You're a five-person creative team behind a DTC brand averaging ~$300K/month, headed for $600–700K by the end of the year. The bottleneck isn't taste, it's hours: you're the strategist, a copywriter, and an editor, and half your week disappears into spreadsheets hunting for insights. What gets installed isn't a tool, it's the system around it, your full ad history pulled and tagged, statics generating on demand, a dashboard that opens with the insights already there, and an install your dev can extend. Three months, built alongside you.
From our call on August 28. A lean five-person creative team, Kris on strategy, copy, and edits, a full-time editor, a full-time designer, a freelance editor, plus a dev already hired to build AI workflows. Two pieces of the call this proposal is built around:
"I'm great at creatives. When I jump into Motion, I get hit by this spreadsheet, and I just gotta start finding these insights, finding these patterns. For me it's a time suck. I want to open something up that has the insights there, so I can get to work."
That's exactly the thing this install ends. "What's working, what's not" should never be a conversation, or an afternoon in a spreadsheet. It should be a screen you open, then you go write, brief, and edit.
You're scaling from ~$300K to $600–700K/month by year end. That scale demands creative volume and diversity your current headcount can't ship by hand, and you've already hired a dev to build AI workflows around it.
Instead of your dev guessing at the architecture for months, we install a system that's already working inside brands and my own agency, and your dev extends it from there, cross-platform pulls, custom views, whatever you dream up.
Every ad you've ever run gets pulled through the Meta API and tagged by persona, sub-persona, angle, emotion, product, and format. Then a time-series analysis of the whole account shows which cohorts have historically won, not anecdotally, analytically.
A protein brand we did this for found gut health was a core persona they'd barely touched. The ads built on that insight were winners right off the bat.
Brands stall by pushing the same audience until CPMs choke. The batch mix this system ships by default: 60% iterations on what's working, 20% reskins of winners into new formats, 20% net-new personas mined from subreddits and voice-of-customer data you've never tested.
The "test something new" lane is the escape hatch from spend purgatory, and it's automated.
No two installs look the same. You asked about comparing Meta vs. YouTube vs. native, the system tags everything the same way across platforms, so you get a clean cross-section of what's winning where, and where an angle crushing on one platform hasn't been tested on another.
And because it lives in your environment, your dev keeps building on it after I'm gone.
Three tranches. Foundation lands first because nothing downstream works without it. Production layers on as we start shipping statics. Optimization closes the loop with the dashboard, cross-platform views, and video.
Weekly cadence in the build month, bi-weekly after, when your install is running and just needs nudging. Slack-embedded throughout, calls booked Bangkok-friendly. I take three consulting clients a quarter, so the work stays deep, not templated.
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 step.
SOW countersigned, M1 invoice paid. Slack channel created. Kickoff call scheduled inside 48 hours, booked Bangkok-friendly.
Meta, Anthropic, Gemini, and fal.ai keys set up live on the call. Starter kit installed in Claude Code. Historical ad pull kicked off, it runs unattended for 8–12 hours. You generate your first statics before we hang up.
Every ad you've run tagged by persona, angle, emotion, product, and format. Time-series winner + fatigue analysis lands. Untested-persona mining live. First AI statics launched into the account.
Batches shipping on the 60/20/20 mix, QC agents regenerating until clean, swipe-file diversity enforced. Volume and persona coverage you haven't had before. Sprint cadence ends, we move to bi-weekly + Slack.
Live dashboard customized to how you think, open it, see the insights, get to work. QC tuning, feedback ingestion, brand-voice agent, subreddit-driven copy refinement.
B-roll bank sliced and tagged, persona-driven video assembly live, YouTube and other platforms pulled in alongside Meta. Month 3 closes with a handoff doc and a "running it without me" checklist.
Take this to your CEO, everything's laid out above. If it's a go, I send the SOW + Stripe link inside 24 hours, kickoff call inside 48, and you're generating statics by the end of call one. Two of my three Q4 consulting spots remain.
