# AI products stop feeling smart when they hide their context > Why MCP wedges, narrow workflows, richer product context, trace reviews, and visible uncertainty make AI products easier to trust. - Canonical HTML: https://growth.iangoh.com/blog/ai-products-stop-feeling-smart-when-they-hide-their-context/ - Published: 2026-05-26 - Updated: 2026-05-26T09:10:00Z - Categories: AI products, product-led growth, brand trust - Niches: AI products, SaaS, developer tools, B2B software, creator tools ## On this page - The first wedge can be smaller than the agent - Breadth is a bad first promise - The agent should know where the user is standing - Trust grows faster when the team watches reality - Visible uncertainty is part of the brand - Where this cluster is most useful ## Start with these related tactics - [MCP server before custom agent](/growth-ideas/mcp-server-before-custom-agent/): Expose the product through an MCP server first so developers can use it in their own agent workflows before you invest in a full in-app agent. - [Workflow-first AI demand validation](/growth-ideas/workflow-first-ai-demand-validation/): Start with one narrow AI workflow that solves a repeated job, then expand into a broader agent only after users pull for adjacent use cases. - [Layered context injection for AI answers](/growth-ideas/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. A lot of AI product pages still sell intelligence as if intelligence were the product. The demo writes something fluent, clicks a few buttons, and makes the room feel briefly futuristic. Then a real user arrives with a messy account, a vague question, and a timeline that does not forgive confusion. That is where many AI products stop feeling smart. The problem is often not the model. It is the missing evidence around the model. What does it know about this user, this workflow, this page, this account, and this job right now? If the product hides that context, the AI starts to look like a stranger wearing the company logo. ## The first wedge can be smaller than the agent That is why [MCP server before custom agent](/growth-ideas/mcp-server-before-custom-agent/) is such a useful move. PostHog makes the practical point: technical users may not need your polished assistant first. They may just need your product to show up inside the workflows they already run. I like this because it is honest. It lets a team learn whether people want agentic access before it spends months pretending a broad assistant is already justified. ## Breadth is a bad first promise The companion move is [workflow-first AI demand validation](/growth-ideas/workflow-first-ai-demand-validation/). PostHog started with a narrower data-question workflow before expanding the broader agent. That sequence matters. A small workflow gives the user one repeatable reason to come back. It also gives the team one place to study where the product is saving time and where it is simply producing clever-looking noise. ## The agent should know where the user is standing The hardest useful lesson in this batch is [layered context injection for AI answers](/growth-ideas/layered-context-injection-for-ai-answers/). A user inside a dashboard is not asking from nowhere. They are asking from a page, a role, a schema, a timezone, a plan, and a recent trail of actions. That is what separates a product-native AI experience from a generic chat tab. The answer should feel like it came from inside the software, not from a polite outsider trying to guess what your app is. ## Trust grows faster when the team watches reality I would pair that with [weekly traces hour for agent quality](/growth-ideas/weekly-traces-hour-for-agent-quality/). A lot of teams talk about evals as if evals are enough. They are not. Someone still has to look at the ugly sessions and the surprisingly good ones. That review habit is useful beyond AI products. It is the same discipline behind good support, good onboarding, and good sales calls. Watch the real interaction. Then make the next system a little less naive. ## Visible uncertainty is part of the brand The most human lesson here may be [uncertainty, source, and progress cues in AI UI](/growth-ideas/uncertainty-source-and-progress-cues-in-ai-ui/). AI trust does not come from a warmer mascot or a larger hero claim. It comes from whether the user can tell what the system is doing and whether that effort deserves belief. If the answer is slow, show progress. If the answer rests on a source, show the source. If the system is unsure, let it admit that early. The product looks more serious when it risks sounding careful. ## Where this cluster is most useful This cluster is strongest for AI products, SaaS tools, and developer platforms where the company is deciding whether to bolt on an assistant or make AI part of the core workflow. It also maps well to creator tools and internal software where users are tolerant of new interfaces but impatient with fake certainty. If an AI feature still feels flimsy, I would not ask first whether the model is strong enough. I would ask how much of the real job the product is willing to show. If you want a second set of eyes on the AI surface, [Ian Goh advisory](https://iangoh.com/advisory) is the clearest next step. ## Related GrowthDex tactics - [MCP server before custom agent](/growth-ideas/mcp-server-before-custom-agent/) - AI Search, Developer Tools, Product - [Workflow-first AI demand validation](/growth-ideas/workflow-first-ai-demand-validation/) - Product, Activation, Retention - [Layered context injection for AI answers](/growth-ideas/layered-context-injection-for-ai-answers/) - Product, Onboarding, AI Search - [Weekly traces hour for agent quality](/growth-ideas/weekly-traces-hour-for-agent-quality/) - Product, Retention, Support - [Uncertainty, source, and progress cues in AI UI](/growth-ideas/uncertainty-source-and-progress-cues-in-ai-ui/) - Product, Conversion, Retention ## Essay chronology - [Newer essay: The launch page should answer the second question](/blog/the-launch-page-should-answer-the-second-question/) - Product Hunt, launches, product marketing - [Older essay: The first growth system usually looks manual](/blog/the-first-growth-system-usually-looks-manual/) - founder-led growth, brand trust, early-stage growth ## Keep reading - [The feedback loop breaks when the middle stays hidden](/blog/the-feedback-loop-breaks-when-the-middle-stays-hidden/) - product-led growth, community-led growth, brand trust - [The Teams app should meet the work before the help doc](/blog/the-teams-app-should-meet-the-work-before-the-help-doc/) - onboarding, product-led growth, brand trust - [The Slack app should start helping before the docs tab opens](/blog/the-slack-app-should-start-helping-before-the-docs-tab-opens/) - onboarding, product-led growth, brand trust ## Continue through the blog - [SaaS](/blog/#path-saas) - 3 essays in this path - [AI products](/blog/#path-ai-products) - 3 essays in this path - [developer tools](/blog/#path-developer-tools) - 3 essays in this path ## Sources - [PostHog Newsletter](https://newsletter.posthog.com/p/what-we-wish-we-knew-before-building) · [GrowthDex source hub](/sources/posthog-newsletter-newsletter-posthog-com/) - [PostHog Newsletter](https://newsletter.posthog.com/p/what-weve-learned-about-building) · [GrowthDex source hub](/sources/posthog-newsletter-newsletter-posthog-com/) - [PostHog Newsletter](https://newsletter.posthog.com/p/the-golden-rules-of-agent-first-product) · [GrowthDex source hub](/sources/posthog-newsletter-newsletter-posthog-com/) ## Editing notes - Kept the essay on one claim about context and trust instead of turning it into a grand theory of AI transformation. - Used plain scenes like messy accounts, vague questions, and slow answers so the argument feels observed rather than promoted. - Let the PostHog examples do the proof work instead of padding the article with general claims about the future of agents. - Closed on one diagnostic question about how much of the real job the product shows rather than a generic AI conclusion. ## Advisory If you want help turning this into a growth system, Ian Goh offers advisory at https://iangoh.com/advisory.