Brands have poured millions into media infrastructure, targeting models, attribution stacks, programmatic pipes, and almost nothing into understanding why their creative works. That’s the creative intelligence gap. And generative AI didn’t create it. It just turned the lights on.
At Junction 37, we see this every week. A brand comes in with solid media fundamentals and creative that’s essentially running on gut feel and brand guidelines. When we start asking questions: what message drives the most qualified traffic? Which visual format performs best at the bottom of funnel? What’s the actual cost of a bad creative decision? The room goes quiet.
The data doesn’t exist. Because the system was never built.
Why Generative AI Is Making This Worse Before It Gets Better
Here’s the uncomfortable truth: giving a brand with no creative intelligence system access to generative AI is like giving someone who can’t drive a faster car.
You don’t get better outcomes. You get more of the same mistakes, faster, at higher volume.
We’ve watched brands use AI tools to flood their ad accounts with creative variants – 40, 60, 100 versions of an ad — without a single hypothesis behind any of them. No framework for what they’re testing. No taxonomy to track learnings. No process to feed results back into the next round.
That’s not creative intelligence. That’s creative noise.
What Creative Intelligence Actually Means
Creative intelligence is the systematic ability to connect creative decisions to business outcomes.
It includes:
- Creative taxonomy — a consistent naming and tagging system so you can actually analyze performance by format, message, visual style, or offer type
- Hypothesis-driven testing — every creative variant exists to answer a specific question, not just fill an ad set
- Signal capture — knowing which metrics (thumb-stop rate, hold rate, CTR, ROAS, downstream conversion) matter at which stage of the funnel
- Iteration loops — a defined process for taking creative learnings and applying them to the next brief
- Cross-channel coherence — understanding how creative performs differently on Meta vs. TikTok vs. YouTube and building for each, not repurposing one asset everywhere
Most CPG and DTC brands have none of this. Some have pieces. Very few have the full system.
Why Performance Media Agencies Are Now On the Hook
This is where it gets interesting for agencies like us.
Brands aren’t just asking their performance media partners to buy media anymore. They’re asking us to be the connective tissue between creative output and business results — because their internal teams and creative agencies aren’t built to do it.
That’s a reasonable ask. But it requires honesty about what performance media strategy actually involves. It’s not just optimizing bids and audiences. It’s building the feedback loop that tells your creative team what to make next.
At Junction 37, our performance media work is built around this directly. We don’t separate media from creative intelligence. We treat them as the same system. Because in a performance context, they are.
The 3 Mistakes CPG and DTC Brands Make Right Now
1. Mistaking creative volume for creative learning.
More ads running doesn’t mean more insight. Without a testing framework, you’re just spending more to learn less.
2. Treating creative as a production problem.*
Brands hire production resources when what they actually need is an analytical framework. Creative intelligence is a strategy function, not a vendor management function.
3. Letting the platform make creative decisions for you.
Meta’s Advantage+ and Google’s Performance Max will optimize toward something — but that something isn’t always aligned with your brand or your actual business goals. Research from the Data & Marketing Association consistently shows that human-guided creative strategy outperforms fully automated approaches on brand metrics that matter long-term.
The Fix Isn’t a Tool. It’s a System.
Generative AI will keep getting better at producing creative. That’s not the question anymore.
The question is whether your organization can tell the difference between creative that’s good and creative that works — and build a system that closes that gap continuously.
That takes human judgment. It takes analytical rigor. And it takes a partner who treats creative intelligence as a core capability, not an afterthought.
If you’re running performance media without a creative intelligence infrastructure underneath it, you’re optimizing the wrong thing.
Let’s fix that. Talk to Junction 37 about building a performance media system that connects creative to outcomes.
FAQ: Creative Intelligence for CPG and DTC Brands
What is creative intelligence in advertising?
Creative intelligence is the ability to systematically connect creative decisions — messaging, format, visual style, offer — to measurable business outcomes. It requires a consistent taxonomy, hypothesis-driven testing, and defined iteration loops.
Why do most CPG brands lack creative intelligence?
Most CPG brands invested heavily in media measurement infrastructure but treated creative as a brand or production function rather than an analytical one. There was no system built to capture and apply creative learnings at scale.
Does generative AI solve the creative intelligence gap?
No. Generative AI produces creative faster, but without a creative intelligence system, it produces more untested, untracked creative at higher volume. The gap widens unless the underlying strategy infrastructure is built first.
What should a performance media agency do about creative intelligence?
A performance media agency should build the feedback loop between creative output and business results — including creative taxonomy, testing frameworks, and signal capture — not just optimize bids and targeting.
Chris Pyne, Founder, Junction 37 – 30+ Years in Performance Media