In partnership with
Podcast ad intelligence, for a fraction of the cost
Radar uses state-of-the-art transcription technology to transcribe and analyze where brands are advertising across 115,000+ podcasts.
See where your clients’ competitors are advertising on podcasts, read the exact copy for their ads, and verify your own clients’ ads have run.
Radar also allows you to search across millions of podcast transcripts to find relevant shows to buy ads on.
New podcast episodes and ad reads are available minutes after an episode airs.
Radar is built by former Twitter and Tesla engineers, using an AI-native transcription pipeline that delivers high accuracy and extensive data enrichment, at a fraction of competitors’ costs.
€50 a day is enough. Five ads are enough.
The reason most small advertisers never get there is not budget.
It is that producing five genuinely different ads used to need a media buyer, a copywriter, a designer, or an agency retainer.
I built a full batch for a client selling metal garden sheds.
Here is the whole process.
This is MVAC, run entirely in statics
Back in issue #04 I published the MVAC framework, the Minimum Viable Ad Campaign.
One campaign, one ad set on Advantage+, three ads.
Those three ads tested three ad formats: static, carousel, video.
This build keeps the same skeleton and changes the test axis.
Format stays fixed at static.
What varies is the message, across five different kinds of buyer.
Same discipline, different question.
MVAC asks which format earns attention.
This asks which message does.
The structure takes fifteen minutes:
Campaign: Sales (or Leads if you're not ecommerce)
Ad set: ONE. Advantage+ Audience on. Broad. No interest stacking.
Budget: All €50/day in that one ad set
Ads: 5 statics inside it
Do not split into three ad sets at €17 each.
Meta wants roughly 25 conversions a week per ad set before it stops guessing, and you will never get there.
Five ads, not one ad five times
If all five say "here is our product" or "here is 10% off", the whole budget talks to the handful of people already shopping.
The five ads should sit at different distances from buying, from people who have never thought about the problem to people ready today.
That is the problem awareness funnel from issue #108, and it is the single idea doing the most work in this build.
Why I prefer Claude AI, and why skills
These five skills are that, built out properly.
Not one of them touches the ad account.
Nothing publishes, nothing changes a budget.
They produce text, and I do the rest by hand in Ads Manager.
Why skills and not just prompting ChatGPT: a skill is a file that installs once and runs the same way every time.
A prompt gets re-pasted, edited, and drifts.
Six weeks later you cannot reproduce what you did.
The skill also reads a context file about the brand, so I stop re-explaining the client in every chat.
I work in the terminal with Claude Code.
If you would rather not, Claude Desktop or Web app runs the same skills with an upload and a toggle.
Install guide for both (with screenshots): digitaladsnack.com/meta-ads-claude-code-skills
It’s also FREE!
Step 1. Build the brand context, once
I pointed the foundation skill at the client's product page and pasted in three real customer reviews.
BUYER: "The Owner With a Cluttered Yard"
Trigger: the garage fills up, a tool rusts outside, or they start landscaping and want order.
Surface pain: "I have nowhere to put my tools"
Deeper pain: "the bikes are rusting outside, the yard looks a mess"
2am pain: "I have a beautiful yard I can't use because it's full of stuff"
Objections: rust, assembly difficulty, wind and snow
BIGGEST GAP: brand advertises only at product-aware and ready-to-buy. Nothing speaks to the ~85% earlier in the funnel.
That last line is the whole strategy, and it came out of their own reviews.
You build this file once and every other skill reads it.
Step 2. Hooks, one per awareness level
32 hooks, then a top five picked deliberately one per level.
That constraint matters.
Ask any AI for your five best hooks and you get five bold product claims, because bold claims read strongest alone.
Five ads for the same small audience.
L1 Unaware
"Every garage starts out as a garage."
→ Lo-Fi Social Proof
L2 Problem-Aware
"You have a garden you cannot use, because it is full of stuff."
→ Before & After, weighted to the "before"
L3 Solution-Aware
"Wooden shed rots. Tarp tears. Metal locks."
→ Us vs. Them
L4 Product-Aware
"Metal, not wood. It does not rot and it does not need a foundation."
→ USP / Core Benefit
L5 Most Aware
"10% off every size while the landscaping season lasts."
→ Deals / FOMO
Note it assigns the ad concept too, from the 8 evergreen concepts in issue #105.
You never have to guess which kind of ad to make.
Step 3. Copy at the right depth
Same product, five genuinely different ads.
L1 · 1-3 lines, product last, closing the thought, "Learn more"
"The garage was supposed to be for the car. Then it slowly became the place everything else goes. A lockable metal shed puts it all back outside."
L3 · long, this reader will actually read it, product mid-body as the category answer, "See how it works"
"A wooden shed rots from the base up, usually around year four, and it needs a foundation before you even start. A tarp lasts a season. Metal is the only one of the three that locks, survives a winter, and goes up in an afternoon on flat ground. That is the whole reason we sell metal and nothing else."
L5 · 2-4 lines, product first, next to the offer, "Shop the sale"
"10% off every size until the season ends. Lockable, weatherproof, assembled in an afternoon."
It also flags which lines came from real reviews. "I finally cleared out my garage, the car fits again" is a customer's sentence, not mine.
Those tend to win.
Step 4. The design brief, the step that used to kill it
You have copy.
You still do not know what the picture looks like.
So you reach for a stock photo and a logo, and the ad dies.
I asked for five ad concepts for the coldest audience.
It chose the concepts itself and briefed each one.
THE FIVE IT PICKED
1 · Lo-Fi Social Proof
strongest L1 fit for this brand, raw phone photo
2 · Lo-Fi Social Proof
same concept, review-screenshot format, uses a real review
3 · Us vs. Them
no new photography, works off the catalogue image
4 · USP / Core Benefit
kills the rust and assembly objections in one claim
5 · Deals / FOMO
live 10% off plus a seasonal deadline
It also told me what it would not do. Meme ads are the usual first choice for cold traffic, but the brand's context rules them out for this category, so it reused the lo-fi format twice rather than reach for the wrong tool.
Here is one of the five in full:
Ad 3 · Us vs. Them, L3 Solution-Aware
Core message: metal is the only one of the three that lasts and locks.
ON IMAGE
Headline: "Three ways to store a mower."
Left column (grey, crossed): Wooden shed, Tarp
Right column (brand, ticked): Metal shed
CTA: "See the difference"
VISUAL
Two-column layout, text-led. Product shot sits in the Metal shed column. Use the existing catalogue image, no new shoot needed.
AVOID
Rows that don't answer each other one to one.
Step 5. Score before you spend
Every finished image through the 26-point Creative Readiness Scorecard from issue #103.
Under 13, fix it before it goes live.
Cheaper to fail there than in the auction.
One setting to switch off
Every new Sales, Leads and App Promotion campaign launches with all Advantage+ Creative enhancements on.
Text rewrites, cropping, background generation.
Turn them off at the ad level.
You just built five ads aimed at five different buyers.
If Meta rewrites the headline on the cold one, it is not the cold one anymore, and the test tells you nothing.
Conclusion
The budget was never the constraint.
Neither was the missing designer.
Producing an ad felt expensive, so you made one idea and dressed it up five ways.
One campaign. One ad set. Five statics that each talk to a different person.
An afternoon of work, and the tools are free.
TLDR:
Keep the MVAC skeleton, one campaign and one ad set, but run five statics that vary the message instead of three that vary the format.
Five ads only works if they sit at different awareness levels.
Skills beat prompting because they install once, run identically, and read a saved brand context instead of you re-explaining the client.
Keep AI outside the ad account. Research, copy and briefs only, never the API.
Install guide for Claude Code and Claude Desktop: digitaladsnack.com/meta-ads-claude-code-skills
P.S. If you want to build statics like these without starting from a blank canvas, my template pack covers all 8 evergreen concepts, editable in Canva:
DAS is also sponsored by:
7 Strategies. Zero Guesswork.
The brands winning on Amazon stopped guessing. They know which external channels pull weight — and which ones don't. Levanta's free playbook breaks down 7 strategies driving millions in off-Amazon revenue.
♥️ Tools we love to use at my agency:
My ultimate list of FREE Ad Libraries for spying and inspiration
Get my FREE Facebook Ads Hook Library and make it yours.
Email Marketing tools for e-commerce Klaviyo and for leads Mailerlite. We choose the one that fits our client’s needs between these two.
Best Pop-up Tool for e-commerce to grow your email list: Optimonk
Our go to No-Code Drag & Drop website and landing page builder with 170+ prebuilt templates: Sitejet
#1 Newsletter tool for founders, creators, writers is Beehiiv.
Get 30-days Free Trial + 20% OFF for 3 months with this link.
For Project Management we use Nifty.
* If you click the links above, we may receive a tiny commission that helps keep Digital Ad Snack running—thank you!