Trust And Verify

Last August, I got a message from a VP of Marketing at a mid-sized cosmetics brand.

"Can we talk?" she wrote. "Sales are off, and I need a second opinion on our ads."

I'll call her Rachel.

The brand's DTC revenue had not been growing quarter over quarter. Their retail partnerships were steady, but the direct channel, which had been their growth engine was sputtering.

She'd hired a performance marketing agency six months earlier to turn things around. And at first, it seemed like they were making progress. Instagram followers were climbing. Engagement looked healthy. The agency's weekly reports showed metrics trending in the right direction.

But sales weren't following. If anything, they'd gotten worse.

Rachel wanted me to review their marketing before our next meeting. She had a feeling something was off, but she couldn't put her finger on what.

I told her I'd take a look.

The story the dashboard told

Rachel gave me access to their Meta ads account, Shopify backend, and the agency's reporting dashboard. We hopped on Zoom, and I started digging while she watched.

The first thing I noticed was the Instagram growth. Their posts went from averaging 30 likes to averaging 750 almost overnight. That's a massive jump for a brand their size, especially in the cosmetics space where organic growth is brutally slow. And they weren't doing anything special.

I showed my wife who is an Instagram coach and creator. I pulled up their recent posts. Decent engagement rates. Nothing spectacular, but nothing alarming either. Then I clicked into the follower list.

We started scrolling together.

Fake profile after fake profile. Accounts with zero posts, generic usernames, no profile pictures. Random combinations of names and numbers like "jessica_2847" and "makeup_lover_9103." Some accounts didn't even have bios.

The further we scrolled, the worse it got.

"I think..." my wife said carefully, "something is off. I think their agency is buying followers."

The Andromeda cover story

I kept digging while Rachel processed what I'd just told her.

The agency had structured their account like a junk drawer. Overlapping audiences, redundant ad sets, campaigns targeting the same people in three different ways. The kind of setup that tells you someone learned Facebook ads in 2018 and never updated their playbook.

But what really got me was their weekly reporting to Rachel.

For the past three months, every report included the same explanation: Meta's Andromeda update was causing "industry-wide disruptions," creative fatigue was "expected during this transition," and rising CPMs were "affecting all beauty brands right now."

And here's the thing—they weren't completely wrong.

Meta did roll out the Andromeda update in April 2025. It fundamentally changed how creative gets distributed in their ad system. For years, you could run one "winning" ad and just swap hooks occasionally. CPMs were predictable. Scaling was straightforward.

Andromeda killed that approach. The algorithm started demanding complete creative diversity—not just different hooks, but entirely different narratives. Hook, body, CTA, all of it had to change. Brands that didn't adapt saw their costs spike and their campaigns collapse almost overnight.

So yes, the cosmetics industry was getting hit hard by this shift. The agency's explanation had just enough truth in it to be believable.

But that wasn't Rachel's problem.

Her problem was that she was paying an agency $12,000 a month to buy fake followers and hide their incompetence behind a real industry disruption. They'd used Andromeda as cover for the fact that they had no idea how to adapt to the new algorithm—or how to actually drive sales for a cosmetics brand.

The creative work they'd produced was solid, honestly. Beautiful product shots, strong testimonials, compelling before-and-afters. But none of it mattered because they weren't running campaigns designed to acquire customers. They were running a vanity metrics scam with gorgeous ads as window dressing.

"So what do I do?" Rachel asked. "I have a board meeting in two weeks."

The real numbers underneath

I spent the next hour breaking down what I'd found.

I showed Rachel the fake followers. I walked her through how the agency's campaign structure was fighting Meta's algorithm instead of working with it. I explained why their "recent boost" in followers had zero correlation to her declining sales—because those followers weren't real people who could buy lipstick.

Then I pulled up her Shopify data and cross-referenced it against the ad spend.

Her actual customer acquisition cost had increased 140% since the agency took over. Not because of Andromeda. Because the agency was optimizing for the wrong metrics and spending her budget on fraud.

Meanwhile, the creative strategy they were running was designed for the old Meta algorithm. Single hero ads with minor hook variations. No narrative diversity. No proactive rotation. They were running a 2023 playbook in a 2025 algorithm environment and charging premium rates to do it.

"Here's what an actual post-Andromeda strategy looks like," I told her.

I walked her through the shift: simplified campaign structure, complete narrative diversity where each ad tells a different story about why someone needs the product, proactive creative rotation every two to three weeks, and proper conversion tracking to separate warm audiences from cold acquisition.

"Can you help me fix this?" she asked.

I wasn't actively looking for new clients. But I also couldn't walk away from this knowing what I knew. Rachel was about to walk into a board meeting with fraudulent data and a failing strategy.

"Yeah," I said. "I can help."

Rachel hired me that afternoon.

What verification actually looks like

I used to think "trust but verify" was corporate-speak for "I don't trust anyone."

But Rachel's situation taught me it's the opposite. Verification isn't about distrust. It's about accountability. It's about caring enough to understand what's actually happening instead of accepting polished reports at face value.

When I started working with Rachel, the first thing we did was tear apart the agency's reporting structure.

No more weekly recap emails with cherry-picked metrics. No more dashboards that highlighted follower growth and impressions while burying customer acquisition costs. We set up direct access to everything: Meta ads, Shopify, Google Analytics...so Rachel could see raw data whenever she wanted.

We fired the agency and rebuilt the creative strategy from scratch. I brought in a media buyer who actually understood the Andromeda shift and knew how to build creative pipelines for beauty brands.

Within six weeks, Rachel's customer acquisition cost dropped by 55%. Her return on ad spend went from 1.8x to 4.2x. Sales started climbing again for the first time in eight months.

But the bigger shift was in how Rachel operated.

She started reviewing the numbers herself every Monday. Not just glancing at summary slides but actually looking at the data. She asked questions when something didn't track. She stopped accepting "it's the algorithm" as an explanation for poor performance without seeing the proof.

She learned to verify everything.

And here's what surprised me: her new media buyer appreciated it. When you're doing good work, you want clients to verify. You want them to see the results and understand what's driving them. You're proud of what you're building, and you welcome the accountability.

It's only the frauds who get defensive when you ask to see the raw data.

What I keep thinking about

Rachel walked into her board meeting two weeks later with completely different numbers than she'd expected to present.

Instead of defending declining sales with vague explanations about algorithm changes, she showed them exactly what had gone wrong, what we'd fixed, and what the early results looked like. She was honest about the money they'd wasted. But she also had a plan and proof it was working.

The board extended her timeline and increased her budget.

But I think about how close she came to a different outcome. If she hadn't reached out. If she'd just accepted the agency's explanations for another quarter. If she'd trusted the fake follower boost as a sign things were turning around.

She would have walked into that board meeting with fraudulent data, defended a failing strategy, and probably lost her job.

And the scary part? She's not unique. I see this pattern constantly. Smart, capable people running businesses or leading departments who've handed over critical functions to agencies or contractors and stopped verifying the work.

Not because they're lazy or incompetent. But because they trusted the expertise they hired and assumed professional-looking reports meant professional-quality work.

How many other Rachels are out there right now, looking at dashboards full of green arrows while their businesses slowly bleed? How many are about to walk into board meetings or investor calls with numbers that don't tell the real story?

I don't know. But I think about it a lot.

Your turn

This week, I want you to pick one area of your business where you've been trusting without verifying.

Maybe it's your marketing agency. Maybe it's your ads manager, your bookkeeper, your developer, or your operations lead. Maybe it's a tool you're paying for that you haven't actually audited in six months.

Ask yourself: When's the last time I looked at the raw data instead of just accepting the summary someone sent me?

If the answer is "I don't remember," it's time to verify.

You don't need to be an expert to ask good questions and spot inconsistencies. You just need to care enough to look.

Write back and tell me what you find. I can't reply to everyone, but I read every response, and I love hearing from you.

That's all for this week.

Have a great one!