I run Meta Ads for two active e-commerce brands simultaneously. VALO Gallery sells illustrated art prints. NeoEssentials is a men’s skincare brand.
Different products, different audiences, different markets, different price points. Both have active campaigns. Neither has an agency. The total time I spend on paid advertising across both accounts is about 2-3 hours per week.
Here’s how that’s possible.
The Setup That Makes It Work
The first thing I did was stop treating each brand as a completely separate system. Yes, the products are different and the audiences are different — but the underlying process of running Meta Ads is the same for both:
Build an audience → test creative → identify what works → scale it → refresh creative before it fatigues → repeat.
Once you have a reliable process, running it for two brands doesn’t take twice as long. It takes maybe 30-40% more time than running it for one, because the thinking is the same and the execution is mechanical.
The second thing I did was automate the reporting.
Both accounts feed into the same Make.com scenario. Every Monday at 9am, the system pulls the past week’s data from both accounts via the Meta Ads API, runs it through Claude AI, and delivers two separate reports — one for VALO, one for NeoEssentials — with executive summaries, key metrics, and three specific actions for each.
I sit down on Monday morning, read two reports (takes about 15 minutes), and execute a combined six actions across both accounts (takes about 20-30 minutes). Done for the week.
Campaign Structure: Keep It Simple
One of the most common mistakes in Meta Ads management is over-complexity. Too many campaigns, too many ad sets, too many variables running simultaneously. This creates a management burden and also fragments your budget in ways that prevent the algorithm from optimizing effectively.
My structure for each brand:
1 cold traffic campaign — targeting new audiences who don’t know the brand. Usually 1-2 ad sets: one with broad targeting, one with a lookalike audience based on past purchasers.
1 retargeting campaign — targeting people who’ve visited the site or engaged with content. Only runs when there’s enough audience (typically 1,000+ people) to make it efficient.
That’s it. Two campaigns per brand, four campaigns total. The simplicity makes them easy to manage, review, and optimize without losing track of what’s doing what.
Creative: the Only Variable That Really Matters
Once campaigns are set up with a sensible structure, creative is where almost all performance variance comes from.
For VALO, I test lifestyle images — the poster in a real interior space — against isolated product shots on clean backgrounds. The lifestyle image wins almost every time. Real context sells physical products better than perfect photography.
For NeoEssentials, I’ve tested product-forward images against benefit-focused copy with minimal imagery. The products are skincare — they don’t have the same immediate visual appeal as art prints — so copy angle matters more.
The discipline I’ve built: never run a single creative for more than 3-4 weeks without testing something new. When frequency hits 2.5-3, creative fatigue is coming. Prepare before it arrives, not after performance has already dropped.
What I Do When Something Isn’t Working
The weekly automated report catches problems before they become expensive. When the Claude analysis says “Worst performing campaign: ROAS 1.4×, CPM €14, recommend pause” — I pause it. I don’t try to fix it immediately. I pause it, think about why it might have underperformed, and launch a new version the following week with one change.
The key is changing one variable at a time. If I change the audience and the creative simultaneously, I don’t know which one caused any improvement. If I change just the creative, I learn something definitive.
This sounds obvious but in practice, when a campaign is underperforming, the temptation is to change everything at once. Discipline here pays off in accumulated knowledge about what actually works for each brand.
What I Outsource and What I Don’t
I do all the campaign management myself. I set up the campaigns, manage the budgets, make the optimization decisions.
What I don’t do: produce the creative. I brief what I need — lifestyle images at specific sizes, specific rooms, specific color palettes — and generate them using AI tools. This isn’t a compromise; AI-generated lifestyle imagery for physical products is genuinely good now, and it’s faster and cheaper than coordinating a photoshoot.
The Meta Ads Report tool is effectively a junior analyst that I never have to manage. It does the data collection, interpretation, and recommendation generation automatically. I make the final calls and execute the actions.
The Honest Limitation
I’m managing brands with modest budgets — €10-20/day per account. At this scale, the management overhead is low because the number of meaningful decisions per week is small.
If I were managing €500/day per account, I’d need more infrastructure — more creative testing, more granular audience segmentation, more sophisticated attribution. The approach I’m describing scales to a point, and that point is somewhere around €100/day total across all accounts before you genuinely need dedicated resources.
For small brands and small agencies managing multiple modest-budget accounts, this approach works well. For enterprise media buying, you need something more sophisticated.
→ See how the automated reporting works at louvrlabs.com/report → Meta Ads management as a service at louvrlabs.com