Embracing AI would lead us to data-driven insights, improved forecasting, anomaly detection and automation. It would remove many of the current time-consuming repetitive tasks. It could potentially release our teams from hours of work pointing them to more value-add projects. If we embrace technology, this is what our future holds. 

For a moment though, let's un-mask the “alphabet soup” of tech. AI, ML, NLP, RPA what does all it all mean?! 

• Artificial Intelligence (AI): algorithms that exhibit ‘smart’ behavior; encompasses machine learning and natural language processing
• Machine Learning (ML): algorithms that detect patterns and use them for predictions or decision making 
• Natural Language Processing (NLP): algorithms that can interpret, transform, and generate human language 
• Robotic Process Automation (RPA): algorithms that mimic human interactions to reduce repetitive, simple tasks 

Individuals often lump all four acronyms and software functions into one AI catch-all. The reality is they function very differently. In fact, RPA is not intelligent at all. It’s simply automated.  

With the advancements being made in technology, specifically artificial intelligence and machine learning, there will be an inevitable knock-on impact to how we negotiate. Yet whilst it will be possible to expedite simple buying and selling tasks AI will not, in the near future, be able to replace people to run complex tender processes and strategic procurement initiatives by itself. As such, the core use and benefit of technology will be in the information it can provide. 

If data-driven decisions are what you need, you will require a rich dataset to draw from. Let’s assume for a minute you have all the time in the world. You would have to parse the data by establishing clusters, trends, identify and understand outliers and the circumstances they present themselves with, predict future trends, and then extrapolate it 5 years. Assuming you don’t have all the time in the world, AI could do that work for you.  

Too little time is spent understanding exactly what AI is. It’s not the all-knowing being that is here to solve your business problems. AI is about helping you solve your business challenges by making an informed decision. As a procurement professional, AI can help you: 

• Analyze responses to a request for proposal that highlight soft or flexible language (NLP) 
• Research traffic patterns and suggest optimal delivery schedules (ML) 
• Predict replenishment or maintenance programs based on season for the next 3 years (AI) 
• Purchase a commodity item every time the price reaches $3/unit (RPA) 

For different tasks, different aspects of AI could be leveraged. Understanding WHY you need a specific piece of technology or software is more important than the software itself. Think deeply about what task you are trying to automate before choosing a desired platform. Once you have done that, you are able to free up human resources to complete more complex and high-value tasks. Imagine you were able to give your procurement team back 20 hours of their week due to automation by AI that still meets your requirements? Imagine what they could accomplish...  

Equally, you can buy the latest groundbreaking technology, but if you don’t take the time to change, adapt, and develop all of the human and manual processes touch points that engage with it it’ll become an isolated and expensive talking point on a marketing brochure. Too often, organizations will make significant investments in technology and not realize how important it is to train the individuals leveraging the platform. Only to realize 12 months later that they need to re-train their staff appropriately because the team’s frustration causes churn.  

How do I get started?

That’s a complicated question. Is your business even ready for AI? Here are a few things to consider: 

• Have I identified the specific problem I need solved? 
• Do I have enough data on that specific issue? 
• Have I narrowed the field enough for an AI to be successful? 
• Do I have the right software that can analyze the dataset provided? 
• What type of oversight is needed? 
• Are my people trained to understand and implement the analysis? 

Those are just a few of the many questions that need answers before you begin to leverage an AI within your process.  

AI isn’t objective, humans aren’t either  

AI isn’t going to replace humans because humans are contextually aware and can problem solve outside of a defined dataset. Case and point, the AI will do exactly what it’s instructed to do given the information it’s provided. Do you want your Roomba to map your first floor, remember where the sofa is, and intelligently change the vacuum mode to deep clean the carpet? Done. If you bring that same Roomba upstairs and expect it to know what that floor looks like based on what’s on the first floor, it will fail. The first floor was the only dataset it had.  

Be extremely careful on the dataset you provide and are asking the AI to analyze. If your information is biased, incomplete, the clusters undefined, you name it, the AI will bring back insights that may be misaligned with reality. It’s like trying to analyze work-week traffic patterns and only giving the system a dataset for Sundays.  

The future is in the net 

AI will fundamentally change how people work quicker than Embracing AI would lead us to data-driven insights, improved forecasting, anomaly detection and automation. It would remove many of the current time-consuming repetitive tasks. It could potentially release our teams from hours of work pointing them to more value-add projects. If we embrace technology, this is what our future holds.

If you are interested in learning more about this topic, please download our report, Procurement: Is there a negotiation gap?, below.