← Prompt library

Google Ads Planning Prompt: Intent, Structure and Evidence

Outline a search campaign using real service areas and available keyword data. Separate hypotheses from measured costs and conversions.

Copyable promptPractical instructions
Google Ads Planning Prompt: Intent, Structure and Evidence

Illustrative cover; not a verified result from this prompt.

Read the full prompt

Help outline a Google Ads search campaign for [BUSINESS] in [SERVICE AREA]. Offer, landing pages and budget context: [DETAILS]. Use supplied keyword or account data: [DATA]. Group a manageable set of relevant queries by intent, suggest negatives with reasons and draft accurate ad text. Identify missing measurement and conversion information before discussing bidding. Check current format constraints against official documentation where available. Do not claim certification or invent CPC, search volume, conversion rates or guaranteed outcomes.

Start with the service area and intent

Describe what the business actually offers and where it can deliver. A huge keyword list can introduce irrelevant traffic if the underlying offer is unclear. Treat a campaign draft as a plan for review, not a launch instruction.

Example: bicycle repairs in Jaipur

Use a shop that handles punctures, brake adjustment and routine servicing within its local area. Separate urgent repair searches from general service searches. Ask why a term such as “bicycle jobs” might be excluded, while avoiding negatives that could block a genuine repair request.

Provide the real opening hours and booking page before drafting those claims into an ad. Leave unavailable services out even if their keywords look attractive.

Verify before spending

Use current account tools or exported data for demand and cost estimates. Check conversion measurement, location settings, search terms and landing-page accuracy. No prompt can infer a reliable cost per lead from the business name. Start any approved campaign with a budget and monitoring plan appropriate to the owner, rather than trusting generated forecasts.