Layered context injection for AI answers
Feed the model the user's current page state, schema, and account context before it answers so the AI can act like part of the product instead of a detached chatbot.
Why this can grow a startup
Users do not experience your product as a blank prompt. They arrive from a page, a role, a dataset, and a business context. When the agent receives that context up front, it can produce answers that fit the real task instead of generic best guesses. This lowers the amount of clarification the user has to provide, makes outputs feel more native to the product, and gives the AI a real advantage over a general model tab.
Company example
PostHog says Max AI receives current dashboard state, visible insights, filters, role, schema details, account tier, timezone, and retention context so requests like why signups dropped last week can be answered inside the product context instead of from a blank chat.
Source and metric
Source: PostHog Newsletter · Browse PostHog Newsletter tactics
PostHog describes page state, schema, role, organization tier, timezone, and retention as standard context sent with Max AI requests.
Source discovered: May 26, 2026
When to use it
Use this when Product, Onboarding, AI Search is relevant to ai products, activation, ux and you can run a bounded test with a medium budget.
When not to use it
Do not use it as a substitute for customer evidence, a clear owner, or a measurable stop condition. Local platform rules and market behavior still need checking.
Founder checklist
- Read PostHog Newsletter and identify what is directly supported.
- Choose one channel context: Product, Onboarding, AI Search.
- Define the test around PostHog describes page state, schema, role, organization tier, timezone, and retention as standard context sent with Max AI requests..
- Set an owner, evidence window, and stop condition before launch.
Explore the context
Apply this with an operator
Turn isolated search tactics into a crawlable visibility system tied to demand and proof.