How assumptions and misguided expectations are holding back businesses in South Africa
Most of us have had a moment with a chatbot that made us want to pull out our hair, or put our phones in the microwave. And it’s usually because it didn’t understand a basic request.
Mention a chatbot in a South African boardroom and you sometimes get rolling eyes and subject changes. But the fact is that some of the best-performing customer teams in South Africa are already using them, and not as a gimmick, but as a core part of how they operate.
So what’s really going on?
The gap isn’t the technology. Sometimes it’s the journey, sometimes it’s implementation, but a lot of the time it’s perception.
Let’s unpack a few of the myths that are stopping SA businesses from creating the best customer service bots humanity has ever known.
Myth 1: Chatbots replace humans
This is usually the first concern and it misses the point entirely. No one is replacing your service team. What’s happening instead is a shift in how that team works.
High-volume, repetitive queries such as balance checks, order tracking, or policy questions, don’t need a human to answer. They need speed, accuracy, and availability.
In our experience with local retail and banking sectors, we see teams drowning in 'Where is my order?' or 'Reset my PIN' queries. These are tasks that eat up around 60% of a human agent's day for no proper gain. When those queries are handled automatically, your team isn’t overwhelmed. They’re fully focused on the conversations that actually need judgement, empathy, and creative problem-solving.
It’s about clearing the decks so your best people aren't stuck doing data entry.
Myth 2: Chatbots are just glorified FAQ pages
That might have been true a few years ago, but today’s chatbots – or as we like to call them intelligent assistants or digital agents – are doing far more than answering basic questions. They’re onboarding customers, qualifying leads, guiding transactions, and resolving service requests, which are often across multiple channels at once.
And they’re much better at resolving queries because of their Natural Language Understanding capabilities, which mean they can comprehend what the user is asking, even if they don’t say it exactly right or choose it off a menu.
Gartner predicts that by 2027, chatbots will become the primary customer service channel for around 25% of organisations globally. In sectors such as telecoms, this shift is already happening. Customers can activate services, update details, or complete processes without ever needing to call a contact centre.
Myth 3: They’re expensive and complicated to implement
This one tends to come from past experience. Early chatbot projects were expensive and often over-engineered.
With cloud-based platforms and more mature AI models, deploying intelligent assistants in South Africa is far more accessible than what it used to be. More importantly, the cost conversation has shifted from “What does it cost?” to “What does it save?”
In some cases, we’ve saved our customers over R100 million in under a year.
Juniper Research estimates that chatbots will help businesses save over $11 billion annually by 2027, driven by efficiencies in customer service operations. When a service interaction drops to a fraction of the cost of a call centre query, the solution becomes fairly straightforward.
(And remember, the human agent’s role then becomes more efficient, and of a better quality for the queries that automations can’t solve.)
