Preparing for Ad Launch
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
A wide-ranging practical stream combining a Burnish website and mobile review with discussions about product launches and AI-assisted work. Luqman examined content density, spacing, motion, input and CTA clarity, and visual identity, while the mobile audit, landing page, and some template regressions remained unfinished. On launch strategy, he argued against letting perfection delay shipping, but the product, lead funnel, email sequences, and ad creatives still required preparation and testing; conversion costs and projected outcomes therefore remained unknown. He also discussed lead qualification, install-to-paid measurement, entering the Shopify ecosystem after competitor research, and balancing report depth against speed. In audience Q&A, he shared personal views on Claude, Codex, Gemini, context compaction, multiple agents, and CLI workflows, stressing that tool choice is workflow-dependent and that these were not systematic comparisons. He warned against exposing API keys or selecting a cheaper provider without checking privacy and reliability. He advocated incremental learning through building rather than theory alone and framed AI as an amplifier of existing skills, not a replacement for human judgment. The conversation also covered education, freelancing, B2B versus B2C, unit economics, focused work and breaks, plus brief practical suggestions for preparing chicken meals.
159 moments from the stream
- 0:02:59
How Fear of Failure Delays a Project Launch
The speaker argues that working on a project for a long time can create fear of seeing the result, turning anxiety into repeated polishing, delayed market entry, or distraction by another idea. He presents possible failure as part of the journey rather than a reason not to start. He also uses baseline, upside, and downside scenarios to frame uncertainty, choosing to proceed while acknowledging that the undesirable downside remains possible.
- 0:07:29
Launching an Imperfect Website with Mobile First
The speaker treats the current version as a first attempt that can change and prioritizes mobile optimization before launch, based on his estimate that mobile accounts for most traffic. Content review and nonessential polish can be deferred so perfectionism does not block publishing.
- 0:09:14
Planning Ad Angles Across Different Landing Destinations
The speaker outlines a launch funnel using several ad creatives and distinct marketing angles: some lead directly to Shopify with an install-focused call to action, while others lead to a website page that asks for the visitor's Shopify store URL. The funnel system's own name is unclear in the captions.
- 0:10:18
A Pre-Launch Lead Funnel Built Around an Audit Report
A plan to exchange a useful PDF audit of a prospect’s website and catalog for an email address, then use a follow-up sequence to demonstrate product value and encourage installation. Pre-launch work includes mobile and desktop site readiness, the funnel, ad creatives, and email sequences; expected conversion performance remains a projection, not a tested result.
- 0:12:31
Filtering Temporary Email and Qualifying Leads
A discussion of reducing low-quality submissions by requiring a domain-based business email and a suitable store URL instead of relying on public or temporary email addresses. The aim is to avoid spending on unqualified requests, while recognizing that validation requires carefully designed gates and rules.
- 0:13:56
An Uncertain Note on Gulf Startup Support Programs
An exploratory mention of programs in Qatar, Dubai, and Saudi Arabia that may support startups relocating or expanding locally. The speaker explicitly said his understanding of the details was incomplete and suggested that some programs may expect operating maturity, such as existing traffic or revenue, before application.
- 0:16:27
Balancing Lead-Qualification Friction Against Cost per Lead
The temporary-email discussion continues with combined checks for a domain-based business email and a verifiable store URL, plus an optional email-domain match. The stated tradeoff is that more validation layers may improve lead quality but add friction and raise cost per lead; the actual cost was not yet known.
- 0:18:43
Fear of the Outcome as a Driver of Detail Fixation
A reflection that internal anxiety about the outcome or failure can pull attention toward minor details and delay progress. It is presented as a possible personal explanation, not a confirmed psychological diagnosis.
- 0:21:28
Beginner Boldness and the Dunning–Kruger Effect
An opinion that limited knowledge can encourage a beginner to enter a project boldly, while awareness of details and complexity may create hesitation; the speaker loosely relates this idea to the Dunning–Kruger effect.
- 0:22:55
Checking Mobile Chat Interface Inconsistencies
A review of why a chat icon appears in a local environment but not on the live version, followed by checking the dashboard on a phone and opening a link to identify remaining issues. Some dashboard work was described as completed, while other elements still needed adjustment.
- 0:28:51
The Idea of Graph Engineering for AI Agents
A brief exchange about a concept described as building a graph-shaped network of AI agents whose roles resemble those of employees. The exact term itself remains uncertain in the captions.
- 0:29:31
Guiding AI with Question-Based Learning and Criteria
A suggestion to challenge an AI assistant with critical questions and give it a Markdown file so it can learn from those questions and understand the user's preferences. The discussion also stresses judging ideas against explicit criteria rather than accepting suggestions out of excitement.
- 0:30:35
An Opinion on Building Businesses That Serve AI
The speaker relays another person’s view that businesses serving AI directly or indirectly may be a long-term opportunity, based on the belief that AI will remain influential.
- 0:31:46
Adding a GitHub-Hosted Skill to a Coding Agent
The speaker explains that an AI coding agent can be given a GitHub repository link and asked to inspect the skill, explain its usefulness, and install it into a repository. Alternatively, the README’s installation instructions and a documented CLI command such as NPX can be followed.
- 0:34:29
Turning Internal E-commerce Tools into SaaS and Open Source
The discussion describes a pattern where e-commerce companies branch out by turning solutions built for internal needs into SaaS products or open-source projects for other stores.
- 0:36:39
Open Debate About Changes in Software Engineering
The speaker welcomes discussion with people who disagree with ideas he has discussed, while acknowledging that claims about coding and software engineering changing can be sensitive. He notes that the argument is not originally his and remains open to multiple viewpoints.
- 0:38:04
Scoring Ideas and Narrowing the Shortlist
The speaker discusses using a scoring function or framework to rank ideas, then narrowing a large list into a smaller shortlist for review against several criteria, including unit economics. Some criterion names are unclear in the captions.
- 0:39:10
Rocket Money’s Model for Detecting Excess Spending
The speaker describes Rocket Money as a consumer app that reviews spending, identifies duplicate or excessive expenses, and suggests savings opportunities. He characterizes the space as competitive and expresses uncertainty about whether a similar model would suit the Egyptian market.
- 0:42:16
Investigating Sections Hidden on Mobile
A screenshot showed several sections hidden or blocked and not working on mobile, prompting a review of how they display on mobile screens; the number of sections and the beginning of the statement are unclear in the captions.
- 0:42:47
When an AI Task Board Becomes a Maintenance Burden
He explains that the task board or “command center” was useful for building a new project with agents and a Kanban board, but he set it aside because continuing to develop it to keep pace with rapid AI-tool changes during the three months before the recording would consume time needed for core work.
Got a question about something in this stream?
Ask in the group