What's up, Marketers! This is Aazar.

This newsletter is about leveling up your creative strategy and media buying skills by analyzing the best brands' paid strategy, tactics, positioning, and value props.

This newsletter is divided into:

I built these skills for myself. My first instinct was to keep them private, use them for my own clients, and leave it at that.

But the output is too good to sit on. And at the end of the day, AI doesn’t have a good judgment, experience, and taste to compete with me.

If you did this manually, here is what it looks like: 15 browser tabs, Amazon reviews, Reddit threads, YouTube notes, everything dumped into a Google Doc, then an attempt to make sense of it all.

It takes four to five hours minimum. Closer to seven if you want it done properly. And you would still likely miss the most important part: the content gap.

I did it in under 30 minutes. The full output is at the bottom of this email.

Here is exactly what I ran, using one product as the example: golden milk latte.

Note: Please understand that I had to do it quickly for the sake of the example, so I haven't given a detailed prompt to the AI while using these skills. However, the more detailed your prompt is about the product and the inputs, the better the results will be for you.

Also, I created this with Fable, but then they banned so it still works with Opus, it would just be less automated. If you don’t 100% get it to work, let me know. I’ll turn into a free training for all.

Let us get into them one by one.

Topic: What verified buyers say about golden milk lattes and the closest competitors

Skill: Review Researcher

Inputs Required:

Output: 60+ tagged reviews from Amazon and Trustpilot, sorted by pain point, desire, objection, and transformation story. Every entry includes the star rating and a direct source link.

Here's a link to the detailed output of the review skill (for one of my brands).

Here’s what I think about this:

Topic: Raw community sentiment around turmeric supplements, wellness drinks, and morning routine identity

Skill: Reddit Researcher

Inputs Required:

Output: Subreddit analysis with verbatim quotes, upvote counts, a pain point map, and an objection bank.

Here’s a link to the detailed output of the Reddit research skill.

Commentary: 

Thanks to my partners, Atria, they perfectly fill the gap of AI automation because I am as short on these skills and Atria does better.

Claude is still not intelligent enough to understand your account and campaigns, watch videos, figure out an audience from your ads, so Atria fits here perfectly imo.

You can now build on the open Model Context Protocol, the Atria MCP Server works with Claude, ChatGPT and any other MCP-compatible AI assistant. Once connected, you can work with your advertising data conversationally:

Start your free trial.

Getting started takes under a minute — follow the quickstart guide: https://docs.tryatria.com/docs/mcp-quickstart

Topic: Creator hooks and comment sections for turmeric, golden milk, and anti-inflammatory content

Skill: YouTube Ad Researcher

Inputs Required:

Output: 15 videos analyzed, 150+ comments extracted and tagged, and a table of high-value channels to monitor.

Here’s a link to a detailed output of YouTube research skill.

Commentary: 

(This was my favorite for the deep research cowork did)

Topic: What is going viral right now in the chai latte (or golden Milk) space on Instagram

Skill: Social Trend Researcher

Inputs Required:

Note: For this example, I selected Quick Research. If you want, you could do deep research with this skill; it would analyze 40-60 pieces of content.

Output: Viral content analysis with hooks, format types, engagement numbers, and an audience language bank from top comments.

Here’s a link to a detailed output of Instagram research skill.

Commentary:

The four steps above collect the data. Now, these three steps turn it into something you can actually use.

Skill: Content Gap Analysis

Inputs Required:

Output: Priority 1, 2, and 3 gaps, each with a creative brief, source evidence, verbatim quotes, and a draft hook.

Here’s the detailed content gap analysis.

Commentary:

This comes directly out of the gap analysis. Every gap is ranked by how many sources confirmed it. Highest confidence gaps go first. No prioritization required on your end.

Skill: Ad Scripter

Inputs Required:

Output: Full script in labelled, timed sections. Each angle is tagged to its research source and the objection it pre-empts.

Here’s the link to the ad scripts

Commentary:

Here is a link to the complete pipeline skill, which takes the necessary inputs and does all the work for you in one go.

Before you use the skill, however, you will have to complete all the research and then try it.

Please reply to this email if you really tried, and like this, it took me 30+ hours to come up with this full flow to help you. Let me know if a video training would help you to make full use of this workflow.

Happy Growing with Paid Social,

Aazar Shad

Since this newsletter is free, I do it to follow my curiosity. But I’d love it if you could leave some feedback so I know if I am helping you or not.