You’ve probably noticed more AI-generated ads showing up in your feed. And on the surface, they look good.

Polished voiceovers. Perfectly manicured hands holding supplements. Hyperreal visuals that mimic high-end lifestyle shoots. They definitely catch the eye.

Scroll-stopping? Definitely. But once you stop scrolling, do you click? Do they convert?

That’s the challenge a lot of CPG marketers are facing right now. Customer acquisition costs keep rising. Margins are tight. Creative fatigue is setting in. And there’s a lot of pressure to produce new content faster and cheaper.

So it’s a fair question: Is AI the solution? Or just the latest shiny thing?

What We’re Seeing Right Now

AI tools are making their way into CPG creative in a few noticeable ways, particularly on TikTok and Meta:

• AI-generated User-Generated Content with scripted voiceovers and glossy product shots

• Digital “creators”: AI-generated faces and bodies that look like influencers but don’t exist

• TikTok Shop-style accounts churning out near-identical videos - same script, different AI face or voice

Some are impressively lifelike. Some less so.

But even when it looks real, there’s something missing. We’re seeing this type of creative lack a sense of real human connection. And that still matters.

And now, platforms are flagging some AI-generated content. That adds friction. If a user sees “Generated by AI”, it can raise skepticism. This is a red flag, especially in categories like wellness, skincare, or nutrition, where trust drives conversion.

And here’s where it starts to show up in performance:

AI-generated assets may look polished, but we’ve seen them underperform in key areas like:

• Thumbstop rate: People pause, but don’t stay

• Click-through rate: viewers hesitate when the content feels impersonal or too scripted

• Conversion rate: even if they click, they often don’t buy because the trust is not quite there

In a space where brand credibility is already hard-won, even subtle signs that something’s artificial can raise doubts. And that kind of doubt hurts perception and trust. It impacts the metrics that matter.

Alpha’s POV:

We’re seeing the rise, but we’re also seeing the red flags.

Why We Haven’t Gone All-In Yet

At Alpha, we test new tools all the time. Of course, we do. But our rule is simple: if it doesn’t convert, it’s not worth the hype.

And right now, AI-driven creative just isn’t delivering yet. Here's why:

• Label accuracy is still hit-or-miss. AI image tools often don’t render packaging or product copy cleanly. This is a problem in categories like wellness where trust, clarity, and compliance matter.

• Creative production is clunky. Image generation can take multiple tries, and even then consistency is still a challenge. It’s not always as quick as it seems.

• Audiences can tell. Especially in categories like wellness, nutrition, or personal care where health claims or trust matter. Here authenticity matters more than polish, and AI content often falls flat.

That said, we’re not entirely ignoring AI. We’re just using it upstream where it adds speed without impacting strategy:

• Pulling insights from reviews

• Sparking ideas for messaging

• Exploring creative angles

• Research into competitors and new trends

• Structuring briefs faster

Those are real time-savers. But final creative? Today, that’s ultimately still a job for humans.

Will It Ever Work? Maybe. Here’s Where It Might Fit.

We’re not writing it off completely. There are a few scenarios where AI could support (not replace) creative production:

• Concept testing

  Use AI to visualize an idea or test messaging before committing to full production.

• Stylized background visuals

  Especially for static ads or design elements where product realism isn’t critical, and the creative doesn’t rely on a spokesperson to build trust.

• Certain verticals

  Categories where trust is built more on utility and polish than personality. Think tech accessories or organizers.

But in wellness, skincare, or anything health-related, realness still wins for now. Buyers in these spaces evaluate more than your product, they make their decisions based on trust.

And when the person or story behind the product doesn’t feel real, conversion suffers. We definitely wouldn’t recommend letting AI be the face of your brand - at least not yet.

Where We’re Seeing Real Wins in CPG Ads (Right Now)

Instead of chasing shortcuts and experimenting with avatars, here’s where we still see consistent wins:

1. Customer-first creative

Think unpolished user-generated content, real testimonials, and voiceovers that feel native and natural. Creative that looks like it came from a real customer not an editor’s room still outperform polished spots.

2. Formats that feel platform-native

You don’t want your ad to disappear, you want it to belong. The best-performing ads don’t try to look like ads.

On TikTok or Meta, content that feels native to the feed earns attention faster than anything overly polished or obviously scripted. This is where many AI-generated creatives still fall short. They look too perfect to feel personal.

3. Strong landing pages and CTAs

Getting the click is only half the job. Even if AI helped generate an ad visual that led to a click-through, it’s the clear messaging and structure of the landing that closes the loop. That’s where most AI-driven creatives stall.

Creating a landing page that actually converts takes clear messaging, an intentional structure, and a CTA that makes the next step obvious and valuable.

Here’s what we look for:

• The page matches the ad’s promise. If the message feels disconnected, users bounce. And when an AI-generated ad is visually or tonally disconnected from the landing page, it breaks trust.

• The CTA says more than “Buy Now”. It should guide action, whether that’s “See what’s inside”, “Take the quiz”, or “Start your 30-day reset”.

• CRO-informed design matters. AI-generated images might look polished, but they’re often built in isolation with no real connection to the landing page they lead to. Layout, copy, and flow need to reduce friction - not create more.

A beautiful image won’t save a broken user journey. Every element, from the visual to the CTA, has to work together to guide action.

A great ad earns the click.

But a smart and connected landing page and a strong call to action closes the gap between interest and conversion. That’s not something you can automate - yet.

4. Messaging that taps into beliefs

By all means, highlight the features of your products. But more importantly, speak to what your audience cares about: their values, motivations, and deeper pain points.

A feature like “clinically-tested” won’t mean much unless it’s tied to something real, like your customer’s desire to feel in control of their health.

The best creative taps into what people believe, not just what the product does. That emotional layer is where trust - and conversions - are built.

AI still misses the mark here. It struggles to understand the emotional context behind a buying decision, because real insight still comes from real people.

Bottom Line: AI Might Save Time - But It Can’t Replace Strategy

Experimenting with AI? Of course.

You can and should be. But do it in the same way you’d test any other tactic:

Set up a clean test. Track real outcomes. Don’t mistake novelty for value.

Because right now, the real wins we’re seeing still come from:

• Smart, channel-specific strategy: Knowing what works on Meta isn’t the same as knowing what works on TikTok. Strategy needs to match the platform, the product, and where the customer is in their decision journey.

• Quick, focused iteration: Speed without insight is just noise. Moving fast only matters if you’re learning as you go. The best brands test small, track closely, and double down on what performs.

• Creative instincts shaped by deep platform experience: Content creation is important. But you also need to know and understand how the algorithm of the platform behaves, how audiences engage, and how to build for attention and conversion. That kind of insight only comes from spending time in the platform - not a prompt.

AI tools can help speed up your workflow, spark ideas, and reduce production costs in some cases.

But the brands that are consistently winning in 2025?

They’re not the ones chasing the latest tool. At least not without gathering valuable insights first.

They’re the ones who know their audience, trust their testing loop, and use AI to support performance not let it define it.

You Want To Find Out What Actually Works? Let’s Talk

We’re keeping tabs on AI tools. We’re testing them where it makes sense. But we’re not building our clients’ strategies around trends. We’re building around what works and gets results.

If you want performance without the fluff, we’re ready when you are. Let’s talk.

👉 Contact Alpha Inbound