AI-Assisted PPC Management Tool Creation

Cliente Freelancer · Remoto · Remoto · freelance · mid · 3000–5000 GBP

Publicada el 2026-07-15

Descripción de la oferta

# Build an AI-Powered PPC Agency Assistant ## Project Overview We are looking for an experienced Python, API and AI automation developer to build the first working version of an AI-powered operating assistant for our PPC agency. The PPC Agent will combine Google Ads and GA4 data, client targets, account history, documents, emails, calendars and AI analysis to support day-to-day PPC management. This is not a basic dashboard or an autonomous campaign management tool. It should help us understand performance, identify issues, manage budget pacing, prepare reports and recommend next steps. All Google Ads changes will remain under our team’s control. ## Core Objectives The PPC Agent should help us: * Monitor Google Ads and GA4 performance * Identify changes, risks and opportunities * Manage monthly budget pacing * Review search terms and impression share * Apply client targets and account history to analysis * Generate weekly, monthly, quarterly and annual reports * Create client-ready emails and internal action lists * Prepare meeting materials and follow-up actions * Create and update spreadsheets, documents and presentations * Log data imports, AI outputs and recommendations The system must remain human-led. AI should analyse, recommend and prepare, while our team remains responsible for account changes and client communication. ## Required Technology and Integrations The solution should use a lightweight, scalable stack based around: * Python * Google Ads API * GA4 Data API * OpenAI API and/or Claude API * Google Workspace APIs * Microsoft Graph API * Secure OAuth * Scheduled cloud jobs * Version-controlled source code The architecture should allow future migration to BigQuery and a custom application. ## Google Ads and GA4 The system should securely pull scheduled data across multiple client accounts, including: * Campaign, ad group and keyword performance * Search terms * Spend, clicks, impressions, CTR and CPC * Conversions, conversion value, CPA, CVR and ROAS * Campaign budgets and bid strategies * Impression share and lost impression share * RSA asset information * Google Ads change history * GA4 sessions, events, conversions, revenue and landing-page data The Google Ads integration must be read-only. Imports should include error handling, duplicate prevention and clear logs. ## Client Memory Each client should have editable settings including: * Monthly budget * Target CPA or ROAS * Priority and excluded services * Primary conversion actions * Reporting tone * Known tracking issues * Protected campaigns * Strategic priorities * Google Drive and OneDrive folders * Reporting templates * Gmail, Outlook and calendar mappings Our team should be able to add and update clients without changing the code. ## Performance Analysis The system should answer: * What changed? * Why does it matter? * How does performance compare with the client target? * Which campaigns require attention? * What should we do next? * What should we communicate to the client? Outputs must distinguish between confirmed data, account changes, client communication, external context and AI inference. ## Budget Pacing and Recommendations The system should calculate: * Month-to-date spend * Remaining budget * Projected month-end spend * Required daily spend * Current and recommended daily budgets * Overspend or underspend risk * Recommended action, reason and confidence Recommendations should consider pacing, CPA, ROAS, conversion quality, client priorities, bid strategy, impression share, seasonality, tracking issues and remaining days in the month. Recommendations should be written to a control sheet for review. The system must not make changes directly in Google Ads. ## Search Term and Impression Share Analysis The search term module should classify queries as: * Negative candidate * Expansion opportunity * High intent * Monitor * Irrelevant * Competitor * Research intent * Job intent * Excluded service The output should include performance data, suggested action, reason, confidence and review status. The impression share module should identify whether campaigns are limited by budget, rank, competition, demand, bid strategy or efficiency. The system should prepare recommendations only. It must not add negatives, create keywords or change campaign settings. ## Google Workspace Integration The system should integrate with: * Google Drive * Gmail * Google Calendar * Google Sheets * Google Docs * Google Slides Required capabilities include: * Read approved files and folders * Save reports to specified Drive locations * Create and update Sheets, Docs and Slides * Use agency-owned templates * Read approved Gmail messages for client context * Prepare Gmail drafts * Read relevant calendar events * Prepare meeting notes and reporting materials Client emails must not be sent automatically. Calendar events should only be created where approved. ## Microsoft 365 Integration The system should support: * Microsoft OneDrive * Microsoft Excel * Microsoft Word * Microsoft PowerPoint * Microsoft Outlook * Microsoft Calendar * Microsoft Teams where agreed Required capabilities include: * Read approved OneDrive files and folders * Create and update Excel workbooks * Generate editable Word documents and PowerPoint presentations * Use agency-owned Microsoft templates * Read approved Outlook messages * Prepare Outlook drafts * Read relevant calendar events * Send internal Teams notifications where included Outputs should include .xlsx, .docx, .pptx, Google Workspace formats, CSV and PDF where appropriate. ## Reporting and Document Generation The PPC Agent should generate editable: * Weekly and monthly reports * Quarterly Business Reviews * Annual reviews * Client presentations * Budget pacing and search term workbooks * Forecasts * Meeting preparation notes * Internal action briefs * Client email drafts Reports should use British English and remain concise, commercial and grounded in data. Generated files should preserve agency branding, templates, layouts and formulas where technically possible. ## Logging, Security and Governance The system must log: * Data imports and API errors * AI outputs * Files accessed or created * Email drafts * Calendar actions * Budget and search term recommendations * User reviews * Data quality issues Security requirements include: * No credentials stored in shared files * No secrets committed to source code * Secure OAuth * Least-privilege access * Client data separation * Controlled email and calendar permissions * Read-only Google Ads access * Clear credential setup and revocation guidance * Ability to disable individual integrations ## Main Deliverables The successful freelancer should provide: * Google Ads and GA4 data integrations * Google Workspace and Microsoft 365 integrations * Structured operating workbook * Client memory and configuration layer * Performance analysis module * Budget pacing and recommendation workflow * Search term and impression share modules * AI reporting and communication system * Logging and error monitoring * Clean source code in a shared Git repository * Setup and security documentation * Data dictionary and prompt templates * Handover and training * An agreed post-launch bug-fix period ## Suggested Milestones 1. Discovery, architecture and security design 2. Google Ads, GA4 and operating workbook 3. Google Workspace and Microsoft 365 integrations 4. Performance, budget and search intelligence modules 5. Reporting and communication workflows 6. Testing, documentation and handover ## Out of Scope for Phase 1 Phase 1 will not include: * Direct changes to Google Ads * Automated budget changes * Automatic keyword, negative or ad publication * Automatic bid strategy or campaign setting changes * Unapproved client emails or calendar changes * A full custom web application * Fully autonomous account management ## Experience Required Please apply only if you have relevant experience with several of the following: * Google Ads API data extraction * GA4 Data API * Python automation * Google Workspace APIs * Google Docs and Slides automation * Microsoft Graph API * OneDrive, Excel, Word and PowerPoint automation * Outlook and Microsoft Calendar * OpenAI or Claude integrations * OAuth and credential security * Automated PPC reporting * Budget pacing and forecasting * Cloud scheduling and error monitoring Experience building PPC, marketing analytics or agency workflow tools will be highly valued. ## What to Include in Your Proposal Please include: 1. Your understanding of the project 2. Relevant Google Ads API experience / Apps Scripts 3. Relevant Google Workspace experience 4. Relevant Microsoft 365 or Microsoft Graph experience 5. Your proposed architecture and technology stack 6. Your security and permissions approach 7. Your PPC budget pacing and forecasting experience 8. Your document and presentation generation approach 9. Your recommended milestones and delivery plan 10. Clear inclusions and exclusions 11. Your post-launch support approach 12. Confirmation that source code and documentation will transfer to us Please start your proposal with: PPC Agent Build Generic proposals will not be considered. ## Final Objective We want a practical and reliable AI assistant that helps us manage PPC accounts more consistently. The first version should improve how we analyse performance, manage budget pacing, prepare reports, communicate with clients and maintain a clear history of account activity. The system should provide recommendations and supporting evidence, while all Google Ads changes remain under our direct control.

Skills

Fuente original: freelancer

Análisis JobHunter