Microsoft Power Platform · Process automation
As a Microsoft Power Automate partner, we build flows that don't break down at the 10,000th run in production. When someone at your company copies data from one system to another every morning, orchestrates an approval round by email, or sends reminder emails by hand — that's exactly a Power Automate case.
Functional scope
Five components make up the platform. Each has its place, each has its trap. Here's the honest take.
Flows that run in the Microsoft cloud and talk via APIs to standard services — SharePoint, Outlook, Teams, Dynamics, Salesforce, SAP, ServiceNow, custom REST APIs. Three trigger types: automatic (event-driven), instant (button click), scheduled (cron). The workhorse of the platform.
Flows that remotely control a real Windows machine — mouse, keyboard, screen. The solution for legacy applications without an API: old SAP GUI clients, green screens, Excel forms, web UIs without an open interface. Attended (with a logged-in user) or unattended (bot on a server).
Guided multi-stage processes directly in model-driven apps. The field worker sees a progress bar at the top of the appointment form: inquiry → quote → negotiation → close. Each stage with required fields, validations, automatic actions. Very useful for sales and service pipelines.
Built-in approval logic with Teams/Outlook cards, multi-approver, parallel or sequential. Audit trail, escalation logic on inactivity, mobile approval. Saves a lot when you'd otherwise organize approvals via email threads with an Excel list.
Form recognition for receipts, invoices, contracts — directly in the flow. Sentiment analysis for customer-service emails. Object detection for image recognition. Works well for the standard cases, gets unreliable with highly variable inputs. Before using it, we check whether the data volume is sufficient for a custom model.
When a third-party system has no pre-built connector: custom connector from OpenAPI definition, auth configuration (OAuth, API key, basic), pagination handling. For one-off calls, the HTTP action with full control over headers and payload. This is often the bridge to in-house developments we build in parallel in Independent Engineering.
Use cases
The following four task types come up most often — and are the most economically clear applications for the platform.
Vacation, travel, investment, supplier release, customer terms. Application from Power App or form → multi-stage approval with Teams card → result back into the source system with audit trail. The classic application most clients start with in Power Automate.
Customer data between Microsoft Dynamics 365 and SAP, orders between web shop and ERP, employees between HRIS and Active Directory. Including conflict handling, error queue, monitoring. A production-grade synchronization is much more than a "copy on change" flow — we know the pitfalls.
New employee created in the HR system → automatically AD user, mailbox, Teams memberships, licenses, hardware order, welcome email with credentials, IT-security-briefing appointment. Saves several hours of manual work per onboarding for the IT team.
Contract expires in 90 days → notify the account manager, in 60 days escalation to lead, in 30 days an automatic calendar appointment. Service ticket without response after 24h → escalation to service lead. Pure time logic that no one else would take on — until it's too late.
What we deliver
Where does which logic live? One big flow or several small ones? Synchronous or asynchronous via service bus? Child processes as reusable sub-flows? These architectural decisions determine whether you have a maintainable landscape in 12 months or a flow graveyard.
Retry logic with exponential backoff. Idempotency, so the same job isn't processed twice. Dead-letter queue for unprocessable inputs. Monitoring emails to the right people, not to everyone. That's the difference between a demo flow and a production flow.
Standard or premium connector? When nothing fits: custom connector from OpenAPI, with auth mapping, pagination, throttling logic. For your own REST APIs, we write the OpenAPI definition along the way, and then the connector is a 30-minute task instead of a multi-day project.
SAP GUI bots, old Windows clients, browser apps without an API. With ID-based element selectors instead of pixel clicks (much more stable). With concurrency locks so two bots don't write into the same window at once. With image recognition as a fallback when the UI has no IDs.
License cost
Three layers drive your Microsoft bill:
Pay-as-you-go via Azure subscription is also available — billed per flow run. Useful for sporadic tasks with small volume. A fully calculated variant with your situation: License Cost Calculator.
FAQ
Cloud Flows run in the Microsoft cloud and talk via APIs to services like SharePoint, Dynamics, Salesforce, SAP. Desktop Flows control a real Windows machine — keyboard, mouse, screen — and are the choice for legacy applications without an API.
Cloud Flows with standard connectors are included in Microsoft 365. Premium connectors (SAP, ServiceNow, Salesforce, custom HTTP APIs) need a Power Automate plan from ~€15 per user per month. RPA with attended Desktop Flows from ~€40, unattended RPA as a bot license from ~€150 per bot per month. More in the License Cost Calculator.
Yes. That's exactly what Power Automate Desktop (RPA) is for. We automate green screens, old Windows clients, browser-based applications without an API — anything that a mouse and keyboard can operate. Important: RPA is a bridge, not an architecture. Where possible, we replace RPA over time with clean API integrations — either with custom connectors or, when complexity warrants, with Independent Engineering.
That depends almost entirely on error handling. We build every production-relevant flow with retry logic, idempotency, dead-letter queue, and monitoring. The difference between a demo flow and a production flow is usually the error handling — and that's our daily business.
A single Cloud Flow with a clear use case and 2–3 connector steps can often be set up in 1–2 days. Complex automations with many branches, error handling, and an RPA component typically come to 5–15 person-days. In the initial conversation we'll give you an honest assessment based on your use case.
We offer maintenance in the Application Care model with a guaranteed hourly allotment or pay-per-hour without contract commitment. When Microsoft makes API changes or throttling adjustments, we contact you proactively and roll out the adjustment as part of the running contract.
30-min initial conversation
Describe the current sequence, and we'll give an honest take — Cloud Flow, Desktop Flow, or better a custom-code solution.
Accompanying services
Engineering projects rarely stand alone — license logic, architecture clarification, quality gates, knowledge transfer, and follow-on operations usually run in parallel. Below are the most common accompanying services we add via discovery spikes, fixed-price sprints, or application-care contracts.
Before · Architecture
Before implementation: tenant structure, data model, security concept, integration mapping. The result is an architecture document any engineering team can continue working with — even one other than us.
Read more →
Before · CSP
Which license bundles for which users, which add-on SKUs are necessary, where you're over- or under-licensed. Sourced as a Microsoft Licensing Partner — with the option to use CSP purely for control without margin maximization.
Read more →
During · Quality gate
Independent second opinion during an active implementation project — whether we run it or another partner does. CMMI-based quality gates, risk reviews, fixed price per gate.
During · Adoption
Not the classic 2-day workshop forgotten after a week — but a dynamic learning program over 4–6 weeks with initial training, application phases, and follow-up sessions. Training matrix for roles and topics.
Read more →
After · Operations
After go-live: a plannable application-care contract with a monthly flat rate, SLA-based. Includes releases, hotfixes, extensions, tenant hardening — and continuous accompaniment instead of mere ticket response.
Read more →
After · Knowledge
When the original developers are gone, the previous partner is no longer reachable, or the documentation is stale — reverse engineering of the existing solution with a documented result: code map, data model, customization inventory.
Read more →
Strategy background · arades topic page
Why we consistently build all our apps and recommendations on a single data foundation — and when the island solution is still the better answer. 2,500 words on architecture, migration paths, and honest boundaries.
Read topic page →