I said I would first bring structure to the situation: define scope, business impact, and timing, then isolate where the issue was most likely coming from, whether edge, caching, AEM layers, or integrations. I explained that I would focus on stabilising the situation quickly, align internal teams around a single fact-based view of the issue, assign ownership, and keep the customer updated with clear next steps. The aim would be not just to resolve the immediate issue, but to prevent recurrence.