As of 13 August 2026, AI can analyse your social media advert results.

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 30 minutes until you can act on the result.

   How to actually do it

   1. Open the reporting area for the social platform where the adverts ran and export campaign or advert-level results for the same date range.
   2. Gather the campaign objective, budget, audience, location, conversion definition, attribution window and any changes to the advert, landing page or tracking.
   3. Remove passwords, access tokens, customer names, email addresses and other unnecessary personal data from the export, then keep the column headings and date ranges intact.
   4. Paste the cleaned table and the campaign context into a chatbot with the prompt above, and ask it to use the platform's reported figures rather than estimates.
   5. Compare every total, spend figure, conversion count and calculated rate in the response with the original platform export, checking the stated formulae on a sample of rows.
   6. Mark conclusions as observations or hypotheses, then choose only tests whose budget, audience, attribution window and success metric you can define before making campaign changes.

   Prompt

   Analyse the social media advert results below. Use only the data I provide and do not invent missing figures. First check for missing columns, inconsistent totals, duplicate rows, different attribution windows or periods that cannot fairly be compared. State those problems before drawing conclusions.
   
   Business objective: [for example, generate qualified leads, online sales or awareness]
   Primary success measure: [metric and definition]
   Conversion and attribution window: [details]
   Campaign dates: [dates]
   Budget and currency: [details]
   Target audience and location: [details]
   Important changes during the period: [creative, landing page, offer, targeting or tracking changes]
   
   Data:
   [Paste the spreadsheet export or table here]
   
   Return:
   1. A concise data-quality check.
   2. A table ranking campaigns or adverts by the primary success measure, with spend, reach or impressions, clicks, conversions and any relevant rates copied or calculated from the data.
   3. The largest reliable differences between campaigns, including the absolute figures behind each comparison.
   4. Separate observations from hypotheses. Do not claim that one change caused a result unless the data supports that conclusion.
   5. Three practical next tests, each with a reason, what to change, what to keep constant and which metric will decide the test.
   6. A short summary for a colleague in plain UK English.
   
   Show the formula for every calculated rate and ask me a question where the data does not support a safe conclusion.

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