Google Ads ROI Calculator

A production-ready web tool for account managers to estimate Google Ads revenue potential for clients based on industry, service mix, cost per lead, and close rate.

Quick Start

npm install
npm run dev

Open http://localhost:3000.

How It Works

The calculator answers: If this client spends $X/month on Google Ads, and their close rate is Y%, and their average job value is $Z, then they can expect approximately $N/month in new revenue potential.

Formulas

leads/month    = monthly ad spend / cost per lead
jobs/month     = leads × close rate
revenue/month  = jobs × average job value

For multiple services, the budget is split by allocation percentages and each service is calculated separately, then summed.

Workflow

1. Client Info — Enter client name, select industry, optionally paste website text or GBP description
2. Service Selection — Select which services the client offers; adjust CPL and job values per service
3. Budget & Sales — Set monthly ad spend, close rate, and optional gross margin
4. Results — See leads, jobs, and revenue estimates with per-service breakdown
5. Export — Copy a client-facing summary or internal notes

Project Structure

src/
  app/
    page.tsx              # Main page (all state management)
    layout.tsx            # Root layout
    globals.css           # Global styles
  components/
    ClientInputs.tsx      # Client name, industry, website text
    ServiceSelection.tsx  # Service checkboxes, CPL, job value per service
    BudgetSalesInputs.tsx # Ad spend, close rate, rounding mode
    Results.tsx           # Results cards and per-service table
    ExportSummary.tsx     # Client-facing and internal summary export
  lib/
    types.ts              # TypeScript type definitions
    benchmarks.ts         # Benchmark data access functions
    calculations.ts       # Core calculation engine + formatters
    service-extraction.ts # Text-based service detection
  data/
    benchmarks.json       # Seed benchmark dataset (EDIT THIS)

Updating Industry Benchmarks

All seed data is placeholder. Before using with real clients, replace the values in src/data/benchmarks.json.

Each benchmark entry:

{
  "industryId": "plumbing",
  "industryName": "Plumbing",
  "serviceName": "Drain Cleaning",
  "cplLow": 20,
  "cplMid": 40,
  "cplHigh": 70,
  "avgJobValue": 350,
  "recommendedMinAdSpend": 800,
  "recommendedTargetAdSpend": 1500,
  "notes": "Source and rationale here",
  "source": "Name of data source",
  "confidence": "high"
}

• confidence: "high", "medium", or "low" — displayed in the UI
• notes / source: shown to the AM as context
• Set cplLow/Mid/High to null if unknown (the UI will prompt for manual entry)

To add a new industry, add entries with a new industryId — the app picks them up automatically.

Accuracy Rules

• No benchmarks are fabricated. All seed data is clearly labeled as demonstration-only.
• Missing benchmarks show "No benchmark available" and require manual entry.
• Custom/manual values are flagged in output as "custom estimate based on user-entered values."
• Formulas are always visible in the UI.
• Confidence levels are shown per service.

Tech Stack

• Next.js 16 (App Router)
• TypeScript
• Tailwind CSS
• No database required — JSON seed data

Build for Production

npm run build
npm start