antrix.dev
Free Guide

DIY Competitor Intel

The manual approach to competitor research on Instagram. Learn the methodology — then decide if you want to automate it.

Reading this alone?

Do it with peers — bi-weekly calls, $29/mo founding.

Join →
1

Find your competitors

Most people think they know their competitors — they follow 5-10 accounts and call it done. Real competitor research starts with casting a wide net, then narrowing down.

You want creators in your niche who are 1-2 steps ahead of you — not the mega-accounts with millions of followers. Look for accounts with 10K-500K followers who are actively posting reels and getting engagement.

Action steps

  1. 1Search Instagram for your primary keyword (e.g., "AI tools", "solopreneur", "content creation")
  2. 2Check the "Suggested" accounts on 3-5 competitors you already follow
  3. 3Look at who your target audience follows — check the following lists of engaged commenters
  4. 4Save 20-30 handles into a spreadsheet with columns: handle, follower count, niche, avg views
  5. 5Prioritize accounts that post reels consistently (3+ per week) with visible engagement

What to look for:

  • Accounts that mix educational + personal content
  • Reels with 2x-10x their average views (viral outliers)
  • Comments that say “saved” or ask follow-up questions (high intent)
2

Extract their strategies

For each competitor, you need to analyze their last 9 reelssystematically. Don't just watch them — break them down into structured data.

For each reel, document: hook type (question, challenge, story, stat, controversy), format (talking-head, b-roll, screen record, tutorial), content pillar (education, motivation, personal story, tips), and performance (views, likes, comments).

Analysis Template

For each competitor reel, record:
- Handle: @username
- Reel URL: [link]
- Hook (first 3 seconds): [exact words]
- Hook Type: [question / challenge / story / stat / controversy]
- Format: [talking-head / b-roll / screen / tutorial / skit]
- Content Pillar: [education / motivation / personal / tips / behind-scenes]
- Views: [number]
- Likes: [number]
- Comments: [number]
- Engagement Rate: [likes + comments / views * 100]
- Key Takeaway: [one sentence — what made this work?]

Action steps

  1. 1Watch each reel twice — once for content, once for structure
  2. 2Write down the exact opening hook (word for word, first 3 seconds)
  3. 3Categorize the hook type and format
  4. 4Record views, likes, and comments
  5. 5Calculate engagement rate: (likes + comments) / views * 100
  6. 6Note any pattern — do their viral reels share a common hook type or format?

Reading this and want to actually do it with others?

The community is bi-weekly live calls, async accountability, and a group of builders shipping in parallel. Founding seats are $29/mo.

3

Identify patterns

Once you've documented 50+ reels across 10+ competitors, patterns emerge. This is where manual research gets powerful— you start seeing things tools can't tell you.

Sort your data by engagement rate. Look at the top 20% of reels. What do they have in common? Typically you'll find 2-3 hook types dominate, one format consistently outperforms, and certain content pillars get 3-5x the engagement of others.

Pattern Analysis Prompt

I have competitor research data for [X] Instagram creators in the [niche] space.

Here's the data: [paste your spreadsheet]

Analyze this data and tell me:
1. Which hook types get the highest average views?
2. Which format (talking-head, b-roll, etc.) performs best?
3. Which content pillars drive the most engagement?
4. What are the top 5 outlier reels and what makes them different?
5. Give me 10 specific techniques I can steal for my own content.

Action steps

  1. 1Sort all documented reels by engagement rate (highest first)
  2. 2Group the top 20% by hook type — which type dominates?
  3. 3Group the top 20% by format — which format dominates?
  4. 4Group the top 20% by content pillar — which pillar dominates?
  5. 5Identify 3-5 outlier reels (10x+ average engagement) and note exactly what they did differently
  6. 6Paste your data into Claude with the analysis prompt above for a structured breakdown
4

Build your playbook

Now turn your analysis into a replication playbook — a document that tells you exactly what to create, how to structure it, and which hooks to use.

Your playbook should include: your top 3 content pillars (ranked by competitor performance), 5-10 hook formulas proven to work in your niche, the format split (e.g., 60% talking-head, 30% tutorial, 10% b-roll), and a weekly production schedule.

Action steps

  1. 1Pick your top 3 content pillars based on competitor data
  2. 2Write 5-10 hook formulas adapted from the best-performing reels
  3. 3Define your format mix based on what works in your niche
  4. 4Create a weekly posting schedule (e.g., 5 reels/week, Mon-Fri)
  5. 5Draft your first 5 reel scripts using your hook formulas + top pillar
  6. 6Post, measure, and iterate — update your playbook monthly with new competitor data

This guide works. It's also slow.

The manual process above takes 20-40 hours per competitor batch. Watching reels, transcribing hooks by hand, building spreadsheets, calculating engagement rates, spotting patterns manually.

The Competitor Intel Complete Edition automates the entire pipeline. It downloads reels, transcribes audio with Whisper, calculates engagement, and runs the full strategy analysis — in under 15 minutes. Same methodology, zero manual work.

The Community — Coming Soon

For AI-first operators who actually build the thing.

Builders, consultants, founders, freelancers — anyone whose income comes from shipping real work, not from selling courses about shipping.

AI isn't going to make you money. Products do. Clients do. Real businesses do. But AI is the unfair advantage — when you wire it into how you build, sell, and serve, you move 10x faster than people still doing it by hand.

Who it's for

Beginners

Learning AI from people actually using it, not just teaching it.

Builders

Ready to automate workflows — yours or your clients'.

Operators

Senior, deep in production, looking for peers and leverage.

You have real skills. You want AI as an unfair advantage on real work — building something of your own, leveling up at your job, or both — not another set of videos to watch.

What's inside

Direct access — me + the network

Bi-weekly Google Meet calls where we actually work. DMs and posts with me and other operators when you're mid-build and stuck.

  • Bi-weekly live calls — builds, Q&A, teardowns
  • Architecture + payment reviews on your code
  • Peer network of operators who actually ship
  • Paid client referrals when I'm overbooked

The production agentic stack

Everything I run on real client work and my own products — installed in your environment, not just shown in a video.

  • Antrix OS — the control room for your agent fleet
  • Installers: Ship Your Product + Run the Machine
  • ax-* skill packs (audit, deliver, simplify, content, more)
  • Multi-agent code review across parallel worktrees
  • Scheduled jobs, overnight pipelines, dual-AI planning
  • Setup library: Stripe, Supabase, Vercel, Resend

Antrix OS — the agent dashboard

Antrix OS — agent fleet dashboard

The control room I actually run my work from — every project, agent, skill, schedule, machine, and secret in one place. Members get the codebase.

Why this one is different

Most hosts make content.
I make products.

The platform doesn't matter — Skool, Circle, Discord, Whop, whatever. The problem is the host.

Most paid-community founders aren't shipping anything outside of the community itself. Their whole business is selling you the community.

I'm in production every day. The people I want next to me are too.

We level up together.

Most communities

  • Revenue IS the membership fee
  • Main output: content about the community
  • You learn from someone who only sells communities
  • Growth metric: new members

This community

  • Revenue is client work + products. Community is a side of it.
  • Main output: shipped code, deployed products, served clients
  • You learn from someone in production every single day
  • Growth metric: what members are actually shipping

Waitlist

Open · first cohort

Get in before the price moves.

Founding pricing exists for the first cohort. When it fills, the next one pays more — and joins behind it. Names get pulled in order.

  • Founding price locked: $29/mo
  • Bi-weekly live calls + brainstorms
  • Direct access: me + operator peers

Work with me directly

Senior engineering, paired with agentic AI.

I take a handful of clients each quarter. Direct work on your codebase, your stack, your business. No agency. No junior handoffs. You hire me, you get me.

I build with you

Pair sessions where we ship the product, the agent, or the integration that's been blocking you.

I wire AI into your business

Agentic workflows, multi-agent pipelines, and scheduled jobs that turn AI from a chat window into 10x leverage on actual revenue work.

I review what you shipped

Architecture, AI usage, payments. I'll tell you what's worth keeping, what to rip out, and what to ship next.

I set you up, you take it from there

Sometimes you don't need me to run it. You need someone in the room while you do. I'll show you the system.

Specialty — payments

Hundreds of millions processed across subscriptions, marketplaces, and migrations. If your product touches money, that's where I help most. Webhook reliability, marketplace splits, billing architectures — the kind of work where shipping wrong costs you money, not just time.

Tell me what you're building.

I reply within 24 hours.