You're a tiny team taking a beauty brand from Bulgaria into Europe, the UK, and eventually the US. The blocker is creative volume, you can't out-headcount the brands you'll be sitting next to in feed. What you need installed isn't a tool, it's the system around it, the brand bible, the tagging pipeline, the brief-to-static loop, the QC agents, and the landing-page generation that pairs with each ad. This is a 3-month engagement to build it from zero to one, alongside you.
From our call on May 1. NL Beauty is a Bulgarian beauty brand led by a tiny team, Bobby on creative strategy, Angel as CEO and co-founder, with real distribution in Bulgaria today and an open lane into Europe, the UK, and the US. Two pieces of the call this proposal is built around:
The rule of thumb in this ecosystem is one net-new ad concept per $1,000 of spend, and a "concept" means a different persona, angle, emotion, or format, not just a new hook on the same ad. A small team can't out-ship that on willpower.
The fix is a system that does the boring half, the tagging, the brief drafting, the prompt assembly, the QC pass, and leaves the taste calls to your humans.
NL Beauty doesn't have a creative ops backlog to migrate, you're early enough to build it right the first time. Bulgaria today, then Europe, the UK, and the US, each market is a fresh launch reusing the same kit, brand bible, tagging, briefs, statics, paired landing pages, localized not rebuilt.
Every install I do is from zero to one. I build it with you, not for you. The system goes into NL Beauty's environment, not mine.
One perfect ad doesn't move the needle in beauty, not when feed is wall-to-wall with brands ten times your size. Volume with the right spread of personas, angles, and formats does. We'll build NL Beauty's install to feed exactly that, statics across founder story, ingredient story, before/after, social proof, demo, education, every week.
At Selfmade, the AI statics now outperform what our editors ship. Not by a little.
You're not buying files-and-peace and you're not buying a workshop. I show up, we build the system inside NL Beauty's environment together, and at the end you own how it runs. Every engagement I take is zero to one for that brand, the markdown files, the asset templates, the agent loops, all tuned to NL Beauty in the open.
Skip the teaching. Build it together.
Bulgaria today. EU, UK, US tomorrow. Your brand bible, your tagging pipeline, your agent loops, your paired-LP workflow, localized into each new market, not rebuilt. Every market you open is faster than the last because the install already knows NL Beauty.
Plus paid social → paired landing pages → conversion is one closed loop in this engagement, not three vendors.
Three tranches. Foundation lands first because nothing downstream works without it. Production layers on as we start shipping briefs and statics. Optimization closes the loop with QC, brand-voice calibration, and the paired-landing-page workflow.
Higher cadence in the build month, lower cadence after, when your install is running and just needs nudging. Slack-embedded throughout. Ongoing additions to your install as new asset types, scoring agents, and model integrations come online.
Heavier in month 1 because that's where the build happens. Lower after because your install runs itself with light calibration once it's built. 3-month minimum committed up front. No long-term lock-in beyond it.
So nobody's surprised in month 2. If any of these become important, we scope them separately.
The sprint pace. Each item below is what's true at the end of that week.
SOW countersigned, M1 invoice paid. Slack channel created. Kickoff call scheduled inside 48 hours.
Meta API access provisioned. NL Beauty brand bible and spec cards drafted from your site, winning organic, and tone. Repo skeleton committed.
Tagging pipeline, gap analysis, and historical winner analysis live in your repo. First gap-analysis dashboard for NL Beauty vs. the beauty category lands in Slack.
Brief-generation agents, prompt converter, and fal.ai static generation all running on NL Beauty's repo. First test briefs go through the 10-agent scoring loop.
Statics generated, QC'd, and live in the account. Sprint cadence ends. We move to bi-weekly + Slack for months 2–3.
QC agent loop, feedback updater, brand-voice agent, and the paired landing-page workflow built into your install. Refinements continue. Month 3 closes with a "running it without me" checklist.
If this is a go, I send the SOW + Stripe link inside 24 hours. Kickoff call scheduled inside 48. First AI statics in your Meta account inside 30 days.
