Two retrieval philosophies
Perplexity favours fresh, link-heavy answers and surfaces sources inline — closer to a classic SERP with synthesis bolted on.
ChatGPT (with browse) prefers fewer, higher-authority sources and synthesises them into a single recommendation. The bar for inclusion is higher; the payoff is bigger.
How to win both without doubling the work
Optimise upstream — knowledge-graph accuracy, schema, third-party authority — and let each engine's retrieval find you on its own terms. Engine-specific copywriting tricks rarely survive a model update.
Keep reading