Microsoft bakes new Copilot tricks into Dynamics 365 and Power Platform
Microsoft just kicked off its 2023 release wave 2 for Dynamics 365 and Power Platform, rolling out fresh Copilot capabilities across ERP, customer service, and low-code development. Teams can now generate Power Automate workflows with plain text and receive automated purchasing recommendations, with features rolling out through March 2024.

For decades, changing an enterprise procurement flow or generating custom reports meant waiting months for IT tickets to clear. Microsoft is now betting that frontline workers can simply describe what they need in plain English.
At its Business Applications Launch Event, Microsoft officially rolled out 2023 release wave 2 for Dynamics 365 and Power Platform. The rollout introduces hundreds of new features slated through March 2024. Notably, the tech giant revealed that over 130,000 organizations have already experimented with its Copilot capabilities.
The clear theme across this wave is deep integration. Rather than treating AI as an isolated chat interface, Microsoft is embedding Copilot directly into enterprise resource planning, customer support, and low-code tooling.
From pizza forecasting to grid management
Enterprise software often bogs down in manual data entry and lagging forecasts. Microsoft brought out heavyweight enterprise examples to prove Copilot can handle the heavy lifting.
In supply chain operations, Domino’s Pizza UK and Ireland is using the new demand planning features to accurately predict ingredient requirements, cutting down food waste while streamlining purchase orders. Copilot can now actively recommend mitigation steps to procurement agents when open purchase orders face disruptions.
In New Zealand, telecommunications provider Spark NZ is automating vendor invoice processing, foreign currency transaction revaluations, reconciliations, and complex tax calculations through Dynamics 365.
Frontline service operations received a parallel overhaul. Wealth manager Northern Trust relies on Copilot for account reviews, case investigations, and wrap-ups. Meanwhile, British energy giant Centrica uses Dynamics 365 Field Service to manage and dispatch its 12,000 on-site field technicians.
Building apps with everyday language
The Power Platform updates are equally aggressive, aiming to turn non-technical employees into citizen developers through conversational prompts.
Suffolk, one of the largest construction firms in the US, demonstrated how its teams use Copilot in Power Automate to generate complex material request workflows simply by typing what they want. They also deploy Power BI to automatically surface safety incident trends and near misses across job sites worldwide.
Elsewhere, Kodak Alaris plugged Copilot into Dynamics 365 Sales and Customer Insights, unearthing legacy customer data to target accounts and draft customized outreach content without manual data mining.
Why it matters
While consumer AI often grabs headlines with conversational novelty, Microsoft is proving that enterprise software is where generative AI actually pays the bills. For business operators and line-of-business workers, the immediate appeal is obvious: Copilot removes the grunt work of wrangling sprawling databases, reconciling ledgers, and building internal forms from scratch.
Strategically, this rollout shows how Microsoft leverages its unique enterprise moat. Competitors like Salesforce, SAP, and Oracle have announced their own generative assistants, but Microsoft uniquely connects Dynamics to Teams, Outlook, and Power Platform under one continuous authentication and data layer. Having your field agents trigger Power Automate flows directly inside Teams is an integration depth that standalone SaaS rivals struggle to match.
Still, we think organizations need to approach this with calculated caution. In ERP and supply chain environments, a single hallucinated figure in demand forecasting or currency revaluation can compound into severe financial discrepancies. Furthermore, empowering citizen developers to spawn automated workflows via natural language risks creating a tangled web of unmonitored shadow IT if administrative governance is not strictly enforced from day one.
Sources (1)
- cloudblogs.microsoft.com(opens in a new tab)Originalcloudblogs.microsoft.com