AI is increasingly going to answer more and more complex questions. The types of questions that technical support organizations field from customers. It is not clear exactly what percentage of questions AI will be able to answer and how soon, but there is no doubt that it is a non-zero number. Every manager who runs a support organization and every technical support person should think about the ramifications of this development. We should ask ourselves why we need a technical support organization if AI can answer all the technical questions.
AI’s ability to answer questions depends on the quality and completeness of the data it has access to. Quality data specific to product technical support does not just appear for free, it needs to be created. You could argue that product documentation should fill this role. But we have product documentation today, and we still need technical support to stitch together disparate sources and translate them into a real-world experience. Product documentation doesn’t usually cover everything that can go wrong and cannot cover novel combinations that emerge over time during a product lifecycle. Someone will need to create the content to feed AI to answer these questions. While this is a shift in focus for technical support organizations, the existing skills of troubleshooting, understanding customers, and clear communication are still exactly what is required to feed AI the right data.
Today technical support organizations monitor the overall quality of the support experience. This business requirement is no different when AI delivers the solutions. I expect that in the same way the DevOps methodology transformed sysadmin work, technical support engineers will shift into the role of orchestrators. The technical support experience will resemble a cloud service will real-time monitoring and tuning.
Even with AI, there is no such thing as a free lunch. Support organizations will need to allocate budget to pay for AI services. This might mean using different cost AI models for different solutions. Organizations will need a business continuity plan for when an AI service goes down or produces undesirable results.
AI is going to change what support looks like both externally for customers and also internally how support organizations run. The change may come quickly, or it may take longer. Support leaders will need to live in a hybrid world and match the dynamic rate of change to ride on the wave of innovation.