I spent 4 years at Tesla building predictive models and some of the company’s first LLM apps for Sales, Delivery, and Service. My last tour was leading the Tesla’s Marketing Data Science team, where the framework I built steered every dollar of ad spend and delivered a steady 1.5× ROI.
Now I’m taking that playbook to performance marketing with an ex-Tesla crew - dropping an army of AI marketing scientists into every growth stack so each dollar works harder. Early users are already seeing 14× returns on spend our system directs.
I usually keep things low-heat, but credibility matters, so I’m turning up the heat while keeping the signal high.
Here’s what I’ll share:
- how our systems reasons and works
- wins, misses, and lessons learned
- my take on the economics of marketing
DMs are open - if my future content resonates or you want to join us.
Install the BlueAlpha Marketing Plugin for Google Ads in Cowork or Claude Code, connect your accounts via MCP, and run your first live audit — all in under two minutes of setup.
The plugin packages ten agent skills that connect directly to the Google Ads API: account audit, budget reallocation, creative fatigue detection, competitive conquest, geo expansion, incrementality test design, and more. Each skill follows the same loop: Analyze the account, produce a recommendation with the data behind it, then execute approved changes on-platform.
This is what BlueAlpha calls the Decision Engine — the system that turns every signal in your ad account into a real decision. Not a dashboard. Not a reporting layer. A team of agents that does the analytical work so your growth lead can focus on the call, not the dig.
📌 What's covered:
0:00 – Intro
0:13 – Install in Cowork (drag-and-drop from GitHub)
1:00 – Install in Claude Code (two slash commands)
2:27 – Connecting the BlueAlpha MCP to your Google Ads accounts
3:30 – Running auto-optimize on a live account
5:04 – MCC support: running across multiple accounts
6:08 – Manager vs child account routing
6:40 – Audit results + markdown report output
7:06 – Other skills you can run
🔗 Links:
Full skill reference + copy-paste prompts → https://t.co/M9FIARE1HA
Setup documentation → https://t.co/NHSq7FYe0n
Plugin download (GitHub) → https://t.co/pzc4OVsSwY
Book a demo → https://t.co/ev759lnGrW
The BlueAlpha Marketing Plugin works inside Cowork and Claude Code. One MCP connector, one sign-in, no developer tokens, no OAuth configuration.
#GoogleAds #Claude #MCP #MarketingAutomation #BlueAlpha #GoogleAdsMCP #AIMarketing
@shannholmberg Correct diagnosis. The missing piece: even when the data is connected, most marketing AI layers are reading correlational outputs and calling them insights. The knowledge layer needs causal grounding or it just gives you a faster version of the same wrong answer.
@saroshws Clean infrastructure still gives you correlational outputs. The measurement infrastructure can be perfect and the attribution model still cannot answer the counterfactual. What would have happened without that touchpoint? That is the question clean data alone cannot answer.
I want ONE engineer at Meta to prove that these suggested budget changes are based in data
Doubling a campaign budget from $600 to $1100 in one day is only going to raise my CPA by 13%?
Show me aggregate advertiser data that supports these projections and I will go live as a sheep herder in New Zealand forever
@helloitsaustin Correct. The 45-skill claim is the same promise as the 47-tool martech stack - maximum surface area, minimum depth.
What matters is whether one skill does the right thing reliably. Most agent demos have not survived their first production workload.
@weird_ceo Correct direction, wrong framing.
MCP is table stakes. The question is what you expose through it. A well-structured MCP server for a measurement platform with clean priors and calibrated models is a moat. An MCP wrapper around a ROAS dashboard is a slightly fancier dashboard
@weird_ceo The dashboard was never the product.
The decision was the product.
The dashboard was just the most expensive way to avoid making one.
If mobile removes that excuse, good."
Kahneman documented the mechanism in 1974. Measurement outputs are System 2 products: statistical, probabilistic, abstract. Marketing decisions are System 1 products: intuitive, narrative, immediate. Data does not update narratives. Narratives filter data. A CMO who has spent two years defending a channel allocation has a story about why it works. Your measurement report is not going to defeat that story. It is going to be absorbed by it.
Our CEO Peter joined the Numbers & Narratives podcast to break down why most brands are overindexed on search, how one of our clients cut Meta spend 50% with no drop in KPIs, and what's actually working in marketing measurement right now.
🎧 https://t.co/Z1azC0Ihjw