Funneling, sequencing, & Going to Market
Want to back the stream? The «داعم» membership on YouTube is EGP 60 a month: badges, emoji, and I answer your questions first on stream. Become a supporter
The gist
The stream centers on preparing a complete go-to-market system before increasing ad spend: reviewing the website, landing and pricing pages, prioritizing mobile UX, and building conversion funnels, email programs, and a unified metrics dashboard. Practical work covers brand consistency, motion, conversion sections, UI and background-job diagnosis, coordination between coding-agent sessions, and choosing AI tools based on quality, price, and usage. Audience discussions address learning marketing and sales through practice, pricing and trials, SEO and merchant visibility in AI answers, testing Google and Meta ads alongside email and SMS, and the importance of margins, supply chain, and early seasonal preparation. The speaker repeatedly stresses that marketing outcomes, pricing, and automation require measurement, market testing, and real user feedback. Several matters remain unresolved, including recurring failures, pricing changes, tool performance, and completion of the website and marketing system.
103 moments from the stream
- 3:23:37
Adapting to an Intensive Routine
The speaker suggested that a strict early schedule and an extended period without phone or internet could positively affect habits and character, while noting that the first two or three weeks may be difficult before adaptation.
- 3:24:55
Refining a Pricing-Tier Section
A practical review of reducing copy density across headings and tier descriptions, tightening alignment and grid geometry, removing overlap, and improving the placement and width of AI logos. The speaker wanted to delegate the task first and see what the agent could produce independently.
- 3:27:26
File Organization and Agent Context Usage
The speaker discussed whether long files and a tangled structure make a coding agent read more code and exhaust its context sooner. He agreed that organization and cleanup could improve, but kept the cause uncertain because context and token complaints are widespread and performance varies even when his workflow stays the same.
- 3:30:59
Skepticism About Model Performance and Benchmarks
The speaker noted favorable reports about DeepSeek but expressed dissatisfaction with another model whose exact name is unclear in the captions, questioning shared benchmarks after it failed to solve a seemingly simple problem. These were personal impressions, not a controlled evaluation or verified benchmark result.
- 3:32:05
Asking an Agent for an Inspectable Visual Rundown
The speaker asked the agent to turn its explanation into a tangible, easy-to-scan format—such as a table, graph, or visual widget—so the situation could be understood clearly instead of through a verbose description.
- 3:33:22
Switching Tools When a Critical Fix Stalls
After an important application bug persisted despite repeated repair attempts, the speaker expressed frustration and considered moving the task to another tool. The captions do not establish the bug’s cause or whether switching tools succeeded.
- 3:37:25
The Complexity of Code-Based Email Marketing
The speaker discusses moving from working with specialized platforms and providers to implementing email marketing directly in code with Resend. He describes the work as complex while expressing the view that it may be better in the long term.
- 3:40:30
Copying and Adapting Web Design Elements with AI
The speaker argues that AI generally does not produce strong design ideas on its own. He describes taking elements he likes from other websites, using particular prompting to get past the tool’s copyright objection and implement them, then adapting them with his own content, branding, and details. This reports the speaker’s opinion and practice without asserting that it is legally permissible.
- 3:43:57
Reviewing Pattern and Alignment During Design
While reviewing the design, the speaker notes that the circular elements are not positioned correctly at the intersections and suggests creating a pattern, mentioning an idea he wants to apply there.
- 3:45:04
AI-Assisted Landing and Pricing Interface Refinement
A comparison between an older visual approach using video backgrounds and effects and a newer design described as cleaner and more mature. The pricing-section review identified excessive spacing, incorrect measurements, and visual bleed, while noting that familiar pricing patterns may help users feel more comfortable.
- 3:50:21
Scoping a Separate Session for the Pricing Matrix and FAQ
A prompt was scoped for a separate session to reproduce a reference pricing-matrix structure while adapting the content, allowing limited additions, and closely understanding the reference design. The scope was expanded to include the FAQ section and its visual dividers.
- 3:53:36
Customer-Service Agents and Expansion into Marketing
The discussion argues that customer service has broad applicability because companies generally need it, then considers a wider vision for marketing agents beyond AI-system visibility. It is presented as a potential opportunity requiring substantial resources, not as a validated outcome.
- 3:55:56
Diagnosing Repeated Crashes Without Sacrificing the UI Effect
A request to move from reactive crash fixes to a historical, deductive analysis of why many crashes appeared after a stable period, followed by a concise explanation and durable remedy. A shortcut that removed the reveal effect was rejected; the effect was to be restored and a reference implementation examined to preserve the intended behavior.
- 4:10:47
Reviewing Website Sections and the Application Preview
A design review distinguishes successful sections from one whose problem is difficult to pinpoint. Showing the application’s interior is considered an important function, while another informative section is described as necessary and reasonably designed but possibly too muted in color, warranting a later revisit.
- 4:14:53
Checking Background-Effect Consistency Across Pages
The start of a review of the footer and background effects asks whether the home page uses a different effect from other pages, including a golden effect. The excerpt raises the verification question but does not include its answer.
- 4:16:20
Aligning the Pricing Hero with the Homepage
A discussion of creating visual continuity between the homepage and pricing-page heroes through related—but not identical—geometry, including a divider below the AI engine strip and reviewing ideas before implementation.
- 4:23:48
Simplifying Five Email-Marketing Streams
As the email-marketing work becomes complicated, the question is asked again and the five streams are requested in a concise, digestible format showing each stream’s step, day, content, and sender.
- 4:27:30
Price as a Factor in Trying an Upcoming Model
The speaker is enthusiastic about an upcoming model mainly because of its price and plans to try it with Codex Everywhere. While testing GPT-5.6, he notes that another model did not complete the same task, though the task details and some model identities are unclear in the captions.
- 4:32:13
Finding the Effort Level in Codex Everywhere
While using GPT-5.6 through Codex Everywhere, the speaker tries to determine the active effort level and whether it is low, high, or extra high. He finds that it is set to Medium and says he would prefer a different level.
- 4:35:24
Trying a Sketch-Led Implementation Idea
The speaker considers drawing his idea so the tool can implement it and notes that a similar attempt the previous day did not work. He then performs another step and reacts with surprise when the step finishes, without enough evidence that the full idea succeeded.
Got a question about something in this stream?
Ask in the group