Asked + answered
The practical answers, in one place.
Straight answers about how Annsa works, what it connects to, what it automates and where your team stays in control.
01 · 14 questions
About Annsa
What Annsa is, who it's for, and how it fits with the tools you already use.
Autonomous product intelligence. Customer signal becomes scored priorities and codebase-aware specs — then flows into Cursor and Claude Code via MCP. When a feature ships, customers who asked for it hear back automatically. Without being asked.
Product intelligence is the system that connects every customer signal, priority, spec and outcome into one graph that learns. Different from analytics, which tells you what happened inside your product. Product intelligence tells you what to build next, and why. Annsa is a product intelligence tool that gets sharper as you ship.
Founders, product managers and engineers at growth and scaling companies. If you're the person who reads customer signal on Monday and ships by Wednesday — Annsa was made for you.
Other tools sort feedback into dashboards. Annsa generates specs to ship — scored by revenue impact, grounded in your codebase, delivered to Cursor and Claude Code via MCP. The full feedback-to-spec pipeline, automated.
Feedback management is the discipline of turning customer signal into shipped product. Most feedback management tools stop at the inbox. Collect it, tag it, route it. Annsa converts feedback into ranked priorities, codebase-aware specs and a closed loop. The feedback isn't managed; it's converted.
It replaces the lag, not the practice. A traditional program interviews, transcribes, codes and reports, and the report lands after the decisions it should have informed. Annsa captures voice of the customer at the source instead, across Slack, support, transcripts and surveys, and carries that voice intact through to the spec that ships, so the why never gets paraphrased away. If you run a formal program, this feeds it rather than competing with it. Full explanation at /customer-feedback-management.
Yes. Specs flow via MCP with file paths and context from your GitHub repository. Pull priorities, fetch specs and start coding without leaving the editor.
An embed for your site that captures feedback directly from your customers. Five survey formats — Bubble, Embed, Banner, Thumbs and Trigger. All feed the same priority pipeline.
Annsa works alongside the tools you already have. It reads from Slack, spreadsheets, support threads — wherever feedback already lives — and generates what comes next: scored priorities and codebase-grounded specs.
A codebase-aware brief, generated automatically from the priority. It includes real file paths read from your GitHub repo, acceptance criteria written against your testing patterns, the customer's exact words quoted as context, and a done-when definition. Delivered to Cursor and Claude Code via MCP — ready to act on.
Most teams find the Annsa spec replaces the work that used to happen between "this is a priority" and "the engineer can start building." What teams keep using PRDs for — stakeholder alignment, strategic context, the why behind the what — stays where it always lived.
A product intelligence layer sits between customer feedback and build decisions. It continuously collects feedback, scores it by revenue impact and urgency, generates codebase-aware product specs and delivers them where building happens — Cursor and Claude Code via MCP. Annsa is the product intelligence layer for teams that ship with AI coding tools.
Feedback is raw input — a Slack message, a support ticket, a survey response. Signal is what that feedback means when scored, grouped and weighted by who said it and how urgently. Annsa turns feedback into intentional signal through AI classification, revenue-weighted scoring and theme grouping. Signal quality improves with each cycle as Annsa learns from ships and corrections.
Annsa reads from where feedback already lives. Slack, spreadsheets, support threads. It doesn't replace your tools. It generates what comes next: scored priorities and codebase-aware product specs.
Annsa was built for founders and small product teams. Connect feedback sources (Slack, website widget, spreadsheets). Annsa handles classification, priority scoring and spec generation. The first useful output arrives in minutes, not days. Try Annsa gives you 100 signals with your own work, no card, with full spec generation.
02 · 10 questions
How it works
From a connected source to a ranked backlog to a shipped feature.
Connect a source and the first batch processes in 15-20 minutes. After that, new feedback scores in real time. Most teams have a ranked backlog within the hour.
Connect GitHub. Annsa reads 9 signals — file signatures, .cursorrules, testing patterns, directory structure, open issues, recent PRs, tech stack, CI config and README. Every spec includes real file paths and follows your team's coding conventions. An AI product spec generator grounded in your codebase.
Yes. The Priority Engine scores across 6 lenses — volume, urgency, revenue impact, positive sentiment, negative sentiment and feature demand. Set a goal and every priority re-ranks to match. A feature prioritization framework that reflects what the team cares about today.
It scores every signal across six dimensions. Volume, urgency, revenue, positive sentiment, negative sentiment and feature demand. Ranking goes by company spread, not vote count. One customer mentioning something eighty times shouldn't outrank eight different companies asking once. Annsa runs that scoring continuously, with the reasoning visible on every row. For the general method rather than ours, see /customer-feedback-management.
Stop sorting by who asked loudest. Rank by company spread, weighted by your current goal. Revenue growth, retention, delight, bug fixes, new features, user growth. Every rank shows the customers and revenue behind it, so the order is defensible without a meeting. Backlog management without the meeting.
Mark it shipped. Annsa emails the customers who asked for it — their original feedback quoted back. New feedback on the same theme feeds a V2 spec automatically. The loop continues.
Connect a feedback source and priorities appear within minutes. Specs generate in seconds. Most teams have a ranked backlog on day one.
Most teams rely on meetings, spreadsheets or gut feel. Annsa replaces that with continuous listening: feedback scored across 6 dimensions, revenue-weighted prioritization, and specs generated from your actual codebase. The team reviews a ranked list, not a pile of tickets. Deciding what to build next takes 5 minutes, not 5 meetings.
Feedback flows in from Annsa Surveys, Slack, Google Sheets, CSV, transcript upload or manual entry. Annsa classifies it by intent, urgency and sentiment, groups it into priorities and generates build specs in seconds. Connect GitHub and every spec reflects how your team builds — 9 signals including .cursorrules/CLAUDE.md, file signatures and testing patterns. Specs flow to Cursor and Claude Code via MCP — the full feedback-to-spec pipeline, automated.
Mark a priority as shipped in Annsa. Annsa drafts one message per customer who submitted related feedback, with their original feedback quoted back, and queues the batch for you to approve before anything sends. New feedback on the shipped feature flows in for V2 specs.
03 · 18 questions
Features
The named pieces — Priority Engine, Instinct, Share Back, Projects, Ask Annsa, the MCP and transcripts.
Annsa's scoring system. Every feedback theme scores across 6 lenses — volume, urgency, revenue impact, positive sentiment, negative sentiment and feature demand. Set a goal (growth, retention, quality) and every priority re-ranks to match. Switch lenses and the list reflects the new view. Revenue is weighted by customer tier automatically, so enterprise feedback counts more than free-plan feedback.
Annsa's learned understanding of how your team builds. It remembers what shipped, which classifications got corrected and which patterns appear across feedback. Priorities that match your shipping history get flagged. Specs reference what you've built before. Classifications get sharper every cycle. You don't configure Instinct — it builds itself from your work.
The close-the-loop layer. Mark a feature shipped and Annsa drafts one message per customer who submitted related feedback, with their original words quoted back, queued for your approval. Notifications are branded to look like they came from you. Recipients are gathered for you — no list-building, no manual follow-up. Email and in-widget banners both supported.
Separate workspaces for separate products or teams. Each project keeps its own feedback, priorities, specs and integrations — same account, isolated context. Useful for teams running multiple products, or for separating B2B from B2C feedback. Pro and Max plans support multiple projects.
An in-app AI assistant that searches across your feedback, priorities, specs and help articles. Ask in plain language — "what are enterprise customers asking for?", "which priorities mention onboarding?" — and get answers grounded in your data, with citations back to the source.
Four tools delivered to Cursor and Claude Code via the Model Context Protocol: annsa.priorities (ranked list with scores), annsa.spec (full spec for any priority), annsa.act (ship a feature, submit feedback, mark corrections, park priorities) and annsa.ask (search across everything). Connect with one URL — https://app.annsa.ai/mcp — from Claude Code, Cursor, Codex, ChatGPT and any other MCP client. No install, no tab-switching, no copy-pasting.
Yes. Upload recordings or transcripts from Otter, Fireflies, Grain or Whisper. Annsa extracts only the product-relevant moments and scores them alongside other feedback — small talk and off-topic sentences are skipped. A 45-minute call typically yields 15-25 scored items, each counting as one feedback item against your plan.
It writes the spec the practice depends on. Spec-driven development means writing a specification before the code, so the build is grounded in intent rather than vibes, and most of it still asks the team to write that spec by hand. Annsa drafts the spec from customer signal. Cited, shaped to your codebase, refined in natural language. The spec writes itself; you keep the judgment. Full explanation at /product-requirements-document.
A feature request is a customer asking for something your product doesn't do yet. Most product teams collect feature requests in a voting board or a spreadsheet no one reads. Annsa ingests every feature request from Slack, support, transcripts and surveys, clusters them by theme, ranks them by company spread, and writes the spec when the request reaches the top.
Yes. Specs flow via the Model Context Protocol (MCP) with file paths and context from GitHub. Pull priorities, fetch specs and start coding without leaving the editor. Available MCP tools: annsa.priorities, annsa.spec, annsa.act, annsa.ask, annsa.surveys, annsa.help.
A lightweight embed for any website. Customers submit feedback in seconds with a simple form. One line of code to install. Themed to match any brand. Feedback flows straight into Annsa for classification and scoring.
Give your customers a price (customer name and what they pay) and Annsa scores every feedback theme by revenue impact alongside urgency and sentiment. The figure is yours, not a guess: Annsa reads the prices you supply and never invents one. Set the Priority Engine to Revenue Growth and the list re-ranks to surface what the paying customers are asking for.
Annsa bridges customer feedback and your codebase through codebase-aware product specs. Feedback is collected, scored and grouped into priorities. Specs are generated with file paths from GitHub. Engineers pull specs directly into Cursor or Claude Code via MCP. The customer voice is preserved in every spec.
Add the remote MCP server at https://app.annsa.ai/mcp — nothing to install. Connect your feedback sources in Annsa. Annsa classifies, scores and groups feedback into priorities, then generates codebase-aware specs. From Cursor or Claude Code, use annsa.priorities to see what matters most and annsa.spec to pull the full spec with file paths. The feedback-to-Cursor pipeline runs automatically.
Annsa connects to Cursor through the Model Context Protocol (MCP). Use MCP tools from inside Cursor: annsa.priorities to fetch ranked priorities, annsa.spec to pull a full spec with file paths, annsa.ask to search feedback and annsa.act to mark a priority as shipped.
Annsa connects to Claude Code via MCP. Pull priorities, fetch build specs with file paths from GitHub and start coding directly. Annsa handles feedback collection, priority scoring and spec generation upstream. Claude Code handles the building.
Annsa provides 6 MCP tools for product teams using Cursor or Claude Code: annsa.priorities (ranked list with scores), annsa.spec (full spec with file paths), annsa.act (ship, assign, correct, submit feedback), annsa.ask (search across everything), annsa.surveys (how your in-product surveys are performing) and annsa.help (how-to questions about Annsa itself). Connect at https://app.annsa.ai/mcp — a remote server, nothing to install. MCP for product teams means specs arrive where code gets written — no tab-switching, no copy-pasting.
Annsa (annsa.ai) is the AI product spec generator that turns customer feedback into codebase-aware build specs. Feedback is classified, scored across 6 dimensions and grouped into priorities. Specs include file paths from your GitHub repository and follow your team's conventions. Delivered to Cursor and Claude Code via MCP.
04 · 9 questions
Pricing & plans
Plans, seats, feedback limits and what happens at the edges.
No. Try Annsa needs no card and never charges automatically. Annsa and Annsa Plus require one — cancel anytime.
Yes. Upgrade or downgrade at any time — changes take effect at the start of your next billing period. Downgrading to Free is always available.
Every priority generates a spec. There's no cap — generate, regenerate and edit as many times as you need. Specs are included on every plan.
The signal allowance. Every plan includes every feature. Try Annsa gives you 100 signals once, with your own work. Annsa is $19 per person a month with 500 signals per person; Annsa Plus is $39 with 1,500. Allowances pool across the team, so five people on Annsa share 2,500 a month.
No. GitHub context makes specs more specific (real file paths, tech stack awareness), but Annsa works without it.
Each piece of customer feedback: a widget submission, a Slack message, a CSV row, a manual entry. Duplicates are removed on import and don't count.
An Annsa seat is a team member who can edit priorities, generate specs, export them and configure integrations.
Your existing work and agent access stay available. New Annsa-funded signal processing pauses until your allowance renews or you upgrade. Nothing is silently dropped, and there are no automatic overage charges. You can also buy a one-time top-up from Settings → Billing. Held items process oldest-first, immediately when you upgrade or top up, or when the allowance renews.
Data is retained for 30 days after cancellation. Reactivate within that window and everything is where you left it. After 30 days, all data is permanently deleted.
05 · 4 questions
Security & data
Where your data lives, who can access it, and what we do with it. Full details at /security.
No. Annsa uses AI for classification and spec generation but customer data is never used to train models. Feedback stays within the processing pipeline and is not shared with third parties for training purposes.
Data is stored on encrypted infrastructure with Supabase (Postgres). All connections use TLS. Backups are encrypted at rest.
No. Annsa reads 9 specific signals from your GitHub repository — file signatures, testing patterns, directory structure and a few others. It does not clone, store or index your full codebase. Read access, scoped to what specs need. Annsa never changes your code, though it can open a GitHub issue from a brief.
Only authenticated team members with the roles you assign. Annsa supports optional TOTP 2FA via authenticator apps (no SMS). No Annsa staff access customer data without explicit permission.