Systems over tactics
A tactic decays. A system compounds. Every engagement ends with infrastructure the client keeps.
I do not run campaigns. I build systems that make campaigns inevitable.
Across a decade of performance marketing, I moved from managing ad accounts to architecting the full growth stack: acquisition, tracking, data, automation, and the AI layer that keeps it all learning.
My work sits at the intersection of three disciplines — media buying that respects unit economics, engineering that respects data, and leadership that respects people.
A tactic decays. A system compounds. Every engagement ends with infrastructure the client keeps.
Server-side tracking, clean attribution, and a single source of truth before a single dollar is spent.
If a human does it twice a week, it becomes a workflow. If it needs judgement, it becomes an AI agent.
Teams move fast when the target, the metric, and the owner are unmistakable.
From a junior seat in Cairo in 2018 to architecting AI-native growth systems. Every year is a chapter — companies, projects, certificates, milestones and the lessons that changed the direction.
A junior seat, an internet connection, and too many questions.
The starting point was not a strategy deck. It was a small studio in Cairo, a shared desk and a mandate to make websites load faster and forms actually convert. Everything since was built on habits formed here: read the data before the opinion, ship, measure, repeat.
Fundamentals of search auctions, keyword strategy, quality score and bidding.
Why it mattered: It was the credential that got permission to touch a live ad account with real client money.
What it changed: Moved the role from 'the person who builds pages' to 'the person who can spend budget'.
Do the work nobody claims. It is where the leverage hides.
From executing tasks to owning outcomes.
The year the job changed shape: full account ownership, direct client calls and a first real failure — a retail account lost after scaling spend without fixing attribution. That failure became the origin of every measurement obsession that followed.
Measurement planning, attribution models, goal and funnel configuration.
Why it mattered: It arrived directly after the attribution failure — theory meeting a scar.
What it changed: Reframed the job from buying media to designing measurement.
Lifecycle stages, nurturing and content-driven demand generation.
Why it mattered: Introduced the idea that acquisition without retention is a leaking bucket.
What it changed: Started building post-conversion flows, not just click-to-form funnels.
Trustworthy data beats clever tactics every single time.
The year every business needed a digital engine — immediately.
Lockdowns turned digital from a channel into survival. Brands with no ecommerce needed one in weeks. This forced a shift from campaign thinking to system thinking: catalogs, feeds, lifecycle flows and automation, all built at speed.
Segmentation, lifecycle flows and owned-channel revenue architecture.
Why it mattered: Owned channels became the margin cushion when paid costs spiked.
What it changed: Made retention a first-class part of every growth plan proposed since.
Store architecture, catalog structure and checkout optimisation.
Why it mattered: Understanding the store meant fixing conversion before buying more traffic.
What it changed: Ended the habit of blaming the ad account for a storefront problem.
Automate the repeatable before you scale the spend.
The first team — and the first lesson in leverage through others.
Promotion to manager meant giving away the work that made the reputation. The hardest transition of the decade: from being the best executor in the room to being the reason other people execute well.
Advanced auction dynamics, account structure and incrementality testing.
Why it mattered: Formalised the intuition built across thousands of live campaign hours.
What it changed: Gave the vocabulary needed to defend budget decisions to executives.
Give away the work that made you valuable. That is how leverage begins.
Dubai. Bigger budgets, harder markets, no room for guessing.
Moving into a regional head-of-performance role meant portfolio thinking: real estate, healthcare and ecommerce running simultaneously across markets, each with different economics, regulations and buying cycles.
Event-based measurement, GA4 modelling and cross-platform reporting.
Why it mattered: The industry's measurement foundation changed; leading it required fluency first.
What it changed: Made migration leadership a service the agency could sell.
Consistency at scale outperforms brilliance in isolation.
Automation stopped being a shortcut and became the architecture.
The bottleneck was no longer media buying — it was everything around it: reporting, QA, briefing, routing, follow-up. This was the year workflows replaced people-hours and the first language models entered production work.
Server containers, data enrichment and first-party measurement design.
Why it mattered: Signal loss made server-side the only reliable measurement path.
What it changed: Became the technical differentiator in every new business conversation.
Workflow orchestration, API integration and error handling at scale.
Why it mattered: Turned ideas about automation into production infrastructure.
What it changed: Shifted the identity from marketer to systems builder.
Automate the workflow, supervise the judgement.
TwinAI — where marketing, data and AI became one machine.
Stepping into a product company changed the horizon. Not client campaigns, but an owned growth engine: acquisition, tracking, automation and an agent layer, all built to compound without adding people.
Enhanced conversions, modelled attribution and privacy-safe measurement design.
Why it mattered: Measurement credibility is the entry ticket to any serious growth conversation.
What it changed: Positioned measurement architecture as the core offer, not a technical add-on.
Build the infrastructure; the campaigns become inevitable.
Systems shipping systems.
With the architecture in place, the work shifted upward: productising playbooks, scaling the agent layer, and turning growth infrastructure into something teams in other markets could adopt without a rebuild.
Agent design, evaluation harnesses, retrieval and guardrail architecture.
Why it mattered: Production AI demands evaluation discipline, not prompt intuition.
What it changed: Made AI a measurable engineering practice inside the growth stack.
If it needs you to run, it is not a system yet.
Growth architecture as a practice.
Today the work is architectural: designing the growth machine, the measurement spine and the agent layer that keeps both learning — and building the platform you are currently exploring.
Architecture is the highest-leverage form of marketing.
The next system has not been built yet.
What comes next: autonomous growth infrastructure, deeper AI-native measurement, and teams operating growth machines instead of campaigns. The chapter stays open — deliberately.
Keep the last chapter open.
Demand, media and the economics of attention.
3 worlds orbiting
← → travel galaxies · ↑ ↓ focus a sphere · Enter to fly in
Challenge, research, strategy, execution, tracking, optimization, results and the lessons that followed.
Executive dashboard
Every number below is derived from the same content database that powers the timeline and the project universe. Filter it the way an operator would.
delivered end to end
SMB to enterprise
verticals with delivered work
MENA, GCC, Europe, Americas
since 2018, unbroken
across seven ad platforms
conversion-first builds
agents, pipelines, copilots
workflows running unattended
logged delivery hours
academies designed and taught
departments, freelancers, partners
I build teams the same way I build systems: clear inputs, visible metrics, and no ambiguity about ownership.
Hire for judgement, train for tooling.
A metric without an owner is decoration.
Documentation is the highest-leverage management act.
Protect the team's focus like it is the budget.
Every system below runs in production, with a human where judgement matters.
Scores and enriches inbound leads in real time, routes them by intent, and writes the first-touch message.
Reads warehouse data weekly and produces the human explanation behind the numbers, not just the numbers.
Turns winning-asset patterns into structured briefs for the creative pod, with hook variants included.
Parses tenders, contracts and RFPs into structured records with scoring and draft responses.
Answers from product documentation with citation, escalating anything below a confidence threshold.
Monitors tracking integrity and alerts before a broken tag becomes a broken month.
The thinking behind the systems — available so a team can run them without me.
How to reconcile platform-reported conversions against CRM revenue and decide which number to trust.
Why creative fatigue is a supply chain constraint, and how to size production against burn rate.
A consent-aware, first-party event architecture that survives browser and privacy changes.
The deployment pattern that gets AI systems adopted by teams instead of quietly abandoned.
Pacing media against sales floor capacity so volume never outruns the ability to close it.
The exact sequence I run when taking over an unfamiliar growth stack.
Four engagements. Each one ends with infrastructure the team owns.
The full stack: acquisition, measurement, automation and reporting designed as one machine you keep.
Fractional direction for teams spending six to seven figures who need discipline, not more dashboards.
Agents and pipelines that remove repetitive operations and give judgement work back to humans.
Data-driven websites and landing systems built for speed, measurement and iteration.
Whether it is a growth stack to rebuild, a team to lead, or an AI layer to ship — start with a conversation.
Cairo · Dubai · Remote