Before & After Examples

Before

“Write a blog post about AI”

After

You are a technology journalist with 10 years of experience writing for Fortune 500 audiences. Write a 1,000-word blog post about how small businesses can use AI tools to save time. Target audience: non-technical business owners aged 35-55. Include: 3 specific tools with real pricing, a concrete ROI example, and one cautionary note. Tone: practical, trustworthy, no hype. Format: engaging headline, intro hook, 3 main points with subheadings, actionable conclusion.

Before

“Write an email to my client”

After

You are a professional consultant. Write a follow-up email to a client who has gone quiet after our proposal 2 weeks ago. Goal: re-engage without being pushy. Tone: warm, confident, concise. Include: a specific reference to their stated goal from our last meeting, one concrete value we provide, a soft call to action with two easy options (a 15-min call or a written response). Max 150 words. Subject line included.

Before

“Write code to handle user authentication”

After

Write a TypeScript Next.js 14 App Router API route handler for user authentication using Supabase Auth. Requirements: (1) Handle POST requests with email/password, (2) Return structured JSON responses with proper HTTP status codes (200, 401, 422, 500), (3) Include input validation using Zod, (4) Set secure HTTP-only cookies for the session, (5) Handle the case where the user does not exist vs wrong password differently, (6) Include JSDoc comments. Export as named export. Do not include any client-side code.

Before

“How do I optimize my database queries?”

After

Act as a senior PostgreSQL database engineer. I have a slow query on a users table with 2M rows. First, explain the top 5 query optimization techniques relevant to read-heavy workloads. Then, for each technique: explain when to apply it, show a before/after SQL example, and estimate the performance impact. Finally, give me a step-by-step diagnostic process using EXPLAIN ANALYZE to identify which technique I need. Be specific — avoid generic advice.

Before

“A futuristic city”

After

Sprawling cyberpunk megacity at twilight, neon-lit skyscrapers reflecting in rain-soaked streets below, flying vehicles with light trails, holographic advertisements in Chinese and English, dense urban canyon perspective from street level, cinematic lighting, ultra-detailed, 8K, inspired by Blade Runner 2049 and Ghost in the Shell, photorealistic rendering --ar 16:9 --v 6 --q 2 --style raw

Before

“A portrait of a woman”

After

Professional portrait photograph of a 35-year-old South Asian woman, natural studio lighting with soft key light from camera left, confident and warm expression, wearing a deep navy blazer, shallow depth of field, bokeh background, shot on Sony A7R IV with 85mm f/1.8 lens, skin texture visible, photorealistic, editorial style --ar 4:5 --v 6 --style raw

Before

“Help me with my business plan”

After

Act as a venture-backed startup advisor who has helped 50+ startups raise Series A funding. I am building a B2B SaaS product for restaurant inventory management. Create a structured business plan outline covering: (1) Problem & market size with specific data points, (2) Solution & unique value proposition, (3) Target customer profile with 3 buyer personas, (4) Revenue model with pricing strategy, (5) Go-to-market strategy for first 100 customers, (6) Key metrics to track in first 6 months. For each section, include 2-3 questions I must be able to answer. Format as a structured document with headers.

Before

“Write a SWOT analysis”

After

Conduct a detailed SWOT analysis for [Company/Product]. Structure your response as follows: For each of the 4 quadrants (Strengths, Weaknesses, Opportunities, Threats): list 5 specific factors (not generic ones), rate each factor High/Medium/Low impact, explain the reasoning in one sentence. Then add a "Strategic Implications" section with 3 concrete actions that emerge from the SWOT intersections (SO strategies, ST strategies, WO strategies). Be critical and objective — include uncomfortable truths in Weaknesses and Threats.

Before

“Explain machine learning”

After

Explain machine learning to a 45-year-old marketing director with no technical background but strong business acumen. Use only business analogies and real-world marketing examples — no mathematical formulas. Cover: (1) What it is in one sentence using an analogy they would understand, (2) How it is already being used in marketing today with 3 specific examples they may have encountered, (3) The difference between traditional software and ML in practical terms, (4) One thing they should start paying attention to in their own marketing data. Max 300 words. Avoid jargon completely.

Before

“Analyze this market”

After

Conduct a comprehensive market analysis of the project management software market for 2024. Structure your analysis using the following framework: (1) Market size & growth (TAM, SAM, SOM with sources), (2) Key player competitive matrix (compare top 5 players on: pricing, target segment, key feature, market share), (3) Customer segmentation (3 distinct buyer types with pain points), (4) Emerging trends (3 trends with supporting evidence), (5) Market gaps & opportunities. For each section, flag assumptions clearly. Present data in tables where possible. Conclude with 3 strategic entry points for a new competitor.