Key Takeaways:
- Applying AI to everything at once isn’t a strategy — brands that scale well pick one specific lever to fix, not all three at once.
- Media efficiency, creative testing, and audience targeting are distinct problems, each needing its own data and success metric — treating them as one issue produces modest gains, not meaningful ones.
- Diagnosis matters more than the tool — brands often reach for a targeting fix when the real issue is unclear creative, wasting months of budget on the wrong lever.
- Before applying AI, answer three questions: what specific metric are we moving, is our data clean enough to learn from, and who owns interpreting the results.
- Media efficiency gains from AI have a ceiling — once bidding and targeting are optimized, creative that evolves with audience response is what actually drives growth past that point.
—
DTC AI scaling is everywhere right now, and the brands doing it well share one thing in common: they picked a lane. The brands struggling? They applied AI to everything at once and called it a strategy. That is not a strategy. That is noise with a monthly retainer attached.
Here is the direct answer: AI can improve media efficiency, creative testing, and audience targeting, but treating them as one problem is why most DTC brands see modest gains instead of meaningful ones.
The Real Problem Is Not the Tool. It Is the Diagnosis.
Every week, another DTC brand announces they are “using AI to scale.” And they are right, technically. But scale toward what, exactly?
AI applied to media buying without a clear efficiency problem to solve is just automation for automation’s sake. AI applied to creative without a testing framework produces faster content, not better-performing content. AI applied to audience targeting without clean first-party data is a confident system making confidently wrong decisions.
The tool is not the issue. The diagnosis is.
What the Three Levers Actually Mean
Before a brand decides where to apply AI, it helps to define what each lever actually does, because the industry conflates them constantly.
Media Efficiency is about where your dollars go and what they return. This is bid strategy, budget allocation, pacing, and channel mix. AI can genuinely improve this, but only if your measurement foundation is solid first.
Creative Testing is about learning which messages, formats, and visuals drive action with specific audiences. AI accelerates the volume of testing and can surface patterns humans would miss. But it still needs a human to define what “winning” means for the brand.
Audience Targeting is about reaching the right person at the right moment. AI-driven targeting has gotten significantly more sophisticated, but it is only as good as the data fed into it. For most DTC brands, that data story is messier than they admit.
Where DTC Brands Get This Wrong
The pattern we see most often at Junction 37 is a brand that has a creative problem, but reaches for a targeting solution. Their ads are not resonating, so they assume the wrong people are seeing them. They invest in audience segmentation. Performance still disappoints. Then they try a new channel.
Meanwhile, the actual issue was that the creative never communicated a clear value proposition to begin with.
This is not a small mistake. It can burn through months of budget and produce a misleading conclusion: that the channel does not work for the brand. The channel was fine. The creative diagnosis was wrong.
Three Questions to Ask Before Applying AI to Media
Before any DTC brand commits to AI-driven scaling, these questions need real answers:
1. What specific metric are we trying to move? Not “improve performance.” A specific number: ROAS, CPM efficiency, creative click-through rate, cost per new customer.
2. Do we have clean enough data for AI to learn from? Attribution gaps, channel mixing, and inconsistent UTM tagging will teach your AI system the wrong lessons.
3. Who owns the interpretation layer? AI surfaces patterns. Humans have to decide what those patterns mean for the brand. If that accountability is unclear, the output will be unclear too.
If a brand cannot answer all three, AI will not save the campaign. It will accelerate the confusion.
The Efficiency Trap That Catches Growing Brands
Here is a take that does not get said enough: media efficiency gains from AI are real, but they have a ceiling. Once you have optimized bidding, reduced wasted impressions, and tightened your targeting parameters, you have extracted most of the efficiency available to you.
What grows a DTC brand past that ceiling is creative. Specifically, creative that evolves with what the audience actually responds to, not what the brand assumes they want.
We wrote about this directly in our piece on why your brand needs a creative intelligence system, not just AI. The short version: efficiency gets you to a baseline. Creative gets you growth.
And if you are curious about how AI and human judgment interact across the full media buying process, our breakdown of where machines stop and humans start in media buying is the most honest answer we have published on that question.
What Good DTC AI Scaling Actually Looks Like
When we see it working, it looks like this:
- A brand identifies one specific bottleneck, not a general desire to “be more efficient”
- AI is applied to that bottleneck with a clear hypothesis: “If we automate bid adjustments by time of day, we expect CPMs to drop by X%”
- Results are reviewed by people who understand the brand’s market position and margin structure
- Learnings feed back into the next hypothesis, not into a dashboard that nobody reads
It is methodical. It is humble about what AI can and cannot do. And it is led by humans who know the brand well enough to catch when the machine is optimizing toward the wrong goal.
That last part matters more than any platform feature or AI upgrade. The question is never just “what can AI do?” It is “what do we need, and is AI actually the right tool to get there?”
—
Ready to identify the right media lever for your brand? The team at Junction 37 works with DTC and CPG brands to build performance media strategies that are precise, not just automated. See how we work or get in touch directly.
—
Frequently Asked Questions
What does DTC AI scaling actually mean?
DTC AI scaling refers to the use of artificial intelligence tools to grow a direct-to-consumer brand’s reach, efficiency, or output without proportionally increasing cost or headcount. In media, this typically applies to bid automation, creative production, or audience segmentation. The term is used broadly and often imprecisely.
Why do most DTC brands struggle with AI in media buying?
Most brands apply AI before diagnosing the actual problem. They treat media efficiency, creative performance, and audience targeting as one combined issue when each requires a different approach, different data, and different success metrics. Applying AI to the wrong lever produces faster spending, not better results.
Should AI replace human judgment in performance media?
No. AI is most effective when it handles high-volume pattern recognition and repetitive optimization tasks. Human judgment is essential for interpreting what those patterns mean, setting the right objectives, and making brand-level decisions that require context a machine does not have. The two work best together, with clear roles defined upfront.
How do I know which AI lever is right for my DTC brand?
Start by identifying the specific metric that is underperforming and trace it back to its root cause. If your creative click-through rates are low, a targeting solution will not fix that. If your CPMs are high relative to benchmarks, creative volume will not solve it either. Diagnosis comes before tool selection, every time.
—
Chris Pyne, Founder, Junction 37 – 30+ Years in Performance Media.