Home / Insights

Power Automate in New Zealand: where it fits, where it struggles, and what to pair it with

Microsoft Power Automate is often the first automation tool a New Zealand organisation already owns. If you run Microsoft 365, you probably have some of it today. That makes it a sensible place to start, but not always the right place to finish. It is excellent at connecting Microsoft and mainstream cloud apps with simple, event-driven flows. It is weaker when processes get long, data gets heavy, legacy systems have no API, or the work needs real judgement. Knowing which side of that line a process sits on saves a lot of rework later.

, 4 minute read

What Power Automate actually is

Power Automate is Microsoft's workflow platform. It covers a few distinct things that are easy to blur together:

  • Cloud flows react to events, such as a new email, a form submission or a file landing in SharePoint, and move data between connected services through connectors.
  • Desktop flows are Microsoft's robotic process automation. They drive applications through the screen, the same way a person would, for systems with no usable API.
  • Approvals and business process flows add structured sign-off and guided stages, mostly inside the Microsoft ecosystem.

It sits alongside Power Apps, Dataverse and Copilot Studio as part of the Power Platform. For many teams the appeal is simple: it is already licensed, already trusted by IT, and close to where the work happens in Outlook, Teams and SharePoint.

Where Power Automate is the right tool

Power Automate earns its place when the process is short, the systems have connectors, and the volume is moderate. Good examples include:

  • Saving email attachments to the right SharePoint folder and notifying the owner
  • Routing a Microsoft Forms request to a manager for approval, then logging the outcome
  • Creating a Planner task or Teams post when a record changes in a cloud app
  • Sending reminders when a contract, certificate or review date is approaching

These are the kinds of jobs where a citizen developer can build something useful in an afternoon, and where the business owns and understands the result. That is real value, and it should not be dismissed just because the flows are small.

Where it starts to struggle

The trouble usually starts when a flow that began small becomes business critical. Common warning signs:

  • Long, branching logic. Flows with dozens of conditions and nested loops become hard to read, test and hand over. Changes start to feel risky.
  • High volume or large data. Throttling limits, run duration caps and per-action costs show up quickly when you process thousands of records a day.
  • Legacy and on-premises systems. Desktop flows can bridge the gap, but screen automation is more fragile and needs proper infrastructure, monitoring and support.
  • Weak lifecycle management. Flows built in personal accounts, without environments, source control or naming standards, become orphaned when the builder leaves.
  • Unstructured inputs. Free-text emails, scanned documents and inconsistent forms need interpretation that simple conditions cannot provide.

None of this means Power Automate is the wrong platform. It means the design, governance and surrounding tools need to match how important the process has become.

Licensing shapes the design more than people expect

Many organisations find out about Power Automate licensing halfway through a build. Standard connectors are generally covered by Microsoft 365, but premium connectors, custom connectors, Dataverse and unattended desktop flows typically need additional licences. Per-user and per-flow or per-process options suit very different patterns of use.

The practical advice is to check licensing at design time, not deployment time. A flow that needs one premium connector for a single step might be better split, moved to a different integration approach, or licensed per process rather than per user. Microsoft changes its plans periodically, so confirm current terms against Microsoft's licensing guide and your agreement before committing.

What to pair it with as processes grow

Mature automation programmes rarely rely on one tool. Power Automate works best as one part of a wider design:

  • API integration for high-volume, system-to-system movement of data, where reliability and performance matter more than ease of building.
  • Enterprise robotic process automation, such as UiPath, where you need robust unattended bots, orchestration and queue management across many legacy systems.
  • AI agents and Copilot Studio for the steps that need interpretation: reading free-text requests, extracting data from documents, or drafting a response for a person to check.
  • Human review at the points where an error would be costly, built in as a deliberate step rather than an afterthought.

The question is not which tool wins. It is which tool suits each task. We cover that decision in more detail in when to use deterministic vs agentic automation.

A sensible way to get started

  1. Map the process first. Know the steps, systems, volumes and exceptions before choosing a tool.
  2. Set up environments and ownership. Separate development and production, use service accounts, and decide who supports each flow.
  3. Check connectors and licensing early. Confirm what is standard, what is premium and what it will cost at your expected volume.
  4. Start with a contained, measurable win. Pick a process where time saved and errors avoided are easy to show.
  5. Plan for growth. Decide in advance what would trigger a move to integration, RPA or AI for parts of the process.

Flex helps New Zealand organisations design and deliver business process automation across Power Automate, enterprise RPA, integration and AI agents. If you have flows that have outgrown their original design, or you are deciding where to start, we are happy to talk it through.

Related reading

Sources and further reading

Questions we are asked

Is Power Automate included in Microsoft 365?

Many Microsoft 365 plans include Power Automate for standard connectors and cloud flows. Premium connectors, Dataverse and unattended desktop flows usually need extra licensing, so check your agreement before building.

Is Power Automate the same as RPA?

Partly. Power Automate desktop flows are Microsoft's RPA, but most Power Automate use is cloud flows that connect apps through connectors and APIs rather than through the screen.

When should we use UiPath instead of Power Automate?

Consider an enterprise RPA platform when you have many unattended bots, complex legacy systems, high volumes, or need strong orchestration and queue management. Many organisations use both.

Can Power Automate use AI?

Yes. Flows can call AI models to classify, extract or draft content, and Copilot Studio can add conversational agents. The design question is which steps genuinely need AI and where a person should review the output.

Talk to us about what you would automate first

Your first consultation is free and comes with no obligation. It is a working session, not a pitch: bring a process that is costing your team time and we will tell you plainly whether automation fits it, which platform we would use, and what a pilot would look like.

Book a free consultation