Subscribe to the blog

Get the latest Procurement insights from our experts

Digital trends affecting Procurement – RPA and P2P (part 3)

by | February 7, 2019

Technology is fascinating. It makes us realise that what we used to consider Sci-Fi as children is becoming normal, everyday reality. We’ve now covered what AI and ML can do for procurement, how they’ve simplified complicated task and automated rigorous processes. And this is merely a small part of what AI and ML technologies can achieve.

Although not as Sci-Fi resembling, robotic process automation is one of the 3 key digital trends that promises to fundamentally influence and change the way we work going forward.

Because sometimes it’s about the tedious, manual and repetitive tasks, like transaction processing, data migration and data inputs, all of which requires time and accuracy. Automating processes in these areas is one of the first steps companies should take on their journey to digital transformation. So, what is RPA and how can it help with business processes in general and procurement processes in particular?

What is RPA

“RPA is software that automates manual and repetitive tasks. It works by mimicking human activity and replicating it as the user takes actions – usually through their desktop. RPA is an emerging form of business process automation technology based on the notion of software robots or AI workers.”


The difference between RPA and AI was the topic of many debates. Although 2 very different technologies, these 2 concepts still cause confusion for many. To make things clear, RPA is software robots that mimic human actions while AI is the simulation of human intelligence by machines.

To explain these 3 digital trends on the most basic level, RPA is concerned with the “doing” of things while AI and ML are concerned with the “thinking” respectively “learning”.

Let’s take invoice processing as a practical example. When you receive the invoice from your suppliers by email, you then need to download it into a folder, extract the right information and create the bills for your accounting software.

In this scenario RPA would be involved in retrieving the invoices from your email (usually based on the subject of the email), downloading the invoices into the appropriate folder and creating the bills in the accounting software (mainly based on copy + paste).

On the other hand, it’s down to AI enabled software when it comes to extracting the appropriate information from the invoices, such as invoice number, the name of the supplier, product description etc. This task needs “intelligent” software as not all invoices have the same structure or contain the same information and not all suppliers use the same templates.

Adding AI to RPA

RPA is considered the first step into digital transformation and adding AI to it seems the next logical transformational step, as more and more business processes are being automated by increasingly sophisticated means.

Researches show that combining these 2 technologies can help automate up to 70% of a company’s manual processes which leads to freeing up employees from doing repetitive, low value tasks and helping them focus on more strategic, value adding tasks. Adding AI to RPA can up your process automation game and help companies automate tasks that were not possible before. For example, applications of advanced AI in RPA could include image or text analysis.

Where to start with RPA adoption?

Before adopting RPA it’s important to know your processes. Have a clear image of how your business is running and a clear documentation of your manual processes. It’s also important to analyse each process and see if there is potential for automation.

A good tip is to start small and don’t engage in large scale automation projects from the beginning. this will allow you to get smaller and quicker wins, demonstrating value to the business along the way.

Starting small doesn’t mean you have to limit yourself to a few process automations here and there. Your vision needs to include the whole business. Your ultimate goal should be to standardise and automate end to end.

Ultimately, involve your team in the process and make sure everyone is aware of the changes and benefits RPA is bringing to the business.

RPA and P2P

There is potential for the use of RPA in procurement, specifically within purchase to pay. However, the best place to start is with those manual, time-consuming and repetitive tasks that require accessing multiple systems or are audited for compliance periodically.

Most often procurement is dealing with high volumes of requisitions, purchase orders and invoices, and RPA can help transfer the data from one system or form to another and can also be helpful in maintaining a vendor database and resolving price or quantity discrepancies.

As a best practice we don’t recommend the implementation of pure RPA in procurement. These digital technologies are better used together as part of an integrated Source to Pay solution.

It is more useful to understand how AI, ML and RPA are becoming ingrained in existing procurement applications, and how these applications support the solving of issues that are specific to your business and your goals.