, the licence model makes it accessible, and the competitive advantage of early adoption is real. Businesses that embrace these capabilities now will operate with leaner teams, faster close cycles, more accurate demand planning, and fewer manual errors — while their competitors are still doing it the old way.
The wave is rolling out now. The question is whether your business will ride it or watch from the shore.
Book a free discovery call and we'll walk you through the Wave 1 updates that matter for your Dynamics 365 environment — and help you plan your adoption roadmap.
Or call us directly: 01625 569 777