Video summary
Give Me 27 Minutes and I'll Make You Disgustingly Good at Digital Marketing
Main summary
Key takeaways
Main ideas, concepts, and lessons
- If starting from zero: The speaker claims they would learn digital marketing by focusing on a small set of fundamentals rather than chasing gurus, courses, or tutorials.
- Two unchanging fundamentals: Digital marketing mastery comes down to:
- Capturing attention
- Converting that attention
- Core mindset: Markets and human psychology drive outcomes; technology changes but psychology doesn’t.
- Practical learning approach: Study proven ad formats (old direct-response ads and modern social-media ads), build “swipe files,” then run lots of split tests using measurable metrics.
- Traffic and conversion strategy: Use the right mix of demand capture (people already searching) and demand generation (you create/awaken demand).
- Ad creation philosophy: Make ads resemble the kinds of content people already consume (news/findings/gossip) and use a structured copy framework.
- Market awareness levels: People vary by how “aware” they are of their problem/solution; different awareness levels require different messaging styles.
- Funnel/copy emphasis: Most conversion impact comes from key funnel elements (especially headline, offer, subheadlines, and PS).
- Continuous improvement loop: Use hypotheses and continual split testing; treat each test as learning.
- Ultimate advice: Stay a lifelong student of markets; even “unrelated” ads can provide transferable insights.
Step-by-step methodology (as presented)
Step 1: Understand the foundation (capture + convert attention)
- Close your laptop and stop consuming “guru” content (no more YouTube tutorials/courses).
- Focus on mastering:
- Attention capture
- Conversion of attention
- View humans as driven by psychological triggers (“chimps in disguise”); understand those triggers to never worry about money (speaker’s framing).
Step 2: Capture attention (study traditional direct-response ads)
- Study old school print direct-response ads (newspapers/magazines from the 1960s–1980s).
- How to do it:
- Google image search for:
- “direct response magazine ads”
- “direct response print ads”
- Read 5–10 ads per day.
- Notice recurring ingredients:
- Big bold headline + unique hook
- Subheadline
- Native image
- Build a swipe file:
- Download ads into a folder
- Read them daily for 30 days
- Google image search for:
- Key concept: ads that look like editorial content increase readership.
Step 3: Convert attention (traffic types + where money is)
- If you can sell, you don’t need to be best at traffic.
- There are two types of traffic:
- Demand capture
- Target people searching already (speaker calls them “prospects with a bleeding neck”).
- Use Google as “lowest lying fruit.”
- Example math given:
- If ~100,000 searches/month and ~4% CTR → ~4,000 clicks/month.
- Limitation: limited search volume.
- Demand generation
- Place ads in front of a broader audience to:
- Capture attention
- Agitate a known or unknown problem
- Educate until they think: “Yes, that’s a problem.”
- Claim: “millions” are made here.
- Place ads in front of a broader audience to:
- Demand capture
- Recommendation if forced to choose: demand generation, because you can create demand.
Step 4: Study modern media (match ad style to what people consume)
- Use Meta’s quarterly reports to see what’s most shared/consumed (examples mentioned: TMZ, Lad Bible, fitness pages).
- Claim: top shared internet content categories are:
- News
- Findings
- Gossip
- Ad framework: ADA (A-D-A) / “ADA”
- A = Attention
- Make something attention-grabbing.
- D = Desire
- Build desire for the solution/offer (speaker uses a “flame the fans” metaphor).
- A = Action
- Tell people exactly what to do (lead gen or purchase).
- A = Attention
- Core execution idea: don’t copy “the summit” (big brands) style; look at what people do “on the climb” (smaller-scale reality).
Step 5: Build your industry swipe file (reverse-engineer active competitors)
- Create a fresh Instagram/Meta presence and engage with the niche so the algorithm serves relevant ads.
- What to do next:
- Identify ads that appear frequently.
- Screenshot each recurring ad into a folder.
- Visit the advertiser’s ads library:
- Track how long the ad has been running and whether it’s still active.
- Build a swipe file of ads running for long periods (implied: “printing money”).
- For each advertiser:
- Go through their landing pages
- Take full-page screenshots of funnels
- Save under a subfolder like “landing pages or funnel”
- Keep a Google doc of competitor names.
- Use AI tools (speaker mentions Claude/ChatGPT) to research the founder and estimate revenue.
- Focus on competitors estimated to be $3M–$100M revenue.
- Add their ads/funnels to your swipe file.
Step 6: Halo strategy (find pains/fears/desires quickly)
- “Halo strategy” = forum forage + detective work to map the market’s emotional drivers.
- Process:
- Find the “coliseum keyword” (the main high-traffic keyword in the niche).
- Example given: “digital marketing,” “protein powders.”
- Search on forums like Reddit and Quora for:
- emotionally charged complaints
- questions/search needs
- reviews of past purchases
- Populate the halo template (as described on screen):
- Pains
- Fears
- Desires
- Find the “coliseum keyword” (the main high-traffic keyword in the niche).
- Goal: learn the niche faster than most newcomers.
Step 7: Launch and learn (reps, static ads, split testing)
- To get good: get reps.
- Creative choice: static ads only (speaker’s recommendation).
- Daily cadence: write 5 new ads every day.
- Execution/testing approach:
- Use broad targeting (speaker claims Meta’s algorithm finds the right people).
- Suggested budget:
- Start around $20/day (about $140/week)
- Allows testing around 25 creatives
- Primary metrics (choose based on objective):
- Selling product: evaluate ROROWAZ (speaker’s best-performing metric term)
- Leads: evaluate lowest cost per lead
- Also track:
- CTR (highest click-through rate indicates a strong angle)
- AI warning: don’t blindly run AI-written ads—know what “good” looks like to avoid burning money.
- Output: continually learn from results.
Step 8: Market sophistication & competition continuum (match message to awareness level)
- Market awareness states (from easiest/smallest to hardest/biggest opportunity):
- Already sold / buying now (red ocean; most contested)
- Product/service aware
- Solution aware
- Problem aware
- Unconsciously unaware (biggest blue ocean; requires highest skill)
- Messaging preference by color/awareness:
- Red end (product/service/price):
- Responds to product, service, price, incentives/offers
- Orange (claims and proof):
- Uses differentiation, proof, hero mechanisms
- Blue (explanation & storytelling):
- Use narrative to move people from “who are you” to “shut up and take my money”
- Red end (product/service/price):
- Core rule: meet people where they are.
Step 9: Assemble copy with a “17-step secret selling system” (headline-first funnel construction)
- Speaker claims personal use in producing 2,000+ funnels and generating $7.8B in sales for clients.
- The system includes (at least) these steps:
- Call out your audience (write for the person responsible for ~80% of revenue)
- Create a headline complex
- Core principle: big bold benefit headline
- Example: “Lose 10 kgs in 6 weeks, guaranteed” (not generic “lose weight”)
- Subheadline
- Back up/expand/support the main benefit without giving the full picture
- Create irresistible intrigue
- Use novelty/loops; don’t reveal everything at once
- Shine a floodlight on their problem
- Emphasize pain vs pleasure motivation
- Reveal the solution
- Explain how it solves problems with existing alternatives
- Lay out credentials
- Proof of authority: clients, case studies, results, evidence
- Detail benefits
- Create a two-column list: features → corresponding benefits
- Emphasize benefits (not features)
- Social proof
- “Unquestionable proof” mindset (jury-convincing)
- Godfather offer (extremely hard to refuse) - Example: “Guarantee to increase revenue by 43% or you don’t pay.”
- Stack bonuses
- Sweet spot: 2–3 additional bonuses
- Stack the value
- Increase perceived value via an inflated anchor, then a smaller price (speaker’s “20%” logic)
- Reveal price
- “Value-based marketing” framing
- Example: free consult (30/45/60 minutes) with real value delivered
- Inject scarcity
- Create urgency (e.g., limited stock)
- Claim: can raise conversions by 30%+
- Powerful guarantee
- Reduce perceived risk to increase conversions and lower CAC
- Insert call to action
- Very clear, forensic-level instruction (no interpretation needed)
- Add a PS
- “TLDR” style summary including risk reversal; acts as final prompt
- Additional emphasis claim: key elements carrying ~95% of conversion “freight”:
- Headline
- Subheadline (headline complex)
- Offer
- Subheadlines sprinkled throughout
- PS
Step 10: Dial in your funnel via split tests (hypothesize → test → iterate)
- Mention of copywriter performance:
- High-paid copywriters: ~30% hit rate
- Best guys: even after writing, ~70% failure
- Testing loop:
- Track metrics:
- CTR
- CPL
- ROROWAZ
- time on site
- conversion rates
- Build hypotheses about what’s happening in the funnel.
- Run split tests against control.
- Repeat endlessly.
- Track metrics:
- Emotional/behavioral note:
- Speaker says they experience anxiety without at least five split tests running.
Step 11: Never stop learning (be a student of markets)
- Click on seemingly irrelevant ads because they can be market research.
- Transfer insights across industries.
- Core principle:
- When you think you’ve learned everything, you stop earning.
- Lifelong learning compounds skill and income.
- Personal motivation for making videos:
- Raised in a single-parent household with limited money.
- If abundance arrives, speaker would “send the elevator back down” by helping others.
Speakers / sources featured
People
- Speaker: Sabri Subub (referred to by name multiple times; author of the “17-step secret selling system”)
- Named industry figure: David Ogilvy (advertising leader mentioned as a key example)
Platforms, tools, and research sources
- Google Images
- Meta (including Meta’s quarterly content reports)
- Google (as a demand-capture channel)
- Meta Ads Library
- Claude (AI tool)
- ChatGPT (AI tool)
Web communities
- Quora
Other references
- Referenced Shark Tank tie-in: “Shark Tank dude” (no specific person named)
Rate this summary
Your feedback will help improve summaries.
Improve this summary
Reprocess with a stronger model when the summary feels incomplete or inaccurate.
Translate summary in another language
Ask questions to this video
Chat for follow-up questions, clarifications, and source-backed answers.