Get in touch
Open your email app

Creative Strategy · Performance · AI Systems

Lucca
Magnata

// Senior Creative Strategist · Performance Intelligence · AI Systems

I build the creative and conversion infrastructure that turns ad spend into documented revenue. B2B high-ticket and DTC eCommerce — narrative architecture, performance intelligence, and AI-augmented systems that compound over time.

29.79x
Peak ROAS
$312k+
Revenue Influenced
7%
CVR vs 2% avg
4.67x
Blended MER · 14mo
13+
Concurrent Clients
−50%
Testing Cycle Cut
View Case Studies
B2B & High-Ticket
Long-cycle, high-stakes conversion
Lead qualification, SQL nurturing, Dual-Layer architecture, CRM integration. Funnel: Ad → Lead → Nurture → Close. Metric that matters: MER and CVR over time.
DTC & eCommerce
Short-cycle, high-velocity creative
Hook engineering, UGC direction, fatigue management at scale, rapid iteration. Funnel: Ad → PDP → Purchase → LTV. Metric that matters: Hook Rate, ROAS by creative, cycle speed.
Both tracks ran concurrently Jul 2024 – Sep 2025 (14 months). Same methodology. Different commercial context. The cross-vertical range is documented, not claimed.
Operating Principles

Not positioning statements.
Principles from real failures.

On Performance
"Performance is not an algorithm problem. It is a narrative problem."
The algorithm distributes. The narrative converts. Most teams optimize distribution endlessly while the message itself is broken. I start with the story, then scale what works.
On Metrics
"Volume is vanity. Profit is sanity."
A CPL of $40 looks great until your sales team can't absorb 400 leads a month and 60% aren't qualified. I optimize for MER protection and operational coherence, not vanity numbers that collapse under scrutiny.
On Efficiency
"I pay more per click to pay less per customer."
High-friction filtering is the lever. Educational content before the sales conversation shifts CVR from the 2% industry standard to a documented 7%. Fewer leads. Better ones. Lower CAC.
On Scale
"The spike proves the concept. Consistency builds the bank."
The Dual-Layer Architecture separates these two goals deliberately. Group A hunts for the unicorn. Group B protects cash flow. Running them as one campaign is the most common and expensive mistake in performance marketing.
Architecture

Four layers.
One machine.

Most specialists go deep on one layer. I operate across all four simultaneously — because disconnected layers produce disconnected results. This applies equally in B2B high-ticket and DTC eCommerce.

01
Narrative Architecture
I build brand belief systems, not just messaging. Using Primal Branding frameworks: Creation Story, Creed, Icons, Rituals, Sacred Words — I create the psychological infrastructure that makes audiences feel they belong before they buy.
Primal BrandingHook EngineeringDirect ResponseBrand Voice
02
Performance Intelligence
CTR, Hold Rate, CPA, CVR, MER, ROAS: I don't just read metrics. I translate cold data into creative briefs. If a hook drops at 3 seconds, I know what the narrative problem is and how to fix it before the next sprint.
ROAS / MER AnalysisFatigue DetectionA/B Narrative Testing
03
Growth & Revenue Systems
Not top-of-funnel thinking. Full journey: MQL → SAL → Deal, or Ad → PDP → LTV. I design qualification logic, nurturing sequences, and reporting that connects every creative decision to a revenue line.
Funnel ArchitectureMQL → SQL LogicLTV StrategyConversion Ops
04
AI-Augmented Infrastructure
I don't use AI to replace thinking. I use it to remove the friction that slows thinking down. Automated pipelines for trend analysis, competitor research, and brief generation. The strategist makes decisions. The system does the paperwork.
Make / n8nOpenAI APIClaude / GeminiCRM Automation
B2B Case Studies

Not screenshots. Not one-week anomalies. These cases are structured to show how narrative, conversion architecture, and performance intelligence interact under real commercial pressure.

Silent
Revenue Leak

A system built on the psychology of invisible loss — naming what the market already felt but couldn't articulate, before selling anything. High-ticket B2B, healthcare vertical. 14 months of documented performance.

AD02 · B2B High-Ticket · Jul 2024 – Sep 2025
Silent Revenue Leak
// 14-Month Consolidated Case · Healthcare · High-Ticket B2B
Architecture
Dual-Layer Creative + Conversion System
1,340+
Total Leads · 14mo
$7.20
Avg CPL (USD)
$312k+
Revenue · 14mo
29.79x
Peak ROAS
4.67x
Blended MER · 14mo
7%
CVR vs 2% avg

The Silent Revenue Leak concept was built around a single insight: most B2B brands were losing revenue to invisible operational gaps — slow follow-up, broken funnels, unconverted leads sitting cold in a CRM. A problem every business has but nobody had named yet.

I translated that invisible problem into an emotionally legible metaphor: a bucket that fills but leaks before it can be spent. The metaphor worked because it gave a name to a pain the audience already felt but couldn't articulate. Naming the problem before selling the solution is the oldest and most underused move in direct response.

Over 14 months the system compounded: creative learnings from each sprint fed directly into the next brief cycle. The conversion architecture: 30-second reply window, dual-channel response, pre-recorded humanized audio, 5-day automated follow-up cadence, monthly reactivation loop, and CRM feedback into the ad platform, which remained stable while the creative layer kept optimizing. Months 1–4 built the playbook. Months 5–10 scaled it. Months 11–14 defended MER at scale while volume climbed.

Thesis: Volume is vanity. Profit is sanity. A cheap CPL is worthless if the operation can't absorb and convert the lead. A good creative system is worthless if the operational layer leaks every lead it generates.
Creative Group Split
Group A: Discovery Engine
Aggressive scale and variance across the full 14-month window. High emotional intensity, pattern-interrupt hooks. Blended MER 6.72x in peak sprints. Responsible for the peak ROAS spikes, seasonal revenue injections, and identifying the winning angles that fed Group B's stable layer.
Group B: Stability System
Qualified-intent creatives running continuously. Mini-VSLs, ROI calculators, educational pre-sell content. CPA sustained at $15–$24 across 14 months. ROAS 8.1x–9.3x. Built to sustain the pipeline without overloading the commercial team regardless of seasonal variance.
14-Month Arc · Phase Breakdown
Months 1–4: concept validation, conversion architecture build, first 200 qualified leads. Months 5–10: system scale, peak ROAS of 29.79x in month 7, volume growth from ~80 to ~120 leads/month. Months 11–14: MER defense at scale, reactivation loop producing 18–22% of monthly revenue from aged leads, CVR stabilized at 7% vs 2% category average.
The 14-month window is what separates a campaign from a system. Most agencies show you month 1. This is what the architecture looked like after it had time to compound.

The False
Demand Problem

A B2B case built on one contrarian diagnosis: the account did not need more leads. It needed protection from the wrong kind of demand. Operational transformation consultancy. 75 days.

AD03 · B2B Operational Transformation · 75 Days
The False Demand Problem
// Demand distortion diagnosis · Qualification-first architecture · B2B consultancy
Architecture
Qualification-First Funnel · Dual-Layer Intent Split
1,280
Total Leads · 75 Days
$14.37
Avg CPL (USD)
$511k
Revenue Generated
27.7x
Peak ROAS
13.9x
Stabilized ROAS
22%
SQL Rate · 282 SQLs

Most B2B teams don't have a demand problem. They have a filtering problem disguised as a growth problem.

The account looked healthy from the outside. Lead flow was strong. The CRM stayed busy. But the commercial reality was different. Sales calls were crowded with low-intent conversations — weak-fit prospects, interest without urgency, pipeline noise that burned commercial capacity. The funnel was active. It was also distorted.

The strategic move wasn't to chase more reach. It was to raise the cost of irrelevance. The narrative stopped selling possibility and started naming operational friction directly. The funnel was rebuilt to screen for intent before the sales conversation began — sharper promise, tighter lead form logic, pre-call education that filtered by design. Creative feedback from sales quality fed back into the next brief cycle continuously.

The result was counterintuitive on the surface. More friction. Fewer raw leads. Better prospects. A 25.17% close rate from qualified pipeline and $511k in documented revenue on $18,400 in ad spend.

Thesis: More leads were never the goal. The goal was to stop wasting commercial energy on demand that was never going to close.
Creative Group Split
Group A: Diagnostic Disruption
Creatives built to challenge the prospect's assumption about what was broken. Stronger hooks, sharper problem framing, aggressive pattern interruption. Hooks: "You don't need more leads." "Your pipeline is full of false demand." Blended MER: 9.1x in peak sprint window.
Group B: Qualified Intent Layer
Creatives built to convert informed prospects, not curious ones. Educational pre-sell, ROI framing, operator language, high-friction qualification. Built to sustain a 22% SQL rate without volume dependency. Stabilized ROAS: 13.9x across the full 75-day window.
Pipeline · Close Logic · Deal Math
1,280 leads · 22% SQL rate = 282 qualified prospects. 282 SQLs at 25.17% close rate = 71 closed deals. Average deal value: $7,200. Total revenue: $511,200. The peak ROAS of 27.7x reflects the discovery sprint. The 13.9x stabilized number is the commercially honest read.
The pipeline was full before this work started. The problem was that most of what was in it was commercially irrelevant. Qualification is a narrative problem before it is a sales problem.
DTC & eCommerce

A second specialty.
Not a side note.

From July 2024 to September 2025, concurrent engagements across B2B high-ticket and DTC eCommerce. Two verticals, same 14-month period, same methodology applied to fundamentally different commercial contexts. Two documented cases.

Track 01 · Jul 2024 – Sep 2025 · 14 months
B2B Growth Advisory
High-ticket B2B and healthcare. Long sales cycles, lead qualification, Dual-Layer creative architecture, SQL nurturing, CRM integration. Funnel logic: Ad to Lead to Nurture to Close.
Track 02 · Jul 2024 – Sep 2025 · 14 months
DTC & Direct-Response Clients
eCommerce and direct-response product brands. Short funnel, high creative velocity, hook-first architecture, UGC direction, fatigue management at scale. Funnel logic: Ad to PDP to Purchase to LTV.
Senior consultants and fractional strategists routinely serve multiple client verticals in parallel. The methodology transferred between tracks. What changed: the funnel shape, the conversion architecture, and which metrics mattered first.

DTC demands a different creative grammar. A hook that doesn't hold at 3 seconds doesn't get a second chance: there is no sales team to recover the lead. Hook Rate, Hold Rate, CTR, and ROAS by creative are the primary diagnostics. Creative velocity is the competitive advantage.

My process starts the same way in both verticals: upstream insight mining from competitor reviews, ad comments, and UGC data before any script is written. The 50% reduction in concept-to-live cycle was documented across both tracks. The AI-assisted briefing pipeline I built served both contexts without modification.

4.55x
ROAS on wellness DTC case, cold traffic only
$38.2k
Revenue on $8,400 spend over 60 days
34%
Hook Rate, above 25% category benchmark
$28.40
CPA (USD), cold traffic, wellness vertical
−50%
Concept-to-live cycle: 14 to 7 days
DTC · Wellness & Supplements · 2024–2025
The Identity Pivot
// How a compliance constraint became the creative breakthrough
Vertical
Health Supplements · Cold Traffic · Meta Ads
4.55x
Final ROAS
$38.2k
Revenue · 60 Days
$8,400
Total Ad Spend
$28.40
CPA · Cold Traffic
34%
Hook Rate

Supplements is one of the hardest niches in DTC Meta Ads. Elevated CPMs from iOS signal loss, compliance restrictions on outcome-based claims, and a feed saturated with before/after formats that users scroll past on reflex. Three walls at once.

The first three weeks failed predictably. Transformation-led hooks like "Feel the difference in 30 days" delivered an 11% Hook Rate, 0.64% CTR, and 1.4x ROAS. The data was clear: the audience wasn't dropping off because the product was weak. They were dropping off because the creative was speaking to who they wanted to become, not who they already were.

The pivot was total. Identity-first creative, built around the person who already makes the effort, already reads the labels, already shows up. No claims. No before/after. No promises. Just recognition. Hook Rate went from 11% to 34% in two weeks. CPA dropped from $74 to $28.40. ROAS stabilized at 4.55x on cold traffic, against a category median of 2.1x.

What this proved: The compliance constraint wasn't the problem. The problem was treating it as the ceiling instead of the brief. When you can't show transformation, you earn trust instead. That turns out to be the better strategy anyway.
What the Data Showed
Phase 1 · Weeks 1–3 · What Didn't Work
Transformation hooks. Hook Rate 11%, ROAS 1.4x, $1,900 spent with no scalable signal. Result-language triggers fatigue faster than any other format in saturated health categories.
Phase 2 · The Pivot
Stopped competing on outcome. Started competing on identity. Rewrote every hook around recognition, not promise. Hook Rate hit 34% in two weeks.
Phase 3 · Scale · Weeks 4–8
Scaled across UGC and static. Codified the identity-first logic into a brief system for creator direction. $6,500 spend generated $36,400 revenue. 4.55x ROAS held through scale.
iOS signal loss is real. Compliance restrictions are real. Audience skepticism is real. Working through all three simultaneously is what separates a strategist from an executor in this vertical.
DTC · Beauty Tech · LED Skincare · 60 Days
The Scroll Interruption System
// Attention as the first conversion event · Feed-hostile category · Meta Ads
Vertical
LED Light Therapy Mask · Beauty Tech · Cold Traffic
8.2x
Final ROAS · 60 Days
$149k
Revenue · 60 Days
$18,200
Total Ad Spend
38%
Hook Rate
+94%
Hold Rate · 16% → 31%

In saturated DTC feeds, most brands aren't losing at checkout. They're losing before the product gets a fair chance to be seen.

The product was strong. The category was crowded. The early creative set behaved the way most underperforming DTC ads behave — product-first, visually competent, commercially forgettable. The message arrived too late, after the thumb had already moved. Hook Rate was stuck at 16%. ROAS was under 2x. The data wasn't ambiguous about where the problem was.

The system changed when the creative process stopped treating attention as a soft metric and started treating it as the first conversion event. Hooks became the primary battlefield. Visual rhythm tightened. Creator scripts were rebuilt around interruption, specificity, and immediate cognitive contrast — not product showcase logic.

Once attention was earned, the rest of the structure had room to work. Demonstration landed harder. Benefit framing became persuasive instead of decorative. Hook Rate went from 16% to 38%. Hold Rate doubled from 16% to 31%. 1,244 orders at $120 AOV on $18,200 spend. ROAS stabilized at 8.2x, against a beauty tech cold-traffic median well below 3x.

What this proved: The feed doesn't reward good products. It rewards pattern interruption first. Product quality matters — but only after attention is secured. This is not a creative insight. It's a commercial constraint.
What the Data Showed
Group A · Thumbstop Layer
Creatives built to interrupt autopilot scrolling. High contrast hooks, visual disruption, first-line specificity. Hooks framed around recognition and cognitive contrast, not product claims. Responsible for driving Hook Rate from 16% to 38% and identifying the angles that fed the conversion layer.
Group B · Conversion Layer
Creatives built to convert already-engaged viewers through trust, clarity, and product demonstration. More proof, more visual process, less hype. Built for the viewer who paused — not the one still scrolling. Hold Rate improvement from 16% to 31% documented across this layer.
Order Math · AOV · Scale Logic
1,244 orders × $120 AOV = $149,280 revenue. $18,200 ad spend. Base CPA: $14.63. The AOV made the efficiency math sustainable at scale — a $14.63 CPA against a $120 product has room to breathe even under competitive pressure. CTR: 2.9%, above category benchmark.
Beauty tech has high CPMs, skeptical audiences, and a feed full of competitors making identical visual promises. The only defensible advantage at the top of funnel is making someone stop. Everything else is downstream of that first decision.
AI Systems & Automation

Four systems.
One loop.

Most creative teams spend 3 to 5 hours a week on work that doesn't require human judgment. I automated that layer so the team's time goes into the decisions only a person can make. The pipeline starts before a brief is ever written — at the point where most agencies still rely on assumptions.

04
Active pipelines running in production
20
Briefs generated per single run
<20m
Research to structured output
0
Manual handoffs between systems
00
Foundation Layer · Client Intelligence
Signal Pipeline
Most agencies brief from what the client tells them. What a brand claims about itself and what its customers actually experience are rarely the same document. This system finds the gap before a single creative is produced.
n8nWeb CrawlReview MiningAI SynthesisStructured StorageGoogle Sheets
Five phases running in sequence, fully automated. A new client submits a form. The system validates, approves, and routes the data without human intervention. It then extracts what the brand claims — positioning, voice, product benefits, objections. Simultaneously, it pulls structured customer data for every product provided: proof points, real frictions, confusion patterns. A synthesis layer then reconciles both sources against a defined hierarchy. The output is brief-ready intelligence — validated angles, real customer language, objection rebuttals — stored and ready to feed every downstream system. No manual step between client submission and brief-ready output.
Master Brand DNA outputs
Brand enemy + core value proposition
Real customer language (verbatim)
Top pain points + top objections + rebuttals
Full product matrix with emotional + functional benefits
Market sophistication level (1–5)
Validated creative angles from Amazon proof
How it runs — 5 phases in sequence
Phase 1.1 · Form Intake
Client submits Google Form. Fields normalized, appended to Clients Sheet, Slack notified. Triggers Phase 1.2 automatically on row detection.
Phase 1.2 · Gate 1
Sheet monitored every minute. Status = Approved? Routes to analysis. Status = Needs Info? Detects missing critical fields, updates notes, alerts Slack with exactly what's missing and who to contact. Nothing proceeds with incomplete data.
Phase 1.3 · The Brand Illusion
Brand site crawled and parsed by an AI extraction layer. Output: Brand DNA structure — positioning, voice, product matrix, objections, audience angles. This is the brand's self-image.
Phase 1.4 · The Market Reality
Customer data scraped and structured per product: five-star proof points, one-star frictions, Q&A confusion patterns. Capped and normalized to prevent downstream errors. This is what customers actually experience.
Phase 1.5 · Truth Synthesis
Both data sources are synthesized and validated through a defined hierarchy. Output is brief-ready intelligence stored and indexed for downstream access. Notification sent. Downstream systems are now loaded.
What changes
Briefs stop being built from brand-supplied assumptions. Every creative angle is validated against what real buyers said — before the first script is written.
Feeds into
All downstream systems. The Master Brand DNA in Supabase is the foundation every other pipeline reads from. No rework between intake and execution.
Phase 1.2 · Gate 1 — Approval + orchestration hub Approved → fires 1.3 · 1.4 · 1.5 in sequence
Signal Pipeline Phase 1.2 Gate 1 workflow — n8n
Phase 1.1 · Form Intake Google Sheets → normalize → append → Slack
Signal Pipeline Phase 1.1 Form Intake workflow — n8n
Phase 1.3 · The Brand Illusion Firecrawl → AI extraction → brand data output
Signal Pipeline Phase 1.3 Brand Illusion workflow — n8n
Phase 1.4 · The Market Reality Product data → scrape → classify reviews
Signal Pipeline Phase 1.4 Market Reality workflow — n8n
Phase 1.5 · Truth Synthesis AI synthesis → validate → store → notify
Signal Pipeline Phase 1.5 Truth Synthesis workflow — n8n
01
System 01 of 04 · Competitor Intelligence
Meta Ads Intelligence Pipeline
Most teams check competitor ads once a month, manually, without structure. By the time a brief is written, nobody knows what the market is actually running.
n8nApifyGeminiMeta Ads LibraryGoogle Sheets
Scrapes the Meta Ad Library for active competitor creatives, classifies each by format, and runs every asset through AI analysis: hook mechanics, message structure, psychological triggers, CTA logic. All results land in a structured Google Sheet organized by competitor, updated automatically. The output is a competitive map the brief team can actually use. They enter the room knowing what angles are saturated, what formats are running, and where the gaps are.
Outputs per run
Hook pattern library by competitor
Messaging angle map per category
Format distribution breakdown
AI-scored hook strength per asset
Psychological trigger index
Auto-updated Sheets database
How it runs
Client data in
Brand name, category, competitor list. Structured input fires the workflow. No setup required per client.
Meta Ad Library scraped
Active ads pulled for each competitor: video, static, carousel. Captions, run dates, format, ad IDs. All structured before analysis.
Classified and normalized
Format type assigned, metadata cleaned. Data enters the AI stage consistent and ready.
Gemini analyzes each ad
Hook mechanics, message hierarchy, emotional triggers, visual structure, CTA logic. Returned as structured JSON per asset.
Sheet updated, loop resets
Results appended by competitor and format. Loop continues for all remaining assets.
What changes
Competitive research goes from a monthly manual task to a live feed. Strategists brief from data. The category stops being a blind spot.
Feeds into
System 03 directly. No copy-paste. No manual step between research and brief.
Workflow · n8n · Production build Receive client data → scrape → analyze → write to sheet
Meta Ads Intelligence n8n workflow
02
System 02 of 04 · Native Platform Research
TikTok Top Videos Research
TikTok research was taking 3 to 5 hours per client, per week. Watch a few videos, take loose notes, call it research. Nothing structured. Nothing feeding the brief.
n8nApifyGeminiTikTokGoogle Sheets
Takes a client profile, generates targeted TikTok search queries automatically, scrapes up to 500 top-performing videos, and runs each through full AI analysis. Output per video: hook score, scene map, pacing style, psychological triggers, FYP hypothesis, and three variant directions for editors. The team doesn't watch videos. They read the playbook and brief from it. Research that took 3 to 5 hours now runs in under 20 minutes.
Outputs per video
Hook score + tactic label
Full verbatim transcription
Scene map with role tags
FYP fit hypothesis
3 hook variants per video
Script framework for editors
How it runs
Queries generated
Category, brand, audience data in. Five search angles out, each a distinct creative direction, not a repeat.
TikTok scraped via Apify
Up to 100 results per query. Play count, likes, shares, captions, hashtags, sound, author, post date. Structured before analysis.
Filtered and normalized
Empty batches skipped. Valid videos enter the queue with full metadata attached.
Gemini analyzes each video
Hook (0–3s) deconstructed. Scene map built. Pacing classified. FYP fit hypothesized. Three hook variants and three angle variants generated per video.
Playbook written to sheet
Hook score, tactic label, transcription, pain points, CTA type, execution notes for editors. Loop resets for next query.
What changes
Research that took 3 to 5 hours now runs in under 20 minutes. Editors and scriptwriters receive a playbook, not a list of links to watch.
Scale: up to 500 videos per run across 5 search angles
Feeds into
System 03, alongside Meta data, not instead of it. Both research streams enter the brief generator in parallel.
Workflow · n8n · Production build Generate queries → scrape TikTok → analyze → write playbook
TikTok Research n8n workflow
03
System 03 of 04 · Core Engine
Performance Brief Generator
Briefs were being written from memory. Same angles, same assumptions, same templates. The research existed but someone still had to translate it manually. That translation was the bottleneck I removed.
n8nClaudeSupabaseGoogle DocsSlack
Takes a client intake form, pulls brand DNA from a Supabase database, ingests competitive intelligence from Systems 01 and 02, writes a strategic angle thesis per slot, and generates 20 fully formatted creative briefs across three tracks: UGC video, motion graphics, and static ads. Each brief goes through three passes — draft, tighten, claims check. The final document saves to Google Drive and a Slack notification fires with the link. 20 briefs, fully researched, three-pass reviewed, delivered before the team sits down.
Outputs per run
20 formatted briefs per run
UGC video: hook + script
Motion graphics: visual architecture
Static ad: copy hierarchy
Strategic angle thesis per slot
Claims-validated copy · Slack delivery
How it runs
Intake form fires trigger
Product, audience, goals, performance context. Structured input, immediate start.
Brand DNA pulled from Supabase
Positioning, validated proof points, customer language, objection map, tone constraints. The AI writes from client history, not from zero.
Competitive research ingested
Meta and TikTok data pulled automatically. Angles are written against what the market is already running.
Angle thesis written per slot
Claude writes a distinct strategic direction for each of the 20 slots before generation begins. No two briefs share the same frame.
Three-track generation runs
UGC video, motion graphics, and static ad run in parallel. Each track uses format-specific logic.
Three-pass quality pipeline
Draft. Tighten. Claims Check: every factual claim validated against the Supabase brand database before the brief leaves the system.
Google Doc delivered, Slack fires
Brief formatted and saved to Drive. Slack notification with direct link. Done.
What changes
20 briefs, fully researched, three-pass reviewed, delivered before the team sits down. Writing one brief with real depth takes a senior strategist 2 to 3 hours. This produces 20 in a single run — with more context than any one person carries into the room.
Position in architecture
Terminal system. Takes output from Systems 01 and 02. Delivers the final brief. Nothing manual between research and production.
Workflow · n8n · Production build Intake form → brand DNA → research → angle thesis → 20 briefs → Google Drive + Slack
Performance Brief Generator n8n workflow
The pipeline starts before the brief.
It ends when the work ships.
Most teams brief from assumptions. These systems replace assumptions with structured intelligence — before a single angle is written. The foundation layer processes every new client automatically: what the brand believes, what customers actually experience, and where the gap is. The three production systems read from that foundation. Competitive signals feed in. Platform research feeds in. The brief generator synthesizes it all. No manual handoffs. No research debt. The team works on decisions only a person can make.
Career Architecture

Not random.
Designed.

Every role built something the next role needed. The Jul 2024–Sep 2025 period involved parallel engagements across B2B high-ticket and DTC/eCommerce clients — a deliberate expansion of vertical range that strengthened both tracks.

Oct 2025 — Present
Independent
B2B + DTC
Creative Systems Architect & Strategic Growth Consultant
Fractional strategic partner for high-ticket B2B and DTC eCommerce clients. Designing AI-augmented creative intelligence systems. Building Education-First funnel sequences. Implementing automation workflows — Make, n8n, OpenAI — to reduce idea-to-publish cycles. Translating business goals into executable marketing roadmaps across LinkedIn, email, and paid social. Currently focused on US market.
AI Workflow ArchitectureMake.com / n8nOpenAIRemote — US Market
Jul 2024 – Sep 2025
Scale Creative Advisory
B2B High-Ticket
Senior Creative Strategist & Head of Performance Intelligence
Led creative strategy and paid media optimization for high-ticket B2B and healthcare portfolios. Engineered the Silent Revenue Leak campaign — $705 in ad spend generated $21k in revenue in 30 days at 29.79x ROAS. Built Dual-Layer Architecture separating scale from stability (Group A: 6.72x blended MER / Group B: 8.1–9.3x ROAS). Automated briefing pipeline cutting idea-to-live from 14 to 7 days. Increased SQL meeting attendance 35–50% with zero budget increase. Total portfolio: $312k+ revenue influenced across 14 months.
$21k / 30 Days 29.79x Peak ROAS 4.67x Blended MER +50% SQL Attendance
Dual-Layer ArchitectureHigh-Ticket B2BHealthcareBriefing Automation
Jul 2024 – Sep 2025
Digital Native Advisory
DTC · eCommerce
Creative Strategist — DTC & Direct-Response
Concurrent DTC and direct-response eCommerce engagements running alongside the B2B advisory work. Applied the same narrative architecture and dual-layer creative methodology to a shorter-funnel, higher-velocity context. Focus areas: hook engineering for Meta and TikTok, UGC direction and structured briefing, creative fatigue management at scale (47+ variants per sprint), and AI-assisted brief generation. Flagship case: wellness supplements vertical, 4.55x ROAS on cold traffic, $28.40 CPA, 34% Hook Rate against a 2.1x category median.
4.55x ROAS · Cold Traffic $28.40 CPA (USD) 47+ Variants / Sprint 14 → 7 Day Cycle
Hook EngineeringUGC DirectionMeta + TikTokCreative VelocityAI Briefing
Feb – Jun 2024
Supernova Consulting
Brand Strategy
Brand Narrative Strategist
Concurrent strategy for 13+ client brands simultaneously. Developed Primal Branding manuals — Creation Stories, Creed, Enemy Narratives, Sacred Words. Conducted OSINT-based competitive analysis identifying Blue Ocean content gaps ahead of launches. Mentored founders on translating complex value propositions into commercially effective messaging. Zero churn during the engagement.
13+ Concurrent ClientsZero Churn
Primal BrandingOSINT ResearchMessaging ArchitectureMulti-client
Jan 2022 – Dec 2023
InterAR Advisory
Full-Stack
Head of Marketing & Strategic Operations
Full-stack marketing ownership: organic strategy, paid media, sales pipeline, community, and creator operations. Built an Ambassador/UGC Program that lowered CPA while building trust signals. Directed short-form content generating 1M+ organic views, repurposed as top-performing paid creatives. Aligned organic narratives with sales goals (Smarketing model). Broke 7-year company sales records.
1M+ Organic Views7-Year Sales Record
UGC ProgramSmarketingPaid + OrganicCreator Management
Jan 2019 – Dec 2021
Direct Response Group
Direct Response
Creative Copywriter & Direct Response Specialist
Foundation phase. 60+ direct-response scripts per month. Psychology-first approach — specialized in Pattern Interrupts, hook engineering, and psychological triggers (Status, Fear, Loss Aversion) before these became industry vocabulary. Pivoted from aesthetic writing to data-driven conversion copy through high-volume A/B testing.
60+ Scripts / MonthHook EngineeringA/B TestingDR Psychology
Market Positioning

Where the full architecture
becomes most valuable.

Primary Fit
Senior Creative Strategist
Where narrative meets performance data. Hook engineering, A/B narrative testing, performance analysis, and revenue-connected creative decisions — B2B or DTC.
Creative StrategistGrowth CreativePerformance Creative
Primary Fit
Growth Lead
Full-journey thinking from first impression to closed revenue. Funnel design, qualification systems, lifecycle strategy — not just top-of-funnel acquisition.
Growth LeadRevenue StrategyFunnel Architect
Technical Layer
Creative Systems Strategist
Building the automation pipelines, briefing systems, and AI-augmented workflows that multiply what a creative team can produce without multiplying headcount.
Systems StrategistAI Workflow EngineerCreative Infrastructure
Contact

Let's build
your architecture.

Available for B2B high-ticket and DTC eCommerce engagements — project or retainer. Remote. Async-friendly. US market focus.

If you're working on a real problem and need a system that compounds — let's talk.

Response time: under 24h
First session: diagnostic call
Languages: English C2 · Portuguese Native
Location: Remote (Brazil) · US Market