Gus Dantas

Gus Dantas, retail media measurement

About

Gus Dantas

Eight years in marketing analytics in Australia, most of it agency side, across media strategy, reporting and measurement for advertisers. Retail media came later, and it is where the two sides met.

At PetAds I build the measurement framework. Off site, on site and in store all report into one place, so the network gets measured as one system, overlaid by transactional data for real incrementality. What counts as results is decided by the framework, not by the ad servers' last click or attributed ROAS.

The newer part is agentic. I build agents and MCP integrations that read the ad platforms directly, implement the honest measurement framework, assemble the post campaign reporting automatically and unlock real insights for the suppliers and categories. The network gets sharper, each time it runs.

  1. 2018 Switched On
  2. 2020 AKQA Media
  3. 2022 The Media Precinct
  4. 2025 oOh!media
  5. 2026 Petbarn PetAds

Work

Microsoft FabricOneLakeIncrementality

This is where PetAds’ reporting comes from. Campaign data, sales and audiences land in one place, and the framework decides what counts as incremental before anything is reported. So a supplier gets one number across onsite, offsite and in store, and it comes off actual sales rather than off what the ad server counted. It runs on Microsoft Fabric and OneLake, and we built it instead of buying one.

This work involves retailer transaction data, so the figures stay private. The architecture is described above.

ClaudeMCPAgent orchestrationAd platform APIs

This is the layer that sits on top of the data. MCP connects agents straight to the ad platforms and to the measurement framework, so they read the same numbers the framework does. A campaign can be watched while it runs rather than explained afterwards. Questions about a category or a supplier get answered when they come up. The reporting assembles itself. I check anything that goes out.

PythonRSQLTableauPower BIRegressionClusteringSegmentation

From the agency years, before retail media. Here for context, not as what I do now.

  • BNPL revenue forecast Afterpay
  • Mortgage customer segmentation Aussie Home Loans
  • Scenario planning tool Aussie Home Loans
  • Cart abandonment clustering Officeworks
  • Creative performance a global FMCG portfolio

Try my RMN MCP

Try me on your LLM.

Paste the message below into ChatGPT, Gemini, Claude or whatever you use.

Read https://ghostavo.com/llms.txt, and connect to https://mcp.ghostavo.com/mcp too if you speak MCP. How would Gus’s measurement framework change my retail media network?