Post-call · July 31, 2026

Built for Joe.
Built for Fable.

You are spending well over $200K a month on Meta with 29 ads live, and the agency making them ships work that, in your words, just looks like a VA made it. Fable's product is premium. The ads should be too, and there should be a lot more of them. This page is the plan for both, with 14 spec ads already built from your real catalog as proof.

$200K+/mo on Meta Canada · US · UK 25 live ads in the public library Meta only. Month to month.
Premium at volume. Explicit personas. No VA energy.
Grab 20 minutes → See the 14 ads See pricing
01. Your plates survived three moves
06. Tuesday deserves the good plates
07. The wine glass that won a Red Dot
08. The dog can stay on the rug
10. Bring the tasting room home
14. Three reasons not to buy
14 BUILT FOR FABLE
What the live library says

Your account, first pass.

Before sending this we pulled Fable Home's public Meta Ad Library: 25 active ads as of July 31. Starter sample, real signal. The full pull at kickoff ingests every ad the account has ever run and tags each one by persona, angle, emotion, and format. Here is what the public set already shows.

What the 25 live ads lean on

Where the library is concentrated right now
Restock scarcity"Now Back In Stock" headlines 12 of 25 live ads. Nearly half the library on one message.
High
Category naming"Washable Wool Rugs," "Everyday Glassware," "Dinnerware For Every Course." The product line as the whole pitch.
High
One quality adjective"Heirloom-Worthy Homeware" carries the entire premium story across the set.
Medium
Concept varietyRoughly five distinct headline concepts across 25 ads, batch-uploaded as variants within days of each other.
Low

What has zero active coverage

Lanes with evidence behind them and no live ad on them
Explicit personasNo ad names its buyer. Hosts, registry couples, parents with dogs, ramen cooks, wine people: all unaddressed.
Absent
Social proofRated 4.9 out of 5, thousands of 5 star reviews on your own site. Not one live ad uses any of it.
Absent
Craft and provenanceHand finished in Portugal, German crystal, hand forged in Japan, New Zealand wool. Your strongest premium proof, unused.
Absent
Award credibilityYour wine glasses won a Red Dot Design Award in 2025. The library never mentions it.
Absent
The 14 spec ads were built against this map: personas named out loud, the 4.9 rating and real reviews on canvas, Portugal and Japan and the Red Dot doing the premium work, hosting and registry lanes opened. Every line traces to your site, a verbatim customer review, or the call.
See the 14 ads →
What we heard on the call

Four things you said. Four answers.

Joe's own words from July 31, and how the engagement is built around each one.

"just looks like a VA made it"
Joe · on the current agency's output

The quality floor is the whole reason our team is 25 to 30 editors and not five. Every asset, including anything AI touched, goes through a human editing pass in Photoshop or After Effects, then a 10-gate creative grade before you ever see it. Nothing ships below the bar, and the bar is set to Fable's positioning, not to volume for volume's sake. The 14 spec ads on the work tab are the floor, built in one day.

"it makes our glasses look curved"
Joe · on AI ad tools

We use AI for the first 10 percent and the last 10 percent. The middle 80 is human. And where rendering is involved, your real product photography is the ground truth: every glassware spec ad on the work tab was generated with your actual catalog photo as the heaviest-weighted reference, then graded against it. Straight stems, true rims, your real geometry. Anything off gets fixed by an editor or killed.

"We can't get out of the learning phase."
Joe · on why more ads felt impossible

The structure from the call: a testing pool where a diverse set of concepts runs wide, spend is the leading indicator, and winners graduate to scale campaigns. It is fine that some ads barely spend, that IS the test. The math we hold to: every thousand dollars of monthly spend wants a net new concept behind it. At your spend, 29 live ads is the bottleneck, not the budget.

"we can test out like a $6,000 a month kind of like starting point"
Joe · on how to start

Exactly the shape this proposal takes. We cover a $40K to $80K slice of the account at $6,000 a month with 14 unique concepts a week, while your team keeps shipping campaign and partnership work in parallel. Your b-roll and existing assets get folded in rather than wasted. Prove the slice, then widen it.

Also confirmed on the call, in writing: everything is month to month, no lock in, and our average client tenure is 1.9 years anyway. Everything runs through Slack like an extension of your team. Briefs land Monday, your notes by Wednesday, finished assets the following Friday. We connect to the ad account for insights and pull your existing b-roll library into the mix from week one.
How we work

The three-piece engine.

Three inputs feed every brief. None of them are opinions, and each carries a fixed weight in how the next test gets chosen.

Piece one · weight 0.5

Time series on every ad you have ever run.

We ingest the Fable account since day one and tag every ad by persona, sub persona, angle, emotion, format, and product. Then we read it over time: which angles carried, which quietly stopped getting made, which winners were abandoned without a reason. Recovered learnings become the first briefs, and usually the first wins.

Delivered as a live dashboard. Yours to keep and share.
Piece two · weight 0.2

Gap analysis against the space.

Your library versus the category: where Fable over indexes, where it under indexes, and which angles and personas the premium home space runs that you have never tested. The 25-ad public pull above is the starter version. The full version runs at kickoff and re-runs quarterly so the map never goes stale.

First pull already done: 25 live ads mapped, concentration named.
Piece three · weight 0.3

Social listening where your buyers actually talk.

Thousands of 5 star reviews on your own site, ad comments, Reddit's home and kitchen threads. Real language becomes real copy: one spec ad's entire headline is a customer's own sentence about the weight of your mugs. Explicit persona callouts get written from that language, because the more precisely an ad names its buyer, the harder it works.

Already in use: two verbatim reviews are on the work tab right now.
The weekly output splits 60/20/20: build on what is working, reskin proven winners into fresh formats, and always test net new personas and angles. That last 20 percent is how accounts get out of the same-audience loop that makes promo spikes feel like the only lever.
Pricing

Start in the six thousand bucket.

You named the starting point on the call, and it is the right one. Slide to the monthly ad spend you want SelfMade creating against. The rule of thumb: every thousand dollars of monthly spend wants a net new concept behind it. At $200K+ total, a $40K to $80K covered slice is the proving ground, with your in-house campaign and partnership work running in parallel.

Every tier includes the full engine: gap-analysis briefs, 9-rubric copy refinement, creative grading on every render before delivery, the weekly cadence, quarterly re-runs of the gap analysis, and time series plus social listening bundled. The dashboard comes with every tier.
Monthly ad spend SelfMade covers
$60K/ month
$6,000 / mo creative
$40K$60K$80K$125K
Your starting point
Standard
$6,000/ mo
Covers $40K to $80K/mo of ad spend
14 unique concepts a week
When the slice widens
Scale
$10,000/ mo
Covers $80K to $125K/mo
30 unique concepts a week
Concepts translate flexibly: six statics equal one video. A 14-concept week can be 8 statics and one video built from the b-roll you already have. You pick the mix week to week; the concept count is the constant.
Add-on · landing pages
+$3,000/ mo

Persona-tuned landing pages

Persona-tuned LPs spun up against winning ad concepts: the registry couple lands on a registry page, the host lands on hosting. Recent comp: roughly 30% CAC reduction from the layer alone.

Add landing pages
Add-on · media buying
+$4,200/ mo

We run the buy

7% of covered spend. As discussed on the call: most clients start creative only and hand us the buy later if they want it. Ad budget always runs from your own account, and creative delivery works the same either way.

Add media buying
Your proposal total
$6,000/ month
STANDARD · 14 CONCEPTS A WEEK
Tier deliverables are fixed: 14 or 30 unique concepts a week. Month to month, no lock in. Final start date locks on the close call.
↑ Spending $200K+ a month across the whole account? The covered slice is the dial. Start at $40K to $80K in the six thousand bucket, prove the slice outperforms, then widen toward Scale on the same per-creative economics. That is a five minute conversation once the first month's numbers are in.

"We'll be looking to make some decisions here like early next week."

So everything is here before the weekend: the live-library first pass, 14 spec ads from your real catalog, and the pricing you asked for. Read it whenever suits, forward it to whoever weighs in, then grab 20 minutes early next week and we will walk the dashboard and lock a start date.

Grab a slot with Will → will@selfmade.co