AI: removing jobs or enabling change?
Belinda West (MCIPS): The Non-Trade Specialist
Written by: Belinda West (MCIPS)
Belinda West (MCIPS) is the Head of Non-Trade Procurement for Woolworths, a leading South African retailer in the food, fashion, homeware and beauty markets. They have over 34,000 staff and have been experimenting with various AI tools for three years.
When you’re talking about the effect of AI on jobs in a South African context, you need to realise that there are fundamental differences between South Africa and other regions. These differences are particularly apparent when comparing South Africa with Europe or the US and can be traced back to the fact that labour costs are so much lower.
AI can be used for buying, with autonomous negotiations potentially removing the need for large numbers of buyers working in low-value areas. I can imagine those jobs might be at risk, especially if you’re in mining, petrochemicals or pharmaceuticals, with particularly deep pockets, that has established a large team of buyers.
In South Africa, however, labour rates are low compared to Europe and the US. That means it will take a longer time before reaching the tipping point of saying that the cost of AI enabled solutions are more cost effective than people-based solutions.
Furthermore, in post-apartheid South Africa, we have legislation around transforming the workforce and a great necessity to reduce unemployment, so for corporates there is a tension here around labour cost and our ethical responsibility to try and protect labour.
We need to consider what change means for the workforce, and where we can utilise employment cost-effectively to put more money into the economy. AI solutions can be more expensive than people solutions and often entail sending payment to an offshore company, bringing foreign exchange risk.
Same technology, different imperatives
That doesn’t mean we don’t experience the same pressure to do more with less, as developed countries have. When it comes to AI, our focus has included restoring work-life balance for exhausted employees through easy access to insight and efficient execution. This has not only enabled us to extend reach and unlock trapped value, but it has also created space for reflection and strategic action.
"When it comes to AI, our focus has included restoring work-life balance for exhausted employees"
Many people in the workforce are from a generation that finds the ever-increasing number of different sources of contact to be overwhelming. Excessive email alone was challenging, but now you also have to stay on top of WhatsApp, Teams, in-app messaging, and a few more apps on top of that. There is an expectation that AI can help us with that.
The limitations of AI
We use the analytical capability of AI to analyse our spend, and the time to achieve the expected level of insight and confidence in the classification has been significantly longer than anticipated.
When it comes to extending our use of AI into other aspects of sourcing, I’ve been known to say that I don't have time for a toddler. I'm going to watch others enjoy realising the benefits of early adoption. As AI hits puberty, I'll lean in further, because there’s clearly an opportunity, but the technology is not quite as plug-and-play as people anticipate. Realising the benefits requires sustained time investment, and the time committed to this investment at implementation directly influences the speed and scale of progress.
"As AI hits puberty, I'll lean in further"
There is more opportunity to implement AI than extending sourcing systems. We’re experimenting with Microsoft Copilot in a closed environment and finding that there's a huge amount of value in just staying on top of your minutes of meetings - and there are many ways Copilot can free up time in your day.
There are a number of things we never reached due to time constraints, that the team is finally starting to work on.
The role of data
Data has always been a challenge in the world of procurement. There is a need for high-quality data, especially when you have decentralised requisitioning, but the adage ‘garbage in, garbage out’ remains true and continues to be the root data challenge. An opportunity for AI in intake orchestration is the possibility of structured ‘quality in’.
I’ve observed data analytics listed as a procurement skill of the future, but I don’t agree. I believe when tech and AI deliver as anticipated, we will need less data-savvy people - AI will be serving up the answers.
What I need is people with uniquely human skills, not people who can manage data and make sense out of it. Skills that enable genuine partnership with the business, people who can take the pictures they are being served by AI, consider the available intelligence and guide the business to the best commercial advantage.
Moving forward
When I first started building the business case for implementing upstream procurement technology, I read all the research I could and went to every webinar on the topic. I wanted to see if such a system would benefit a company like ours that had a mature procurement team, which had already taken significant fallow value out of contracts.
Everyone who was talking about such technology, was emphasising the need for AI, and if you didn’t have it, you were going to be severely limited in the short term. Now, years on, the tone in webinars is so much more moderate about what benefit we can expect in the short term. Clearly, the pace of development has not met expectations.
There is also plenty that can change before advances in AI make it possible to consider reducing resourcing. When that moment arrives, we’ll likely be running after other things. As much as AI is coming in and changing things, I’m not convinced that it’s going to be as simple as swapping humans for technology - not just yet, anyway.
The Brand Champion: Sarah Simpson
“There are still going to be entry-level jobs. They’ll just be different jobs that demand different skills”
The Public Sector Voice: Liam Osborn
"It isn’t AI that’s the risk, so much as how we use it"
The Procurement Manager: Georgia Hennessey
"Human connection can’t be replaced"
The Industry Influencer: Imran Shareef (FCIPS)
"AI is already changing the conversation at the leadership table"
The Tail Spend Guru: Oliver Norman
“What AI is going to do for the procurement industry is elevate people into positions where they can do the jobs that technology can’t”
The Practitioner: Wame Sedirwa (MCIPS)
“We must balance AI’s efficiencies with human oversight”
The Thought Leader: Ram Trivedi (FCIPS)
“Any change brings resistance, and AI is going to bring big changes to our way of life”
The Expert: Joseph A. Yacura
If generative agents are adopted, we could have a supply chain function operating in a ‘lights out’ way by 2028”