[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_231":232,"_233":234},"routes/_app.x.$handle.$name",{"_96":95,"_32":44,"_105":86,"_27":106,"_13":14,"_107":-5,"_11":108,"_109":110,"_111":112,"_113":114,"_115":116,"_20":-5,"_117":-5,"_118":-5,"_119":25,"_120":121,"_122":123,"_124":-5,"_125":-5,"_126":127},"routes/_app.x.$handle.$name.branch.$branch.version.$version.diff",{"_11":12,"_13":14,"_15":16,"_17":-5,"_18":-5},"project",{"_42":43,"_32":44,"_37":45,"_46":-5,"_47":25,"_48":49,"_50":49,"_51":52,"_53":-5,"_18":54,"_55":-5,"_56":25,"_57":-5,"_58":-5,"_59":60,"_61":62,"_63":64,"_65":66,"_67":68,"_69":-5,"_70":71,"_72":71,"_73":74,"_75":74,"_76":74,"_77":71,"_78":71,"_79":80,"_81":82,"_83":84,"_85":25},"version",21,"changes",[19],"aiChatInfo","userId",{"_20":21,"_22":23,"_24":25,"_26":25,"_27":28,"_29":28},"file",{"_30":31,"_32":28,"_33":20,"_34":36,"_37":38,"_39":14,"_40":41,"_24":25},"previous",{"_30":31,"_32":28,"_33":20,"_34":35},"deleted",false,"deletedHere","path","README.md","latestPath","fileId","88f3b05c-fa75-11f0-831a-42dde27851f2","name","fileType","content","# PMF Survey\n\nMeasure and optimize your product-market fit using the Superhuman framework from [First Round Review](https://review.firstround.com/how-superhuman-built-an-engine-to-find-product-market-fit/).\n\n![image.png](https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/9ff5b55c-a166-4f47-759c-6726b7477800/public)\n\n## What this does\n\n1. Collect survey responses with the 4 proven PMF questions\n2. Calculate your PMF Score (benchmark: 40%)\n3. AI analyzes responses to detect personas and themes (optional)\n4. Generate actionable insights:\n - High-Expectation Customer (HXC) profile\n - What users love (from \"very disappointed\" responses)\n - What holds users back (from \"somewhat disappointed\" responses)\n - Roadmap recommendations (50% double down / 50% address gaps)\n\n## Getting started\n\n1. Click [**Remix**](/?intent=remix)\n2. Customize `PRODUCT_NAME` in [`main.ts`](./main.ts) line 9\n3. Customize [`PROMPT.txt`](./PROMPT.txt) with your product context\n4. Share `/survey` with your users\n5. Watch your PMF score update at `/dashboard`\n\n**Optional:** Add `OPENAI_API_KEY` for AI-powered persona detection and HXC profile generation. The survey works without it — AI just enhances the insights.\n\n## The Framework\n\nThe Superhuman PMF engine uses Sean Ellis's leading indicator: **% of users who would be \"very disappointed\"** without your product.\n\n**Benchmark:** Companies with strong traction exceed **40%**. Companies struggling to grow are below 40%.\n\n### The 4 Survey Questions\n\n1. **How would you feel if you could no longer use [product]?**\n - Very disappointed\n - Somewhat disappointed\n - Not disappointed\n\n2. **What type of people do you think would most benefit from [product]?**\n\n3. **What is the main benefit you receive from [product]?**\n\n4. **How can we improve [product] for you?**\n\n## Features\n\n### Works Without AI\n- Survey collection\n- PMF Score calculation\n- Response breakdown (very/somewhat/not)\n- Basic persona detection (keyword-based)\n- Basic theme extraction\n\n### Enhanced With AI (add `OPENAI_API_KEY`)\n- Rich persona detection\n- Detailed HXC profile generation\n- Smart roadmap recommendations\n- Nuanced sentiment analysis\n\nAfter adding your API key, POST to `/reprocess` to analyze existing responses.\n\n## How it works\n\n**Survey**\n- Users complete the 4-question survey\n- Responses are stored immediately\n\n**Analysis**\n- Basic persona/theme extraction runs automatically\n- If `OPENAI_API_KEY` is set, AI enhances the analysis\n\n**Dashboard**\n- PMF Score: % of \"very disappointed\" responses\n- Segmented PMF: Score filtered to ideal personas\n- Response breakdown with visual bars\n- HXC Profile (AI-enhanced)\n- Top benefits and improvements\n- 50/50 roadmap recommendations\n\n## Key Insights from the Framework\n\n> \"If you only double down on what users love, your product-market fit score won't increase. If you only address what holds users back, your competition will likely overtake you.\"\n\nThe framework teaches you to:\n- **Ignore** \"not disappointed\" users — they won't convert\n- **Focus on** \"very disappointed\" users to understand what to double down on\n- **Convert** \"somewhat disappointed\" users who already value your main benefit\n","# PMF Survey\n\nMeasure and optimize your product-market fit using the Superhuman framework from\n[First Round Review](https://review.firstround.com/how-superhuman-built-an-engine-to-find-product-market-fit/).\n\n## What this does\n\n1. Collect survey responses with the 4 proven PMF questions\n2. Calculate your PMF Score (benchmark: 40%)\n3. AI analyzes responses to detect personas and themes (optional)\n4. Generate actionable insights:\n - High-Expectation Customer (HXC) profile\n - What users love (from \"very disappointed\" responses)\n - What holds users back (from \"somewhat disappointed\" responses)\n - Roadmap recommendations (50% double down / 50% address gaps)\n\n## Getting started\n\n1. Click [**Remix**](/?intent=remix)\n2. Customize `PRODUCT_NAME` in [`main.ts`](./main.ts) line 9\n3. Customize [`PROMPT.txt`](./PROMPT.txt) with your product context\n4. Share `/survey` with your users\n5. Watch your PMF score update at `/dashboard`\n\n**Optional:** Add `OPENAI_API_KEY` for AI-powered persona detection and HXC\nprofile generation. The survey works without it — AI just enhances the insights.\n\n## The Framework\n\nThe Superhuman PMF engine uses Sean Ellis's leading indicator: **% of users who\nwould be \"very disappointed\"** without your product.\n\n**Benchmark:** Companies with strong traction exceed **40%**. Companies\nstruggling to grow are below 40%.\n\n### The 4 Survey Questions\n\n1. **How would you feel if you could no longer use [product]?**\n - Very disappointed\n - Somewhat disappointed\n - Not disappointed\n\n2. **What type of people do you think would most benefit from [product]?**\n\n3. **What is the main benefit you receive from [product]?**\n\n4. **How can we improve [product] for you?**\n\n## Features\n\n### Works Without AI\n\n- Survey collection\n- PMF Score calculation\n- Response breakdown (very/somewhat/not)\n- Basic persona detection (keyword-based)\n- Basic theme extraction\n\n### Enhanced With AI (add `OPENAI_API_KEY`)\n\n- Rich persona detection\n- Detailed HXC profile generation\n- Smart roadmap recommendations\n- Nuanced sentiment analysis\n\nAfter adding your API key, POST to `/reprocess` to analyze existing responses.\n\n## How it works\n\n**Survey**\n\n- Users complete the 4-question survey\n- Responses are stored immediately\n\n**Analysis**\n\n- Basic persona/theme extraction runs automatically\n- If `OPENAI_API_KEY` is set, AI enhances the analysis\n\n**Dashboard**\n\n- PMF Score: % of \"very disappointed\" responses\n- Segmented PMF: Score filtered to ideal personas\n- Response breakdown with visual bars\n- HXC Profile (AI-enhanced)\n- Top benefits and improvements\n- 50/50 roadmap recommendations\n\n## Key Insights from the Framework\n\n> \"If you only double down on what users love, your product-market fit score\n> won't increase. If you only address what holds users back, your competition\n> will likely overtake you.\"\n\nThe framework teaches you to:\n\n- **Ignore** \"not disappointed\" users — they won't convert\n- **Focus on** \"very disappointed\" users to understand what to double down on\n- **Convert** \"somewhat disappointed\" users who already value your main benefit","createdAt",["D",1769443591308],"branchVersion","branchId","7d938fa2-fa75-11f0-aded-42dde27851f2","id","7d9310a4-fa75-11f0-aded-42dde27851f2","pmf-survey",["D",1769404297336],"deletedAt","legacyEnvVars","privacy","public","httpPrivacy","description","Superhuman PMF Framework - measure product-market fit","imageUrl","9319c175-2593-4f88-ba21-04f2ef7b1c28","pinnedFileId","pinPreview","pinMigrated","screenshotVersion","likeCount",0,"orgId","d663a026-6449-42f6-9156-1d4a2d64d38c","tursoName","7d9310a4fa7511f0aded42dde27851f2","tursoHostname","7d9310a4fa7511f0aded42dde27851f2-val-town.aws-us-east-2.turso.io","tursoMeasuredAt",["D",1789446061244],"pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_96":95,"_104":100},"org",{"_96":95,"_97":98,"_99":100,"_18":54,"_101":102,"_103":25},"branch",{"_13":14,"_32":86,"_42":41,"_87":43,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":-5,"_93":44,"_94":95},"isOwned","main","projectId","forkedBranchId","forkedBranchVersion","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","dcm31","handle","displayName","Charlie","avatar","https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/3e904faa-3764-4629-c475-f561a49b0000/square512?exp=1789531199&sig=70ba01c808225325e3ce9e8a7dc96e189f737eaf3e04a60a5d4bb714dd4d654a","tier","pro","canRestrictVals","profileImageUrl","branchName","","adoptOrgId",{"_42":43,"_32":44,"_37":226,"_46":-5,"_47":25,"_48":49,"_50":49,"_51":52,"_53":-5,"_18":54,"_55":-5,"_56":25,"_57":-5,"_58":-5,"_59":60,"_61":62,"_63":64,"_65":66,"_67":227,"_69":-5,"_70":71,"_72":71,"_73":74,"_75":74,"_76":74,"_77":71,"_78":71,"_79":228,"_81":229,"_83":230,"_85":25},"tree",[197,198,199,200,201,202,203],"files",[197,198,199,200,201,202,203],"counts",{"_187":60,"_188":167,"_189":60,"_190":191,"_192":193},"contents",[128,129,130,131,132,133,134],"upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles",true,"grants","orgs","customDomains",{},{"_42":181,"_32":182,"_30":183,"_138":-5,"_40":41,"_39":179,"_140":123,"_37":184,"_13":60,"_33":20,"_34":185,"_24":25,"_144":-5,"_27":186},{"_42":177,"_32":28,"_30":31,"_138":-5,"_40":41,"_39":14,"_140":123,"_37":178,"_13":179,"_33":20,"_34":36,"_24":25,"_144":-5,"_27":180},{"_42":170,"_32":171,"_30":172,"_138":-5,"_40":41,"_39":173,"_140":123,"_37":174,"_13":151,"_33":142,"_34":175,"_24":25,"_144":-5,"_27":176},{"_42":162,"_32":163,"_30":164,"_138":-5,"_40":41,"_39":165,"_140":123,"_37":166,"_13":167,"_33":142,"_34":168,"_24":25,"_144":-5,"_27":169},{"_42":155,"_32":156,"_30":157,"_138":-5,"_40":41,"_39":158,"_140":123,"_37":159,"_13":139,"_33":142,"_34":160,"_24":25,"_144":-5,"_27":161},{"_42":146,"_32":147,"_30":148,"_138":-5,"_40":41,"_39":149,"_140":123,"_37":150,"_13":151,"_33":152,"_34":153,"_24":25,"_144":-5,"_27":154},{"_42":135,"_32":136,"_30":137,"_138":-5,"_40":41,"_39":139,"_140":123,"_37":141,"_13":60,"_33":142,"_34":143,"_24":25,"_144":-5,"_27":145},"c5ebcb70-fa75-11f0-8810-42dde27851f2","survey.tsx","c5ebcc6a-fa75-11f0-8810-42dde27851f2","parentId",5,"latest",["D",1769404418711],"script","/** @jsxImportSource npm:hono@4/jsx */\n\ninterface SurveyFormProps {\n productName: string;\n submitUrl: string;\n}\n\nexport function SurveyForm({ productName, submitUrl }: SurveyFormProps) {\n return (\n \n \n \n \n Product Feedback Survey\n \n \n \n

\n Help us improve {productName}\n

\n Your feedback helps us build a better product. This takes about 2 minutes.\n

\n\n

\n {/* Question 1: Disappointment Level */}\n

\n \n \n 1. How would you feel if you could no longer use {productName}?\n \n Required\n \n

\n {[\n { value: \"very\", label: \"Very disappointed\" },\n { value: \"somewhat\", label: \"Somewhat disappointed\" },\n { value: \"not\", label: \"Not disappointed\" },\n ].map((option) => (\n \n \n {option.label}\n \n ))}\n

\n\n {/* Question 2: User Type */}\n

\n \n \n 2. What type of people do you think would most benefit from {productName}?\n \n \n Describe the ideal user in your own words\n \n \n \n \n

\n\n {/* Question 3: Main Benefit */}\n

\n \n \n 3. What is the main benefit you receive from {productName}?\n \n \n What value does it provide that keeps you using it?\n \n \n \n \n

\n\n {/* Question 4: Improvements */}\n

\n \n \n 4. How can we improve {productName} for you?\n \n \n What's missing? What would make it even better?\n \n \n \n \n

\n\n {/* Optional: Email */}\n

\n \n \n Email (optional)\n \n \n If you'd like us to follow up with you\n \n \n \n

\n\n {/* Submit */}\n

\n \n Submit Feedback\n \n

\n\n {/* Success State */}\n

\n \n \n \n

Thank you!

\n Your feedback helps us build a better product for you.\n

\n\n

\n Powered by the{\" \"}\n \n Superhuman PMF Framework\n \n

\n\n {\n e.preventDefault();\n \n const form = e.target;\n const submitBtn = document.getElementById('submit-btn');\n const formData = new FormData(form);\n \n submitBtn.disabled = true;\n submitBtn.textContent = 'Submitting...';\n \n try {\n const response = await fetch('${submitUrl}', {\n method: 'POST',\n headers: { 'Content-Type': 'application/json' },\n body: JSON.stringify({\n disappointment: formData.get('disappointment'),\n user_type: formData.get('user_type'),\n main_benefit: formData.get('main_benefit'),\n improvement: formData.get('improvement'),\n email: formData.get('email') || undefined\n })\n });\n \n if (response.ok) {\n form.classList.add('hidden');\n document.getElementById('success-state').classList.remove('hidden');\n } else {\n throw new Error('Submission failed');\n }\n } catch (err) {\n submitBtn.disabled = false;\n submitBtn.textContent = 'Submit Feedback';\n alert('Something went wrong. Please try again.');\n }\n });\n `,\n }}\n />\n \n \n );\n}\n\nexport function SurveyThankYou() {\n return (\n \n \n \n \n Thank You\n \n \n \n

\n \n \n \n

Thank you!

\n Your feedback has been recorded and will help us improve.\n

\n \n \n );\n}\n","label",[136],"4932980c-fa7a-11f0-a99d-42dde27851f2","main.ts","f8e04c22-fa75-11f0-898c-42dde27851f2",16,["D",1769406356849],3,"http","/** @jsxImportSource npm:hono@4/jsx */\nimport { Hono } from \"npm:hono@4\";\nimport { processNewResponse } from \"./analysis.ts\";\nimport { app as dashboard } from \"./dashboard.tsx\";\nimport { getResponseById, storeResponse } from \"./db.ts\";\nimport { SurveyForm, SurveyThankYou } from \"./survey.tsx\";\n\nconst app = new Hono();\n\n// Configuration - customize this for your product\nconst PRODUCT_NAME = \"Your Product\"; // Change this to your product name\n\n// Survey form\napp.get(\"/survey\", (c) => {\n const submitUrl = new URL(c.req.url).origin + \"/submit\";\n return c.html();\n});\n\n// Survey submission endpoint\napp.post(\"/submit\", async (c) => {\n try {\n const body = await c.req.json();\n\n // Validate required fields\n const { disappointment, user_type, main_benefit, improvement, email } = body;\n\n if (!disappointment || ![\"very\", \"somewhat\", \"not\"].includes(disappointment)) {\n return c.json({ error: \"Invalid disappointment level\" }, 400);\n }\n if (!user_type || typeof user_type !== \"string\") {\n return c.json({ error: \"user_type is required\" }, 400);\n }\n if (!main_benefit || typeof main_benefit !== \"string\") {\n return c.json({ error: \"main_benefit is required\" }, 400);\n }\n if (!improvement || typeof improvement !== \"string\") {\n return c.json({ error: \"improvement is required\" }, 400);\n }\n\n // Store the response (works without AI)\n const id = await storeResponse({\n disappointment,\n user_type,\n main_benefit,\n improvement,\n email: email || undefined,\n });\n\n // Fire-and-forget the AI analysis (optional, won't block)\n const url = new URL(c.req.url);\n fetch(`${url.origin}/process/${id}`, {\n method: \"POST\",\n }).catch(() => {}); // Silently ignore if AI processing fails\n\n return c.json({ ok: true, id });\n } catch (err) {\n console.error(\"Error submitting survey:\", err);\n return c.json({ error: \"Failed to submit survey\" }, 500);\n }\n});\n\n// Background processing endpoint (AI-powered, optional)\napp.post(\"/process/:id\", async (c) => {\n // Skip if no API key - this is optional\n if (!Deno.env.get(\"OPENAI_API_KEY\")) {\n return c.json({ ok: true, skipped: true, reason: \"No OPENAI_API_KEY\" });\n }\n\n const id = parseInt(c.req.param(\"id\"));\n const response = await getResponseById(id);\n\n if (!response) {\n return c.json({ error: \"Response not found\" }, 404);\n }\n\n try {\n await processNewResponse(response);\n return c.json({ ok: true });\n } catch (err) {\n console.error(\"Error processing response:\", err);\n // Don't fail the request - AI is optional\n return c.json({ ok: true, skipped: true, reason: \"AI processing failed\" });\n }\n});\n\n// Manual trigger to reprocess all responses with AI\napp.post(\"/reprocess\", async (c) => {\n if (!Deno.env.get(\"OPENAI_API_KEY\")) {\n return c.json({ error: \"Add OPENAI_API_KEY to enable AI features\" }, 400);\n }\n\n const { reprocessAll } = await import(\"./analysis.ts\");\n try {\n const result = await reprocessAll();\n return c.json({ ok: true, ...result });\n } catch (err) {\n console.error(\"Error reprocessing:\", err);\n return c.json({ error: \"Reprocessing failed\" }, 500);\n }\n});\n\n// Thank you page\napp.get(\"/thanks\", (c) => {\n return c.html();\n});\n\n// Mount the dashboard\napp.route(\"/\", dashboard);\n\n// Root redirects to dashboard\napp.get(\"/\", (c) => {\n return c.redirect(\"/dashboard\");\n});\n\nexport default app.fetch;\n",[147],"99299eb2-fa7c-11f0-a539-42dde27851f2","db.ts","a019aea8-fa75-11f0-8990-42dde27851f2",20,["D",1769407350079],"import sqlite from \"https://esm.town/v/std/sqlite/main.ts\";\n\nconst TABLE_NAME = \"responses\";\nconst INSIGHTS_TABLE = \"insights\";\n\nlet tablesCreated = false;\n\nexport interface SurveyResponse {\n id: number;\n timestamp: number;\n disappointment: \"very\" | \"somewhat\" | \"not\";\n user_type: string;\n main_benefit: string;\n improvement: string;\n email?: string;\n persona?: string;\n analysis?: string;\n}\n\nexport interface Insights {\n id: number;\n timestamp: number;\n pmf_score: number;\n segmented_pmf_score: number;\n total_responses: number;\n very_disappointed_count: number;\n somewhat_disappointed_count: number;\n not_disappointed_count: number;\n hxc_profile: string;\n top_benefits: string;\n top_improvements: string;\n roadmap_double_down: string;\n roadmap_address_gaps: string;\n personas_in_segment: string;\n}\n\nasync function ensureTables(): Promise {\n if (tablesCreated) return;\n \n await sqlite.execute(`\n CREATE TABLE IF NOT EXISTS ${TABLE_NAME} (\n id INTEGER PRIMARY KEY AUTOINCREMENT,\n timestamp INTEGER NOT NULL,\n disappointment TEXT NOT NULL,\n user_type TEXT NOT NULL,\n main_benefit TEXT NOT NULL,\n improvement TEXT NOT NULL,\n email TEXT,\n persona TEXT,\n analysis TEXT\n )\n `);\n\n await sqlite.execute(`\n CREATE TABLE IF NOT EXISTS ${INSIGHTS_TABLE} (\n id INTEGER PRIMARY KEY AUTOINCREMENT,\n timestamp INTEGER NOT NULL,\n pmf_score REAL NOT NULL,\n segmented_pmf_score REAL NOT NULL,\n total_responses INTEGER NOT NULL,\n very_disappointed_count INTEGER NOT NULL,\n somewhat_disappointed_count INTEGER NOT NULL,\n not_disappointed_count INTEGER NOT NULL,\n hxc_profile TEXT,\n top_benefits TEXT,\n top_improvements TEXT,\n roadmap_double_down TEXT,\n roadmap_address_gaps TEXT,\n personas_in_segment TEXT\n )\n `);\n \n tablesCreated = true;\n}\n\nexport async function storeResponse(data: {\n disappointment: \"very\" | \"somewhat\" | \"not\";\n user_type: string;\n main_benefit: string;\n improvement: string;\n email?: string;\n}): Promise {\n await ensureTables();\n \n const result = await sqlite.execute(\n `INSERT INTO ${TABLE_NAME} (timestamp, disappointment, user_type, main_benefit, improvement, email)\n VALUES (?, ?, ?, ?, ?, ?)`,\n [Date.now(), data.disappointment, data.user_type, data.main_benefit, data.improvement, data.email || null],\n );\n return Number(result.lastInsertRowid);\n}\n\nexport async function updateResponseAnalysis(id: number, persona: string, analysis: string): Promise {\n await ensureTables();\n await sqlite.execute(\n `UPDATE ${TABLE_NAME} SET persona = ?, analysis = ? WHERE id = ?`,\n [persona, analysis, id],\n );\n}\n\nexport async function getAllResponses(): Promise {\n await ensureTables();\n const result = await sqlite.execute(`SELECT * FROM ${TABLE_NAME} ORDER BY timestamp DESC`);\n return rowsToObjects(result) as SurveyResponse[];\n}\n\nexport async function getResponseById(id: number): Promise {\n await ensureTables();\n const result = await sqlite.execute(`SELECT * FROM ${TABLE_NAME} WHERE id = ?`, [id]);\n const rows = rowsToObjects(result);\n return (rows[0] as SurveyResponse) || null;\n}\n\nexport async function getResponsesByDisappointment(level: \"very\" | \"somewhat\" | \"not\"): Promise {\n await ensureTables();\n const result = await sqlite.execute(\n `SELECT * FROM ${TABLE_NAME} WHERE disappointment = ? ORDER BY timestamp DESC`,\n [level],\n );\n return rowsToObjects(result) as SurveyResponse[];\n}\n\nexport async function getResponseCounts(): Promise<{\n total: number;\n very: number;\n somewhat: number;\n not: number;\n}> {\n await ensureTables();\n const result = await sqlite.execute(`\n SELECT \n COUNT(*) as total,\n SUM(CASE WHEN disappointment = 'very' THEN 1 ELSE 0 END) as very,\n SUM(CASE WHEN disappointment = 'somewhat' THEN 1 ELSE 0 END) as somewhat,\n SUM(CASE WHEN disappointment = 'not' THEN 1 ELSE 0 END) as not_count\n FROM ${TABLE_NAME}\n `);\n const rows = rowsToObjects(result);\n const row = rows[0] as any;\n return {\n total: Number(row?.total) || 0,\n very: Number(row?.very) || 0,\n somewhat: Number(row?.somewhat) || 0,\n not: Number(row?.not_count) || 0,\n };\n}\n\nexport async function storeInsights(data: Omit): Promise {\n await ensureTables();\n await sqlite.execute(\n `INSERT INTO ${INSIGHTS_TABLE} \n (timestamp, pmf_score, segmented_pmf_score, total_responses, \n very_disappointed_count, somewhat_disappointed_count, not_disappointed_count,\n hxc_profile, top_benefits, top_improvements, roadmap_double_down, roadmap_address_gaps, personas_in_segment)\n VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)`,\n [\n Date.now(),\n data.pmf_score,\n data.segmented_pmf_score,\n data.total_responses,\n data.very_disappointed_count,\n data.somewhat_disappointed_count,\n data.not_disappointed_count,\n data.hxc_profile,\n data.top_benefits,\n data.top_improvements,\n data.roadmap_double_down,\n data.roadmap_address_gaps,\n data.personas_in_segment,\n ],\n );\n}\n\nexport async function getLatestInsights(): Promise {\n await ensureTables();\n const result = await sqlite.execute(\n `SELECT * FROM ${INSIGHTS_TABLE} ORDER BY timestamp DESC LIMIT 1`,\n );\n const rows = rowsToObjects(result);\n return (rows[0] as Insights) || null;\n}\n\n// Helper to convert rows array to objects\nfunction rowsToObjects(result: { columns: string[]; rows: unknown[][] }): Record[] {\n return result.rows.map((row) => {\n const obj: Record = {};\n result.columns.forEach((col, idx) => {\n obj[col] = row[idx];\n });\n return obj;\n });\n}\n",[156],"57f3fcd4-fa78-11f0-bb58-42dde27851f2","dashboard.tsx","f0b30ec2-fa75-11f0-be01-42dde27851f2",12,["D",1769405522703],1,"/** @jsxImportSource npm:hono@4/jsx */\nimport { Hono } from \"npm:hono@4\";\nimport {\n getAllResponses,\n getLatestInsights,\n getResponseById,\n getResponseCounts,\n type Insights,\n type SurveyResponse,\n} from \"./db.ts\";\n\nexport const app = new Hono();\n\nfunction hasAIEnabled(): boolean {\n return !!Deno.env.get(\"OPENAI_API_KEY\");\n}\n\nfunction Layout({ children, title }: { children: any; title: string }) {\n return (\n \n \n \n \n {title}\n \n \n {children}\n \n );\n}\n\nfunction AIBanner({ hasResponses }: { hasResponses: boolean }) {\n const aiEnabled = hasAIEnabled();\n \n if (aiEnabled) return null;\n \n return (\n

\n \n

AI Features Disabled

\n Add OPENAI_API_KEY to enable AI-powered persona detection, HXC profiles, and smart recommendations. \n PMF scoring works without it.\n

\n {hasResponses && (\n

\n After adding your API key, POST to /reprocess to analyze existing responses.\n

\n )}\n

\n );\n}\n\nfunction PMFScoreCard({\n score,\n label,\n benchmark = 40,\n}: {\n score: number;\n label: string;\n benchmark?: number;\n}) {\n const isGood = score >= benchmark;\n const color = isGood ? \"green\" : score >= benchmark * 0.75 ? \"yellow\" : \"red\";\n const bgColors: Record = {\n green: \"bg-green-50 border-green-200\",\n yellow: \"bg-yellow-50 border-yellow-200\",\n red: \"bg-red-50 border-red-200\",\n };\n const textColors: Record = {\n green: \"text-green-700\",\n yellow: \"text-yellow-700\",\n red: \"text-red-700\",\n };\n\n return (\n

{label}

{score.toFixed(1)}%

\n Benchmark: {benchmark}%{\" \"}\n {isGood ? (\n ✓ Above\n ) : (\n Below\n )}\n

\n );\n}\n\nfunction StatCard({ value, label }: { value: number | string; label: string }) {\n return (\n

{label}

{value}

\n );\n}\n\nfunction ResponseBreakdown({\n very,\n somewhat,\n not,\n total,\n}: {\n very: number;\n somewhat: number;\n not: number;\n total: number;\n}) {\n const veryPct = total > 0 ? (very / total) * 100 : 0;\n const somewhatPct = total > 0 ? (somewhat / total) * 100 : 0;\n const notPct = total > 0 ? (not / total) * 100 : 0;\n\n return (\n

Response Breakdown

\n Very Disappointed\n \n {very} ({veryPct.toFixed(1)}%)\n \n

\n Somewhat Disappointed\n \n {somewhat} ({somewhatPct.toFixed(1)}%)\n \n

\n Not Disappointed\n \n {not} ({notPct.toFixed(1)}%)\n \n

\n );\n}\n\nfunction WordCloud({ words, title }: { words: string[]; title: string }) {\n if (words.length === 0) {\n return (\n

{title}

No data yet

\n );\n }\n\n return (\n

{title}

\n {words.map((word, i) => {\n const sizes = [\"text-lg\", \"text-base\", \"text-sm\"];\n const size = sizes[Math.min(i, sizes.length - 1)];\n return (\n \n {word}\n \n );\n })}\n

\n );\n}\n\nfunction RoadmapSection({\n items,\n title,\n description,\n color,\n}: {\n items: string[];\n title: string;\n description: string;\n color: \"green\" | \"blue\";\n}) {\n const colors = {\n green: {\n bg: \"bg-green-50\",\n border: \"border-green-200\",\n bullet: \"bg-green-500\",\n text: \"text-green-800\",\n },\n blue: {\n bg: \"bg-blue-50\",\n border: \"border-blue-200\",\n bullet: \"bg-blue-500\",\n text: \"text-blue-800\",\n },\n };\n const c = colors[color];\n\n return (\n

{title}

{description}

\n {items.length === 0 ? (\n

No recommendations yet

\n ) : (\n

\n \n {item}\n

\n )}\n

\n );\n}\n\nfunction ResponseRow({ response }: { response: SurveyResponse }) {\n const colors: Record = {\n very: { bg: \"bg-green-500\", text: \"text-green-700\" },\n somewhat: { bg: \"bg-yellow-500\", text: \"text-yellow-700\" },\n not: { bg: \"bg-gray-400\", text: \"text-gray-500\" },\n };\n const c = colors[response.disappointment];\n const date = new Date(response.timestamp).toLocaleDateString(\"en-US\", {\n month: \"short\",\n day: \"numeric\",\n year: \"numeric\",\n });\n\n return (\n \n \n \n {response.persona || \"Unprocessed\"}\n \n {response.disappointment}\n {date}\n \n );\n}\n\napp.get(\"/dashboard\", async (c) => {\n const responses = await getAllResponses();\n const counts = await getResponseCounts();\n const insights = await getLatestInsights();\n const surveyUrl = new URL(c.req.url).origin + \"/survey\";\n\n // Calculate PMF score from raw data if no insights yet\n const rawPmfScore = counts.total > 0 ? (counts.very / counts.total) * 100 : 0;\n\n // Parse JSON fields from insights\n let topBenefits: string[] = [];\n let topImprovements: string[] = [];\n let doubleDown: string[] = [];\n let addressGaps: string[] = [];\n let personas: string[] = [];\n\n if (insights) {\n try {\n topBenefits = JSON.parse(insights.top_benefits || \"[]\");\n topImprovements = JSON.parse(insights.top_improvements || \"[]\");\n doubleDown = JSON.parse(insights.roadmap_double_down || \"[]\");\n addressGaps = JSON.parse(insights.roadmap_address_gaps || \"[]\");\n personas = JSON.parse(insights.personas_in_segment || \"[]\");\n } catch {\n // Keep defaults\n }\n }\n\n return c.html(\n \n

\n {/* Header */}\n

PMF Survey Dashboard

\n Track your product-market fit using the Superhuman framework\n

\n\n {/* AI Status Banner */}\n 0} />\n\n {/* Survey Link */}\n

Share Your Survey

{surveyUrl}

\n this.textContent = 'Copy Link', 2000)`}\n class=\"px-4 py-2 bg-indigo-600 hover:bg-indigo-700 text-white font-medium rounded-lg transition-colors\"\n >\n Copy Link\n \n

\n\n {responses.length === 0 ? (\n

\n \n \n \n

No responses yet

\n Share your survey link to start collecting feedback\n

\n \n Preview Survey\n \n

\n ) : (\n <>\n {/* PMF Scores */}\n

\n \n \n \n \n

\n\n {/* Response Breakdown */}\n

\n \n

\n Personas in \"Very Disappointed\" Segment\n

\n {personas.length === 0 ? (\n

\n {hasAIEnabled() \n ? \"No personas identified yet\" \n : \"Add OPENAI_API_KEY for persona detection\"}\n

\n ) : (\n

\n {personas.map((persona) => (\n \n {persona}\n \n ))}\n

\n )}\n

\n These are the personas that appear in your \"very disappointed\" group. The\n segmented PMF score filters to only include these personas.\n

\n\n {/* HXC Profile */}\n {insights?.hxc_profile && (\n

\n High-Expectation Customer (HXC) Profile\n

\n Built from \"very disappointed\" users describing who would benefit most\n

\n {insights.hxc_profile.split(\"\\n\\n\").map((paragraph) => (\n

{paragraph}

\n ))}\n

\n )}\n\n {/* Word Clouds */}\n

\n \n \n

\n\n {/* Roadmap Recommendations */}\n

Roadmap Recommendations

\n Split 50/50 between doubling down on what users love and addressing what holds\n others back\n

\n \n \n

\n\n {/* Responses List */}\n

\n All Responses ({responses.length})\n

\n {responses.map((response) => (\n \n ))}\n

\n \n )}\n\n {/* Footer */}\n

\n Built with the{\" \"}\n \n Superhuman PMF Framework\n \n

\n ,\n );\n});\n\napp.get(\"/response/:id\", async (c) => {\n const id = parseInt(c.req.param(\"id\"));\n const response = await getResponseById(id);\n\n if (!response) {\n return c.html(\n \n

Response Not Found

\n \n ← Back to Dashboard\n \n

\n ,\n 404,\n );\n }\n\n const colors: Record = {\n very: { bg: \"bg-green-50\", border: \"border-green-200\", text: \"text-green-700\" },\n somewhat: { bg: \"bg-yellow-50\", border: \"border-yellow-200\", text: \"text-yellow-700\" },\n not: { bg: \"bg-gray-50\", border: \"border-gray-200\", text: \"text-gray-600\" },\n };\n const c2 = colors[response.disappointment];\n\n let analysis = null;\n if (response.analysis) {\n try {\n analysis = JSON.parse(response.analysis);\n } catch {\n // Keep null\n }\n }\n\n return c.html(\n \n

\n \n ← Back to Dashboard\n \n\n

{response.persona || \"Unprocessed\"}

\n {new Date(response.timestamp).toLocaleString()}\n

\n\n {/* Disappointment Level */}\n

\n \n \n {response.disappointment} Disappointed\n \n

\n\n {/* Survey Responses */}\n

\n Survey Responses\n

\n What type of people would benefit most?\n

{response.user_type}

Main benefit received

{response.main_benefit}

How to improve

{response.improvement}

\n {response.email && (\n

{response.email}

\n )}\n

\n\n {/* Analysis */}\n {analysis && (\n

\n Analysis {!hasAIEnabled() && (Basic)}\n

Assigned Persona

{analysis.persona}

\n {analysis.themes?.benefits?.length > 0 && (\n

Benefit Themes

\n {analysis.themes.benefits.map((b: string) => (\n \n {b}\n \n ))}\n

\n )}\n {analysis.themes?.improvements?.length > 0 && (\n

Improvement Themes

\n {analysis.themes.improvements.map((i: string) => (\n \n {i}\n \n ))}\n

\n )}\n {analysis.sentiment_notes && (\n

Sentiment Notes

{analysis.sentiment_notes}

\n )}\n

\n ,\n );\n});\n\nexport default app.fetch;\n",[163],"399efc82-fa7a-11f0-a11a-42dde27851f2","analysis.ts","b29bf888-fa75-11f0-85d4-42dde27851f2",15,["D",1769406330805],"import { readFile } from \"https://esm.town/v/std/utils/index.ts\";\nimport OpenAI from \"npm:openai\";\nimport {\n getAllResponses,\n getResponsesByDisappointment,\n storeInsights,\n updateResponseAnalysis,\n type SurveyResponse,\n} from \"./db.ts\";\n\nlet cachedPrompt: string | null = null;\n\nasync function getPrompt(): Promise {\n if (cachedPrompt !== null) return cachedPrompt;\n try {\n cachedPrompt = await readFile(\"/PROMPT.txt\", import.meta.url);\n } catch {\n cachedPrompt = \"You are an expert product analyst.\";\n }\n return cachedPrompt;\n}\n\nfunction getOpenAI(): OpenAI | null {\n const apiKey = Deno.env.get(\"OPENAI_API_KEY\");\n if (!apiKey) return null;\n return new OpenAI({ apiKey });\n}\n\ninterface ResponseAnalysis {\n persona: string;\n themes: {\n benefits: string[];\n improvements: string[];\n };\n sentiment_notes: string;\n}\n\n// Simple keyword-based persona extraction (no AI needed)\nfunction extractSimplePersona(userType: string): string {\n const text = userType.toLowerCase();\n \n const roles = [\n { keywords: [\"founder\", \"ceo\", \"co-founder\", \"cofounder\"], persona: \"Founder\" },\n { keywords: [\"executive\", \"c-level\", \"cto\", \"cfo\", \"coo\", \"vp\"], persona: \"Executive\" },\n { keywords: [\"manager\", \"lead\", \"head of\", \"director\"], persona: \"Manager\" },\n { keywords: [\"engineer\", \"developer\", \"programmer\", \"dev\", \"software\"], persona: \"Engineer\" },\n { keywords: [\"designer\", \"ux\", \"ui\", \"product design\"], persona: \"Designer\" },\n { keywords: [\"product manager\", \"pm\", \"product owner\"], persona: \"Product Manager\" },\n { keywords: [\"marketer\", \"marketing\", \"growth\", \"content\"], persona: \"Marketer\" },\n { keywords: [\"sales\", \"account executive\", \"ae\", \"sdr\", \"bdr\"], persona: \"Sales\" },\n { keywords: [\"ops\", \"operations\"], persona: \"Operations\" },\n { keywords: [\"analyst\", \"data\", \"analytics\"], persona: \"Analyst\" },\n { keywords: [\"freelance\", \"consultant\", \"contractor\", \"indie\"], persona: \"Freelancer\" },\n { keywords: [\"student\", \"learning\", \"beginner\"], persona: \"Student\" },\n { keywords: [\"startup\", \"small team\", \"early stage\"], persona: \"Startup\" },\n ];\n \n for (const { keywords, persona } of roles) {\n if (keywords.some(k => text.includes(k))) {\n return persona;\n }\n }\n \n return \"Professional\";\n}\n\n// Extract simple themes from text (no AI needed)\nfunction extractSimpleThemes(text: string): string[] {\n const themes: string[] = [];\n const lower = text.toLowerCase();\n \n const themeKeywords = [\n { keywords: [\"fast\", \"speed\", \"quick\", \"time-sav\"], theme: \"Speed\" },\n { keywords: [\"easy\", \"simple\", \"intuitive\", \"user-friendly\"], theme: \"Ease of use\" },\n { keywords: [\"automat\", \"workflow\", \"trigger\"], theme: \"Automation\" },\n { keywords: [\"integrat\", \"connect\", \"sync\"], theme: \"Integrations\" },\n { keywords: [\"mobile\", \"phone\", \"app\", \"ios\", \"android\"], theme: \"Mobile\" },\n { keywords: [\"collaborat\", \"team\", \"share\", \"together\"], theme: \"Collaboration\" },\n { keywords: [\"organiz\", \"manage\", \"track\", \"overview\"], theme: \"Organization\" },\n { keywords: [\"productiv\", \"efficient\", \"streamlin\"], theme: \"Productivity\" },\n { keywords: [\"reliable\", \"stable\", \"trust\", \"depend\"], theme: \"Reliability\" },\n { keywords: [\"support\", \"help\", \"service\", \"responsive\"], theme: \"Support\" },\n { keywords: [\"price\", \"cost\", \"affordable\", \"expensive\", \"cheap\"], theme: \"Pricing\" },\n { keywords: [\"api\", \"developer\", \"custom\", \"extend\", \"code\"], theme: \"Developer tools\" },\n { keywords: [\"report\", \"analytics\", \"insight\", \"dashboard\"], theme: \"Analytics\" },\n { keywords: [\"secure\", \"privacy\", \"safe\"], theme: \"Security\" },\n { keywords: [\"design\", \"beautiful\", \"clean\", \"ui\"], theme: \"Design\" },\n ];\n \n for (const { keywords, theme } of themeKeywords) {\n if (keywords.some(k => lower.includes(k)) && !themes.includes(theme)) {\n themes.push(theme);\n }\n }\n \n return themes.slice(0, 5);\n}\n\nexport async function analyzeResponse(response: SurveyResponse): Promise {\n const openai = getOpenAI();\n \n // Always extract basic persona (works without AI)\n const basicPersona = extractSimplePersona(response.user_type);\n const basicBenefits = extractSimpleThemes(response.main_benefit);\n const basicImprovements = extractSimpleThemes(response.improvement);\n \n // If no API key, use simple extraction\n if (!openai) {\n const analysis: ResponseAnalysis = {\n persona: basicPersona,\n themes: { benefits: basicBenefits, improvements: basicImprovements },\n sentiment_notes: `${response.disappointment} disappointed`,\n };\n \n await updateResponseAnalysis(response.id, basicPersona, JSON.stringify(analysis));\n return analysis;\n }\n \n // Use AI for richer analysis\n try {\n const PROMPT = await getPrompt();\n const prompt = `${PROMPT}\n\nAnalyze this single survey response:\n\n**Disappointment Level:** ${response.disappointment}\n**What type of people would benefit most?** ${response.user_type}\n**Main benefit received:** ${response.main_benefit}\n**How to improve?** ${response.improvement}\n\nReturn ONLY valid JSON with this exact structure:\n{\n \"persona\": \"string - 1-3 word professional role\",\n \"themes\": {\n \"benefits\": [\"array of key benefit themes mentioned\"],\n \"improvements\": [\"array of improvement requests\"]\n },\n \"sentiment_notes\": \"brief observation about user sentiment\"\n}`;\n\n const completion = await openai.chat.completions.create({\n model: \"gpt-4o-mini\",\n messages: [{ role: \"user\", content: prompt }],\n response_format: { type: \"json_object\" },\n max_tokens: 500,\n });\n\n const content = completion.choices[0].message.content || \"{}\";\n const analysis = JSON.parse(content) as ResponseAnalysis;\n\n await updateResponseAnalysis(response.id, analysis.persona, JSON.stringify(analysis));\n return analysis;\n } catch (err) {\n console.error(\"AI analysis failed, using simple extraction:\", err);\n const analysis: ResponseAnalysis = {\n persona: basicPersona,\n themes: { benefits: basicBenefits, improvements: basicImprovements },\n sentiment_notes: `${response.disappointment} disappointed`,\n };\n await updateResponseAnalysis(response.id, basicPersona, JSON.stringify(analysis));\n return analysis;\n }\n}\n\nexport async function reprocessAll(): Promise<{ processed: number; aiEnabled: boolean }> {\n const allResponses = await getAllResponses();\n const aiEnabled = !!getOpenAI();\n \n for (const response of allResponses) {\n await analyzeResponse(response);\n }\n \n await generateAggregateInsights();\n \n return { processed: allResponses.length, aiEnabled };\n}\n\nexport async function generateAggregateInsights(): Promise {\n const allResponses = await getAllResponses();\n const veryDisappointed = await getResponsesByDisappointment(\"very\");\n const somewhatDisappointed = await getResponsesByDisappointment(\"somewhat\");\n\n const total = allResponses.length;\n const veryCount = veryDisappointed.length;\n const somewhatCount = somewhatDisappointed.length;\n const notCount = total - veryCount - somewhatCount;\n\n const pmfScore = total > 0 ? (veryCount / total) * 100 : 0;\n\n const vdPersonas = new Set();\n for (const r of veryDisappointed) {\n if (r.persona) vdPersonas.add(r.persona);\n }\n\n const segmentedResponses = allResponses.filter(\n (r) => r.persona && vdPersonas.has(r.persona),\n );\n const segmentedVD = segmentedResponses.filter((r) => r.disappointment === \"very\").length;\n const segmentedTotal = segmentedResponses.length;\n const segmentedPmfScore = segmentedTotal > 0 ? (segmentedVD / segmentedTotal) * 100 : 0;\n\n const insights = await generateInsights(veryDisappointed, somewhatDisappointed, Array.from(vdPersonas));\n\n await storeInsights({\n pmf_score: Math.round(pmfScore * 10) / 10,\n segmented_pmf_score: Math.round(segmentedPmfScore * 10) / 10,\n total_responses: total,\n very_disappointed_count: veryCount,\n somewhat_disappointed_count: somewhatCount,\n not_disappointed_count: notCount,\n hxc_profile: insights.hxc_profile,\n top_benefits: JSON.stringify(insights.top_benefits),\n top_improvements: JSON.stringify(insights.top_improvements),\n roadmap_double_down: JSON.stringify(insights.roadmap_double_down),\n roadmap_address_gaps: JSON.stringify(insights.roadmap_address_gaps),\n personas_in_segment: JSON.stringify(Array.from(vdPersonas)),\n });\n}\n\nasync function generateInsights(\n veryDisappointed: SurveyResponse[],\n somewhatDisappointed: SurveyResponse[],\n vdPersonas: string[],\n): Promise<{\n hxc_profile: string;\n top_benefits: string[];\n top_improvements: string[];\n roadmap_double_down: string[];\n roadmap_address_gaps: string[];\n}> {\n if (veryDisappointed.length === 0) {\n return {\n hxc_profile: \"Not enough data yet.\",\n top_benefits: [],\n top_improvements: [],\n roadmap_double_down: [],\n roadmap_address_gaps: [],\n };\n }\n\n const benefitCounts: Record = {};\n for (const r of veryDisappointed) {\n for (const theme of extractSimpleThemes(r.main_benefit)) {\n benefitCounts[theme] = (benefitCounts[theme] || 0) + 1;\n }\n }\n \n const improvementCounts: Record = {};\n for (const r of somewhatDisappointed) {\n for (const theme of extractSimpleThemes(r.improvement)) {\n improvementCounts[theme] = (improvementCounts[theme] || 0) + 1;\n }\n }\n \n const topBenefits = Object.entries(benefitCounts)\n .sort((a, b) => b[1] - a[1])\n .slice(0, 5)\n .map(([theme]) => theme);\n \n const topImprovements = Object.entries(improvementCounts)\n .sort((a, b) => b[1] - a[1])\n .slice(0, 5)\n .map(([theme]) => theme);\n\n const openai = getOpenAI();\n \n if (!openai) {\n return {\n hxc_profile: vdPersonas.length > 0 \n ? `Your most engaged users are: ${vdPersonas.join(\", \")}. Add OPENAI_API_KEY for detailed insights.`\n : \"Add OPENAI_API_KEY to generate HXC profile.\",\n top_benefits: topBenefits,\n top_improvements: topImprovements,\n roadmap_double_down: topBenefits.slice(0, 3).map(b => `Enhance ${b.toLowerCase()}`),\n roadmap_address_gaps: topImprovements.slice(0, 3).map(i => `Improve ${i.toLowerCase()}`),\n };\n }\n\n try {\n const PROMPT = await getPrompt();\n const vdBenefits = veryDisappointed.map((r) => r.main_benefit).join(\"\\n- \");\n const vdUserTypes = veryDisappointed.map((r) => r.user_type).join(\"\\n- \");\n const sdImprovements = somewhatDisappointed.map((r) => r.improvement).join(\"\\n- \");\n\n const prompt = `${PROMPT}\n\n## Survey Data\n\n### \"Very Disappointed\" Users (${veryDisappointed.length})\n**Who benefits most:** ${vdUserTypes}\n**Main benefits:** ${vdBenefits}\n\n### \"Somewhat Disappointed\" Users (${somewhatDisappointed.length})\n**Improvements wanted:** ${sdImprovements || \"None yet\"}\n\nGenerate insights as JSON:\n{\n \"hxc_profile\": \"2-3 paragraph High-Expectation Customer profile\",\n \"top_benefits\": [\"benefit 1\", ...],\n \"top_improvements\": [\"improvement 1\", ...],\n \"roadmap_double_down\": [\"feature 1\", ...],\n \"roadmap_address_gaps\": [\"feature 1\", ...]\n}`;\n\n const completion = await openai.chat.completions.create({\n model: \"gpt-4o\",\n messages: [{ role: \"user\", content: prompt }],\n response_format: { type: \"json_object\" },\n max_tokens: 2000,\n });\n\n return JSON.parse(completion.choices[0].message.content || \"{}\");\n } catch (err) {\n console.error(\"AI insights failed:\", err);\n return {\n hxc_profile: `Your most engaged users are: ${vdPersonas.join(\", \") || \"various professionals\"}.`,\n top_benefits: topBenefits,\n top_improvements: topImprovements,\n roadmap_double_down: topBenefits.slice(0, 3).map(b => `Enhance ${b.toLowerCase()}`),\n roadmap_address_gaps: topImprovements.slice(0, 3).map(i => `Improve ${i.toLowerCase()}`),\n };\n }\n}\n\nexport async function processNewResponse(response: SurveyResponse): Promise {\n await analyzeResponse(response);\n await generateAggregateInsights();\n}\n",[171],"fa9d0356-fad0-11f0-98c8-42dde27851f2",["D",1769443591308],2,[28],"93ab5522-fa75-11f0-bd78-42dde27851f2","PROMPT.txt","93ab55ea-fa75-11f0-bd78-42dde27851f2",["D",1769404334403],"You are an expert product analyst specializing in the Superhuman PMF (Product-Market Fit) framework.\n\nYour job is to analyze survey responses and extract key insights for product-market fit optimization.\n\n## Your Product Context\n\nProduct Name: [YOUR PRODUCT NAME]\nProduct Description: [Brief description of what your product does]\n\n## Survey Response Analysis\n\nFor each survey response, you will:\n\n1. **Assign a Persona** based on the user's answer to \"What type of people would benefit most?\"\n - Extract the most relevant professional role or user type\n - Examples: \"Founder\", \"Manager\", \"Engineer\", \"Sales Rep\", \"Designer\", \"Executive\"\n - Keep personas concise (1-3 words)\n\n2. **Extract Key Themes** from the main benefit (Q3) and improvement requests (Q4)\n - Identify the core value propositions mentioned\n - Note specific features or capabilities valued\n - Flag concrete improvement requests\n\nReturn your analysis as JSON:\n\n{\n \"persona\": string,\n \"themes\": {\n \"benefits\": string[],\n \"improvements\": string[]\n },\n \"sentiment_notes\": string\n}\n\n## Aggregate Analysis\n\nWhen analyzing all responses together for insights, follow the Superhuman framework:\n\n### PMF Score Calculation\n- Count responses by disappointment level\n- PMF Score = (Very Disappointed / Total) × 100\n- Benchmark: 40% indicates strong product-market fit\n\n### Segmentation Strategy\n1. Identify all personas from \"very disappointed\" responses\n2. Filter to only include users with those personas\n3. Recalculate PMF Score for segmented group\n4. This reveals true fit within your ideal user base\n\n### High-Expectation Customer (HXC) Profile\nUsing ONLY \"very disappointed\" responses to Q2 (\"What type of people...\"):\n- Users describe themselves when answering this question\n- Synthesize into a rich, detailed persona description\n- Include: role, work patterns, pain points, goals, mindset\n\n### Benefits Analysis (What to Double Down On)\nFrom \"very disappointed\" users' Q3 answers:\n- Identify the most common benefits mentioned\n- These are your core value propositions\n- These should inform 50% of your roadmap\n\n### Improvements Analysis (What Holds Users Back)\nFrom \"somewhat disappointed\" users who value your main benefit:\n- First filter to those whose Q3 answer mentions your core benefits\n- Then analyze their Q4 answers\n- These are the barriers to deeper adoption\n- These should inform the other 50% of your roadmap\n\n### Roadmap Recommendations\nSplit evenly:\n- **Double Down (50%)**: Features/improvements that enhance what users already love\n- **Address Gaps (50%)**: Features that remove barriers for almost-fans\n\n### Important Notes\n- IGNORE \"not disappointed\" users entirely - they won't convert\n- The \"somewhat disappointed\" group matters only if they value your main benefit\n- Track PMF Score over time to measure progress\n",[182],"pulls","branches","forks","envVars",{"_194":60,"_196":60},"connections",{"_194":60,"_195":60},"count","unfilledCount","emptyCount",{"_42":181,"_32":182,"_30":183,"_138":-5,"_40":41,"_39":179,"_140":123,"_37":223,"_13":60,"_33":20,"_34":185,"_24":25,"_144":-5,"_205":224,"_27":225},{"_42":177,"_32":28,"_30":31,"_138":-5,"_40":41,"_39":14,"_140":123,"_37":220,"_13":179,"_33":20,"_34":36,"_24":25,"_144":-5,"_205":221,"_27":222},{"_42":170,"_32":171,"_30":172,"_138":-5,"_40":41,"_39":173,"_140":123,"_37":217,"_13":151,"_33":142,"_34":175,"_24":25,"_144":-5,"_205":218,"_27":219},{"_42":162,"_32":163,"_30":164,"_138":-5,"_40":41,"_39":165,"_140":123,"_37":214,"_13":167,"_33":142,"_34":168,"_24":25,"_144":-5,"_205":215,"_27":216},{"_42":155,"_32":156,"_30":157,"_138":-5,"_40":41,"_39":158,"_140":123,"_37":211,"_13":139,"_33":142,"_34":160,"_24":25,"_144":-5,"_205":212,"_27":213},{"_42":146,"_32":147,"_30":148,"_138":-5,"_40":41,"_39":149,"_140":123,"_37":208,"_13":151,"_33":152,"_34":153,"_24":25,"_144":-5,"_205":209,"_27":210},{"_42":135,"_32":136,"_30":137,"_138":-5,"_40":41,"_39":139,"_140":123,"_37":204,"_13":60,"_33":142,"_34":143,"_24":25,"_144":-5,"_205":206,"_27":207},["D",1769404418711],"children",[],[136],["D",1769406356849],[],[147],["D",1769407350079],[],[156],["D",1769405522703],[],[163],["D",1769406330805],[],[171],["D",1769443591308],[],[28],["D",1769404334403],[],[182],["D",1769404297336],["D",1789446061244],{"_96":95,"_104":100},{"_96":95,"_97":98,"_99":100,"_18":54,"_101":102,"_103":25},{"_13":14,"_32":86,"_42":41,"_87":43,"_88":-5,"_89":-5,"_90":-5,"_91":-5,"_92":-5,"_93":44,"_94":95},"ctx",{"_79":-5,"_125":264,"_265":266,"_267":268,"_269":25,"_270":25,"_271":25,"_272":123,"_273":-5,"_274":-7,"_275":-7,"_276":60,"_277":25,"_278":25,"_279":-5,"_280":25,"_281":25,"_282":25,"_283":-5},"ENV",{"_235":236,"_237":238,"_239":86,"_240":241,"_242":243,"_244":245,"_246":247,"_248":249,"_250":251,"_252":253,"_254":255,"_256":257,"_258":259,"_260":261,"_262":263},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","e87eb04e707cf64e7e532c8e8ec395f209e17f15","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town",[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","isBusinessOrgMember","businessOrgHandle","isBusinessOrgAdmin","isHigherTownieTier","navCollapsed","currentOrgId"]

Diff · v21