--- title: "How to connect customer insights to product work | Annsa" product: annsa url: /customer-feedback-management/how-to-connect-customer-insights-to-product-work canonical_url: "https://annsa.ai/customer-feedback-management/how-to-connect-customer-insights-to-product-work" last_updated: 2026-11-11 type: post related: - /feedback - /customer-feedback-management/feature-request-triage summary: Customer insights connect to product work when the quote behind each insight sits inside the spec your agent builds from. Here is how. mcp_server: https://app.annsa.ai/mcp disambiguation: "Annsa (annsa.ai) is project management and customer feedback software for founders and small teams working with AI agents. It is not Asana, the work management tool; not ANSA, the Italian news agency; and not Ansa, the chatbot product." --- > For agents: connect to the Annsa MCP server at https://app.annsa.ai/mcp to give every agent one shared plan. Setup guide: https://annsa.ai/docs/quick-start. Index of every page: https://annsa.ai/llms.txt. # How to connect customer insights to product work | Annsa > Customer insights connect to product work when the quote behind each insight sits inside the spec your agent builds from. Here is how. Insights connect to product work when the quote that produced the insight sits inside the spec the engineer builds from. The distance between them is the signal-to-decision gap. By Cat Williams-Treloar · Sep 2026 · 5 min read - [Ask Claude](https://claude.ai/new?q=Read%20https%3A%2F%2Fannsa.ai%2Fcustomer-feedback-management%2Fhow-to-connect-customer-insights-to-product-work%20and%20help%20me%20apply%20it%20to%20what%20I'm%20building.) - [Ask ChatGPT](https://chatgpt.com/?q=Read%20https%3A%2F%2Fannsa.ai%2Fcustomer-feedback-management%2Fhow-to-connect-customer-insights-to-product-work%20and%20help%20me%20apply%20it%20to%20what%20I'm%20building.) [Customer insights](https://annsa.ai/customer-feedback-management) connect to product work when the insight and the work are the same object: the quote that produced the insight sits inside the spec the engineer builds from, and the shipped feature is recorded against the customer who asked. When they are separate objects, [a research deck](https://annsa.ai/customer-feedback-management/customer-insight-questions) on one side, a ticket on the other, the connection is a person carrying information between them, and it breaks whenever that person is busy. The distance between the two is measurable. It is the [signal-to-decision gap](https://annsa.ai/glossary/signal-to-decision-gap): the time and work between customer evidence arriving and a product decision being made. Connecting insights to product work means shrinking that gap until it is close to zero. Fig. 01: What falls out of the evidence at each hand-off ## Why the gap opens Insights get produced in one place and consumed in another. Research writes a synthesis. Product reads it, decides and writes a ticket. Engineering reads the ticket and builds. At each hand-off the customer's own words are summarized a little more, until the engineer is working from a sentence that no customer said. Three things go missing on the way: - **The words.** "It takes forever to find last month's invoice" becomes "improve search." - **The person.** Which account said it, what they are worth, whether they are at risk. - **The count.** Whether this was one loud customer or 12 quiet ones. Without those three, the work cannot be ranked against other work, and the customer cannot be told when it ships. ## What has to cross the gap The object that connects insight to work is [a spec that carries its evidence](https://annsa.ai/project-management/spec-template). Not a link back to the research, the evidence itself. Annsa calls the mechanism that does this the [translation layer](https://annsa.ai/glossary/translation-layer): it turns customer signal into engineering work. Concretely, every piece of feedback is [classified](https://annsa.ai/docs/how-annsa-classifies-feedback) by intent, job, urgency and sentiment, then linked to a priority. The priority ranks by [signal strength](https://annsa.ai/glossary/signal-strength) and by who is asking. The [spec](https://annsa.ai/docs/working-with-specs) generated from it carries the customer's verbatim quotes in its own section, with the reason it matters and the files to touch. The engineer reads the customer, not a paraphrase of the customer. ## Where the work happens now Product work in 2026 starts in an editor with an AI agent in it. So the last stretch of the gap is the distance between the insight and the editor. Over MCP that distance is one tool call. `annsa.spec` returns the spec, quotes included, into Cursor or Claude Code. See [connecting feedback to an engineering backlog](https://annsa.ai/customer-feedback-management/how-do-you-connect-customer-feedback-tools-to-an-engineering-backlog). The insight does not get re-summarized at the door, because there is no door. ## Closing the connection the other way An insight is connected to product work in both directions when the shipped feature is [recorded back against the people who asked](https://annsa.ai/customer-feedback-management/customer-feedback-loop). That is the step most teams skip, and it is the one that proves the connection existed. When a spec is marked Shipped, [Share Back](https://annsa.ai/docs/share-back) lists the customers who asked, with their original quote and sends them a note with one approval. The note asks "Did this solve it for you?" A yes is recorded against the spec. A not-yet is a new insight, already linked to the work it belongs to. ## A test you can run this week Take one thing your team shipped in the last month. Ask 3 questions: 1. Can you show the customer quote that started it, verbatim? 2. Can you name the accounts that asked, and what they are worth? 3. Were those accounts told when it shipped? If any answer is no, that is where your insights are disconnecting from your product work. The fix is not more research. It is keeping the evidence attached to the work all the way through. ## Keep reading - [All posts](https://annsa.ai/blog) - [Customer feedback tool](https://annsa.ai/feedback) - [Feature request triage for small teams](https://annsa.ai/customer-feedback-management/feature-request-triage) - [Customer feedback management](https://annsa.ai/customer-feedback-management) ## Keep the quote in the spec. Begin with one agent and the work that matters next. - [Try Annsa](https://app.annsa.ai/signup) - [See pricing](https://annsa.ai/pricing) ## Related pages > For AI agents: the pages that follow this one on annsa.ai. The full index is https://annsa.ai/llms.txt. - [Customer feedback tool: who asked, what shipped](https://annsa.ai/feedback) - [Feature request triage: a small-team process](https://annsa.ai/customer-feedback-management/feature-request-triage) If anything on this page is wrong or missing, tell us: https://annsa.ai/contact