Nutricost
What each of four ad platforms actually drove
I built the measurement for a paid portfolio running about $2M a month across Amazon, Google, Meta, and TikTok. The biggest finding: Amazon Ads was taking credit for sales organic search would have brought in anyway.
Advertising Data Scientist · Nutricost · Aug 2024 – Feb 2025
martysmithson04@gmail.comLinkedIn (opens in a new tab)Resume PDFMcfly Analytics Shopify App (opens in a new tab)
Frameworks
I built statistical measurement frameworks covering Amazon, Google, Meta, and TikTok.
Incrementality
Separating Amazon's own claims from what organic search would have done anyway, so the ads got judged on what they added.
Contribution
MTA, cohort, and predictive analyses that sized each channel's contribution and guided spend decisions across teams.
Growth
For DTC sub-brands like Ketologic and Primaforce, I scaled daily spend once the data backed the expansion.
The measurement work
Four platforms, four different stories about what they drove. I built the statistical frameworks that sat across all of them and used incrementality testing to separate what each channel really did from what it reported.
The biggest finding was on Amazon. Amazon Ads was claiming sales organic search would have delivered anyway, so its own reporting overstated what the ads added. I also ran Google Ads campaigns that sent shoppers to our Amazon listings to push organic rank up. That spend pays off on a different platform, so its own ROAS can't judge it. The MTA, cohort, and predictive work put a size on each channel's contribution and fed spend decisions across teams.
Hands on the accounts
I also worked on day-to-day optimization for some of the ad accounts. Data first, and hands-on with the media when it was needed.
Scaling on evidence
For the DTC sub-brands, Ketologic and Primaforce among them, daily spend went up when the data said the expansion would hold.
Related