RPA First, Then AI.

"He who would learn to fly one day must first learn to stand and walk and run and climb and dance; one cannot fly into flying." Friedrich Nietzsche

We’re often told that mastering prompt engineering is the key to thriving in the AI era.

But in my experience, the real power to optimise and streamline business processes has been at our fingertips for years.

The RPA Alert

This potential is called the RPA - Robotic Process Automation.

The hype around RPA was there about a decade ago, and robots were predicted to take over back-office jobs. Around the same time (November 2016), Microsoft launched Flow, which later became Power Automate.

How many of us were aware of Microsoft Flow at the time? Even fewer actively used it.

Nevertheless, the "robot takes your job" alerts continued to flash until AI entered the picture. Now, it’s AI we’re told to fear.

It's not the Revelation, just an Opinion.

The following observations are based on my personal experience, gained from operating Power Automate, both with and without AI elements.

RPA has long provided about 90% of what’s needed for business process automation.

AI spices up the RPA and adds topical improvements. e.g., large document summary, comparison of newly submitted documents against reference one, data extraction from structured forms, etc.

But AI hasn’t replaced RPA. Not yet, at least from what I’ve seen.

Instead of diving headfirst into prompt engineering, it may be wiser to first master basic process automation.

It’s intuitive, cost-effective, and readily available within the Office 365 suite. Plus, it integrates seamlessly with existing apps and offers robust security.

JSON news for Prompt Learners

Learning the art of prompting is essential, but not sufficient.

Ideally, you should study JSON (JavaScript Object Notation), which is a universal language that both AI and RPA understand.

I started with Copilot and ChatGPT by my side, helping me to structure JSON expressions, which can be particularly tricky due to their complex syntax rules.

Dynamic AI prompts

If you want AI to really help out with your workflows, your prompts need to be dynamic. This means they should change and adapt based on new inputs in your process.

Check out the screenshot below—it shows a prompt with four dynamic elements that I've used within my flow.

Power Automate interface showing a "Category Plan Update" prompt with various dynamic content fields and functions, including a conditional expression for handling suggested edits.

RFQ Evaluation Automation with Power Automate and AI Builder

This case involves a blend of JSON, multiple variables, arrays, and content controls. It’s genuine automation with a touch of AI, where I trained a model to extract data from standardised Word forms.

You’ll need to:

• Describe the process as a Business Analyst

• Tailor it to AI’s need for structured data

• Manage GPTs to help assemble the flow and navigate the JSON intricacies

• Run numerous test loops

• Safeguard the process logic

Before you aim to fly, ensure you’ve mastered walking. Let’s not rush into AI without first grasping the fundamentals of process automation.

P.S. Intelligent Automation.

Visual representation of Intelligent Automation, highlighting its intersection with Robotic Process Automation (RPA) and Artificial Intelligence.

When I wrote the above, intuitively, I tried to explain the Intelligent Automation.

It's nice that such a concept exists. Yet, it suggests that we have to develop ourselves in a balanced manner, not just getting overweight on the AI side.