As of 13 August 2026, AI can only partly track the return on investment from your paid ads.

What to actually do

1. Hand it to a person

   The route this page recommends

   A person who owns the outcome does this end to end, worth it when the failure is dear.

2. Use a tool built for this

   Second choice

3. Do it yourself

   The distant third

   A chat interface, chat-fluent skill, and roughly 1 hour until you can act on the result.

   How to actually do it

   1. Open each relevant advertising platform and export the reporting period with campaign name, spend, clicks, conversions, conversion value and attribution window.
   2. Export matching orders or leads from your website analytics, CRM or shop system, including date, campaign identifiers, revenue, refunds and customer status where available.
   3. Gather gross margin, payment fees, agency fees and other directly attributable costs from your finance records, and label any field that is unavailable.
   4. Paste the exports, business context and attribution rule into a chatbot with the supplied prompt, keeping the column headings and source names intact.
   5. Ask the chatbot to produce the campaign table and formula for each result, then save the output with the reporting period and data sources.
   6. Compare every spend and conversion total with the original advertising-platform reports and compare revenue, refunds and costs with your finance or CRM records.
   7. Check that the stated attribution window, margin assumptions and currency match your internal reporting rules before sharing the ROI report or changing any budget.

   Prompt

   Analyse the paid advertising data below and calculate return on investment for each campaign, ad group and channel where the data supports it.
   
   Use this definition unless the data makes it impossible: ROI = (attributed gross profit - advertising spend - directly attributable campaign costs) / (advertising spend + directly attributable campaign costs). Also show ROAS separately as attributed revenue / advertising spend. Do not confuse revenue, profit, ROAS and ROI.
   
   Before calculating, list the fields you found and identify missing, duplicated, inconsistent or likely unreliable data. State the attribution window, currency, reporting period and whether revenue is gross or net. If gross margin, refunds, fees or other costs are missing, do not invent them. Show a revenue-based result if possible, label it clearly, and explain why a profit-based ROI cannot be calculated.
   
   For every result, show the formula, the input values, the source row or source report, and any assumptions. Flag campaigns with too little data, zero spend, zero conversions or attribution ambiguity. Do not claim that one campaign caused a sale unless the supplied data supports that attribution.
   
   Return:
   1. A concise data-quality assessment.
   2. A table by campaign with spend, attributed conversions, attributed revenue, attributed gross profit if supplied, ROAS and ROI where calculable.
   3. A comparison of campaigns, with no recommendation to increase or reduce spend unless the data supports it.
   4. A list of checks I must perform in the advertising platform, analytics system and finance records before using the results.
   5. A short explanation for a non-specialist colleague.
   
   Business context: [describe your business, sales cycle and main conversion].
   Attribution rule used by the business: [enter the rule or write unknown].
   Directly attributable costs: [enter them or write unknown].
   Data:
   [paste exports or tables here]

   Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.