12 February 2026

This AI Workflow Replaces 10 Hours of Ad Research

2Frameworks

9Insights

Frameworks in this episode

Strategy5 steps

Automation Quality Boundary

Automate structured AI strengths while keeping humans inside fragile creative work.

Read →

Marketing6 steps

Trend-to-Brief AI Workflow

Turn emerging social trends into data-backed ad briefs in minutes.

Read →

Insights & moments

The myth-busts, hot takes, explainers, and tools worth keeping.

Myth Buster· 1

Myth Buster20:30

An AI Workflow Is Not Automatically an AI Agent

Mike separates autonomous agents from predetermined AI automations. A workflow executes a known sequence of AI-assisted steps, while an agent is expected to make decisions more independently; confusing the two can inflate expectations without adding business value.

Agents and automations have different decision authority.
Predetermined workflows execute known steps in order.
Agent hype does not guarantee useful business outcomes.
A conventional workflow may be the more reliable solution.

“I think there is a big distinction between like an AI agent, which is a very hot thing right now.”
Mike Foutia · 21:30

“what I you know an AI automation right which is more predeterministic right it's like we we have certain steps that we want the AI…”
Mike Foutia · 22:00

#ai agents#automation#workflows

Hot Take· 2

Hot Take18:00

AI Finds the Average, but Great Marketing Must Escape It

Kip cautions that AI is naturally effective at reproducing established patterns and moving work toward an average result. Distinctive marketing still depends on experienced humans who can recognize when to reject the mean and create something surprising.

AI efficiently reproduces common patterns.
Average output can be competent without being distinctive.
Marketing performance often requires deliberate deviation.
Human experience and creativity remain differentiators.

“AI is really good at like getting you to the mean, getting you to like what average looks like.”
Kip · 18:00

“In marketing, it's about deviating from the mean and how you actually stand out.”
Kip · 18:30

#creativity#differentiation#human judgment

Hot Take24:00

Vibe Coding May Be Easier Than Visual Workflow Builders

Mike believes the future of internal AI tools may shift away from sprawling node-based canvases toward conversational coding environments. For ordinary marketers, describing and iterating on a focused tool in Claude Code can be easier than mastering a complex workflow builder.

Visual workflow builders can become technically demanding.
Conversational coding tools reduce the barrier to custom software.
Internal tools can target one team's exact pain point.
n8n remains powerful for technically proficient users.

“I think there's like a turning point going on right now where, you know, is this really the future of building AI like these scraggly…”
Mike Foutia · 24:30

“I think it's way easier.”
Kip · 24:00

#vibe coding#claude code#internal tools

Explainer· 1

Explainer02:30

Why Meta Advertisers Need a Constant Supply of New Concepts

Mike explains that paid-social performance increasingly depends on creative volume rather than relying on a few enduring ads. Teams need a steady stream of new hooks, angles, and concepts, making faster creative research commercially valuable.

Meta advertisers need frequent net-new concepts.
Organic TikTok trends can supply hooks for rapid paid testing.
Research speed matters because creative demand is continuous.

“the big thing in Facebook ads right now is creative volume, testing new hooks, testing new angles, putting out a lot of ads.”
Mike Foutia · 02:30

#meta ads#creative volume#paid social

Story· 1

Story22:30

It Took Mike Months to Feel Comfortable Building in n8n

Mike describes initially avoiding n8n because its API keys, webhooks, and technical workflow concepts felt intimidating to a non-coder. He eventually spent about four months building flows with AI assistance before becoming comfortable, illustrating the platform's power and learning cost.

Mike began without a coding background.
API keys and webhooks made n8n intimidating.
He initially used less technical alternatives.
Practical competence took months of building and debugging.
ChatGPT and Claude supported the learning process.

“when I first started with let's call it AI automation I was terrified of NAN because I'm not a coder terrified of it”
Mike Foutia · 23:00

“It took me I don't know four months of learn like truly building these types of flows.”
Mike Foutia · 23:30

#n8n#learning curve#no-code#career story

Tool· 1

Tool26:00

Record an Annoying Task and Let Gemini Map the Workflow

Kip suggests recording yourself performing a repetitive task and giving the video to Gemini for analysis. Because Gemini can consume a long video, it can identify the steps, reason about the process, and suggest which portions could be simplified or eliminated.

Screen-record a repetitive task in full.
Give the recording to Gemini for process analysis.
Ask the model to enumerate the steps.
Use the analysis to identify automation opportunities.

“you can literally just record yourself doing something you hate.”
Kip · 26:00

“It can watch videos up to an hour long. So, it'll just watch and break down everything and then reason through what you can do.”
Kip · 26:30

#gemini#process mapping#screen recording#automation ideas

Takeaway· 3

Takeaway07:30

The Real AI Advantage Is Scaling an Expert Point of View

Kip argues that useful AI research depends on encoding an expert's judgment rather than merely collecting information. A well-designed prompt makes the expert's evaluation criteria accessible to the wider team without requiring that expert to inspect every asset manually.

Raw information collection is not the main advantage.
Expert criteria determine which details the AI should extract.
Prompts can distribute specialized judgment across a team.
Point-of-view-driven analysis is more useful than generic summaries.

“And I think that's part of the modern AI process is like how do you scale expertise?”
Kip · 07:30

“you can basically scale what you think is the optimal process just for you in in doing all this”
Kip · 08:00

#expertise#prompting#team enablement

Takeaway27:30

Start With the Business Pain, Not the AI Technology

Mike's path into e-commerce automation accelerated when he understood the recurring problems customers were trying to solve. Domain knowledge makes it easier to choose useful inputs, design relevant outputs, and build tools that create genuine business value.

Technology alone does not define a useful product.
Customer pain points reveal worthwhile use cases.
Specialization emerged from inbound e-commerce demand.
Business understanding improves solution design.

“understanding the business case like what pain points do they have um will help you build these solutions a lot easier because you understand what…”
Mike Foutia · 27:30

#domain expertise#customer pain#product strategy

Takeaway28:00

Begin Vibe Coding With One API Call and a Small Problem

For beginners, Mike recommends skipping complex automation platforms and starting with a tiny, concrete coding task. He uses one AI coding environment to build and a conversational AI window as a teacher, debugger, and source of instructions.

Start with a single API call.
Choose a small problem that already consumes time.
Use conversational AI to explain and debug the build.
Avoid beginning with a large multi-system workflow.
Expand only after the first useful solution works.

“I would start learning how to build something very simple like just call an API, right? Like your first API call”
Mike Foutia · 28:00

“just start very small with like a small use case, something that's taking up a little bit of your time.”
Mike Foutia · 29:00

#vibe coding#api#learning#beginners