There’s a lot of excitement around AI everywhere right now.
The rise of Large Language Models (LLMs) is a powerful moment in tech history, with incredible potential when applied to the right business contexts. But if business is to gain the full advantage of generative AI, it must maintain a clear focus on measurable outcomes.
The contact centre is a perfect example of an opportunity to enhance customer journeys by bringing together real-time, multimodal interactions across channels to unlock new insights and opportunities for better staff and customer experiences. However, making AI a truly powerful integration with contact centre systems takes a lot more than a shiny new plug-in.
If we want a successful integration of AI/gen AI technology, we must overcome the excitement of the new, and maintain a focus on the objective we are trying to meet to make the investment deliver value.
Does generative AI hold this potential for measurable improvement in contact centre objectives? Absolutely. However, it’s critical that robust design processes are put in place to ensure they meet appropriate targets and measurable objectives.
Success exists in measurement: Streamlining Call Trees, Boosting self-service and enabling call centre teams to spend more time resolving complex calls.
In one example outlined in our new ebook on generative AI in the contact centre, we have supported a major retailer in its transformation of an outdated on-prem contact centre platform to embrace Google Cloud CCAI. The design redefined interactive voice response (IVR) call flows into a voice experience that increased self-service by ~40%, cutting average hold times by 20% while reducing total cost of ownership by 40%.
And locally, we worked with a leading Australian telecommunications company on its own contact centre transformation, touching voice, chatbots, and a mobile app. We have seen powerful results in boosting customer satisfaction scores, reducing friction in access to support services across voice and web channels. In turn, these business improvements create space for human support agents to focus on the higher value needs of customers instead of answering the most basic and repetitive questions that are now quickly resolved through AI supported channels.
These successful outcomes are built on more than general promises of “improved experience” or “enhanced intelligence”. By pursuing a careful, human-centred design process where Cognizant brings together multiskilled project teams, we worked with our clients to develop integrated solutions where AI is part of a wider transformation in contact centre operations.
Successfull implementations though need to blend the human centred design to be anchored with the metrics you already use for key performance indicators in your contact centre. Introducing and integrating generative AI should have a business outcome focus. Whether that is
- CSAT / NPS
- Average hold times
- First Contact Resolution
- Containment rate
Enterprise AI needs an enterprise approach
The explosion in the use of tools like ChatGPT and Google Gemini have been driven by end-user adoption, with humans in the loop. It has been a highly experimental approach, with users exploring what’s possible and benefitting from years of research and enormous computing power to bring about new ways of quickly integrating concepts to move the boundaries of creativity.
While getting teams/staff used to these new tools through experimentation and exploration is an important part of bringing these capabilities into the business environment, it is not the optimal path for achieving and sustaining business value. True adoption within your enterprise processes will require a clear understanding of your business objectives and addressing key concerns. These include grounding models to minimise hallucinations and feeding language models with the most relevant enterprise context.
Working with the right technology partner to bring a multidisciplinary approach to design and deliver trusted, risk-managed, customised AI experiences at the heart of your contact centre is one example of a more direct path to a successful AI future which paradoxically may seem like a longer journey in the rush to leverage the opportunity GenAI provides.
