In the first article of this 3-part series, we explored the opportunities that AI can provide to create a more valuable customer experience through scalability, pattern recognition, and providing more immediate responses to customer needs and challenges. All of that sounds great, right? That said, artificial intelligence is not without its challenges. Now let’s discuss some things to watch out for when introducing AI into your CX strategy.
AI can help to easily scale your marketing, sales, and service very effectively in some ways but there are some other ways where scale is still at least partially dependent on humans to manage and understand. For instance, AI can create an unlimited number of dynamic audience segments based on behavior patterns, but in order to make the most use of those segments, your brand would need to create content variations and offers that would appeal specifically to them.
Thus, until generative AI tools that dynamically create text and imagery are dependable enough to reliably create brand-safe content in a scalable manner, the work would fall on human teams that will quickly be overwhelmed by an infinite number of text and graphics combinations. Additionally, the time spent on each combination may not be economically feasible to create. This was a significant challenge in the digital advertising ecosystem for many years where the cost of content was too high relative to the revenue generated for thinly sliced audiences. Thus, uniform scalability between areas that AI can easily manage and the areas that humans are still needed to manage can prove challenging for some types of content.
That said, scalability may be easier to pull off for content that demands a lower production quality level. For example, ecommerce detail pages and digital shopper marketing content have simpler requirements than video content and interactive advertising. Approaching personalized content in this way accounts for a larger share of the growth of digital content and presents significant opportunities to tailor content and calls to action to micro-segments as you can test on distinct cohorts that arrive regularly over time. AI can help fuel a virtuous cycle where you both lower the cost of research and testing, while reinvesting the savings in scalable approaches to deliver more targeted and personalized experiences that deliver a higher ROI and keep the flywheel turning.
This could be said for any small or large change among teams, but AI-based platforms can cause many internal changes to both processes as well as the way of thinking through problem-solving. This means that people’s roles and responsibilities might shift in small or large ways, and that many individuals that have grown used to a specific way of working will need to start thinking differently about their work. If you’ve been through change management before, you already know that this can be challenging even for agile, high-performing teams, but change always comes with its challenges.
While we’re talking about change and CX, we should also talk about the change that is sometimes required on the part of the customer as well. Even though AI-based tools like conversational AI, or self-service tools can be time-savers and get to quick solutions, they are still often changes from the status quo. Thus, your change management needs to incorporate customer communication and insights in addition to internal communication and training.
One effective way to approach this is through the use of design thinking methods that can enable teams and their leaders to see the behaviors they want, and then test the content and calls to action that best drive that desired behavior. This also allows you to better link content, personalization, and modifications to the customer journey with business outcomes.
It can be very rewarding for both customers and the business when AI-based tools provide quick and meaningful solutions to challenges, relevant content or offers, or other benefits. These rewards can be short-lived, one-off, or impossible to replicate, however, if the logic behind them is impossible to decipher.
In terms of artificial intelligence, this is often referred to as “explainability,” and it refers to the ability for humans to decipher the reasoning and logic behind a decision made by AI. The more difficult it is to determine why a decision has been made, it can pose challenges in replicating success, and to allow humans to learn what they can do better in related areas. In the worst case, machine learning can sometimes take the wrong lessons from customer experiences and unwanted biases can get introduced that can be hard to detect.
Despite these challenges, there are many opportunities for AI to play a key role in the customer experience. The key is to strike the right balance between more advanced analytics and narrowly targeted experiences and simpler to explain approaches to gain buy in and broader organizational adoption. Don’t let the perfect be the enemy of the good! This is particularly true when you need to consider changes to incentives or other aspects of performance management. Next, we’re going to talk more about how the two can work together for CX success.
In the third and final part of this series, we’re going to talk about how to bring it all together and best integrate artificial intelligence with your customer experience program.