AI in CX: you need to lean in or lose out. - guest blog by Jonas Berggren

2023 was the year when artificial intelligence entered the boardroom. An enormous amount has been written about how this will change the customer experience of brands and the role of the contact center agent. Some of the commentaries have been about the end of the contact center as we know it, the typical doom-laden media stories. Other commentary has focused on the weaknesses of AI and how it can’t always be guaranteed to be correct.

The reality is somewhere in the middle. AI hallucination will be tackled by creating more specialized language models (as I wrote about recently here) and AI isn’t about to instantly replace millions of jobs, but it can replace or automate specific tasks and therefore make workers much more productive.

The CX focus for AI has been in two areas:

• Customer interactions: chatbots can be improved using AI and Generative AI (GenAI) so they are essentially very interactive knowledge bases filled with product information that the customer can query and search using natural language interactions.
• Agent augmentation: manual and repetitive processes that agents need to manage inside the contact center - such as writing up notes on a call - can be automated by AI. AI can become a digital assistant for each agent, helping with suggestions and even coaching tips.

My main message to executives on this subject is that the world is changing rapidly, and technology is accelerating. It’s impossible to build a five-year business strategy based on what we know today. Customers may have entirely different expectations a year from today.

On the plus side, off-the-shelf technology is now extremely good. There was a time when any company that wanted very sophisticated bespoke software needed to hire a technology specialist like IBM. This is no longer true.

You can pilot and test ideas around how AI can improve your customer service processes without spending millions of dollars - especially if you are working with a partner that has already been focused on testing innovation. The key is that you need to adopt an exploratory culture that truly embraces innovation.

What do I mean by this?

When most companies plan a customer service process, either internally or through a customer service specialist, they plan it, design how it will work, test it, then implement the new service or process. It’s very much like the old school ‘waterfall model’ used in software engineering, where a system is planned, designed, coded, tested, and then delivered. Each phase is a distinct step that can only be started when the previous one is complete.

With AI and other new technology developments, the world is moving too fast for this kind of design methodology. We need to take ideas and be able to implement and test them rapidly - reversing the ones that don’t work and reinforcing the ones that do.

This goes against the grain of most clients' relationships with their customer service specialists. It is easy to imagine customer service directors all saying, ‘We didn’t contract your services so you could deliver a function that doesn’t work!’

But this is how innovation works.

Innovation relies on ideas - lots of them. When you are creating new ideas for how AI might be able to improve your customer service processes, it will be impossible to know which will work technically and which might work in development. Still, they just don’t work for the customer. You can’t know in advance which ideas will work without testing them, and this means that you need a new culture that tests ideas and quickly rejects failures.

Amazon is known globally as one of the most innovative companies in the world. They are highly successful in e-commerce and cloud computing, and almost every home now has a smart device - Alexa or Kindles. However, there is also a very long list of Amazon failures. The only reason that Amazon has all their successful ventures is because it keeps trying new ideas and quickly retiring the ones that don’t work out.

The rapid growth of AI in the customer service environment means that everyone designing customer service processes will need to start embracing this culture of innovation to a much greater extent.

I don’t mean that reliable business processes will change daily and service outages will become common because it’s all in the name of innovation. However, company leaders must test more ideas because the software and platforms are very flexible, and every company has different requirements.

AI for customer service demands experimentation and testing. Pilots need to become a normal process as new ideas are tested on small groups of customers. Brands need to adopt this innovative mindset.

The bottom line is that AI is reshaping quality, productivity, and efficiency in the customer service environment. Agents can serve more customers, get the right information the first time, and ensure that quality is constantly improved.

Companies that don’t lean into AI for CX will simply lose out