/serp/ - SERP Analysis

lowkey the focus on referral traffic is the right move because citations are too easily gamed by high-volume, low-intent content. tracking direct clicks from llm interfaces tells u if ur content actually answers a query well enough to trigger an action. ive started using a custom python script to scrape search results for specific brand queries and then manually check if the generative summary includes our key product features. it helps identify if we are being omitted by design due to lack of structured data. one thing that is still tricky is correlating these clicks back to specific lead sources in our CRM. do u have a preferred way to attribute those sesisons when they come through via generic ai search proxies?