I sat down with Kara Redman from Backroom agency to discuss things like using AI for research and discovery (not creative output), why your proprietary IP is the only lens that makes AI useful, the growing contradiction between efficiency and human point of view, AI transparency as an ethical obligation for agencies, and why discerning taste remains the ultimate premium in an AI-saturated market. Plus: the “software with a service” model, how enterprise clients are still lost on AI adoption, and why in-person dialogue beats chatbot-filtered conversations.
About the guest
Kara Redman is the CEO and founder of Backroom, an award-winning brand strategy and activation agency headquartered in Baltimore. Backroom has built a reputation as a “people-first” agency known for its collaborative culture and high-quality work. The firm’s small team partners closely with clients — from local startups to national lifestyle brands — to craft compelling brand narratives and targeted campaigns across social media, web, and paid channels.
Backroom specializes in brand strategy and portfolio architecture, helping companies through M&A transitions, rebrand strategies, and market positioning. Kara’s philosophy: make the client look good, not the agency. Her mission is to eradicate sameness.
→ Connect with Kara on LinkedIn, her site, or read her newsletter.
About the host
Peter Benei is the co-founder AI-Ready CMO, a daily AI marketing intelligence platform for senior leaders. Peter has been serving as a CMO, marketing leader, and consultant to high-growth B2B scaleups for the past 10+ years. He has a background in advertising, working with Fortune 500 brands.
→ Connect with Peter on LinkedIn or read his newsletter.
Top 10 Takeaways
AI compresses research, not creativity. Brand discovery that used to take weeks now takes days or even hours. AI scrapes earnings calls, analyzes competitor data, and synthesizes CRM inputs. But the strategic recommendations and creative output stay human.
Your IP is the lens. AI is the accelerator. Without a proprietary framework to filter client data through, you’re just the middleman between a client and a robot.
The middle is getting squished. Mediocre content creators will produce mediocre content faster. Those with real strategic thinking and creative IP will use AI to think and critically analyze faster. The gap widens.
Most companies are still in discovery mode. Despite the LinkedIn narrative, clients are mostly arriving with vague directives from leadership. “We need to use AI.” That’s as far as they’ve gotten.
AI layoffs are largely an excuse. Like the pandemic before it, AI is being used as a cover for workforce reductions that were happening anyway. Senior hiring is growing. Junior hiring is not.
“Software with a service” beats “software as a service.” The human account manager who knows when data is no longer relevant is the actual value layer. Standalone tools strip away the judgment premium.
AI content feels like a betrayal of audience trust. If you built a following on your authentic voice, optimizing that through AI feels like a bait-and-switch. Kara reworks 80% of anything AI touches.
Input matters more than output. Everyone wants AI to produce. The competitive advantage is using it to consume: forming a point of view through conversations, reading, and critical thinking. Then the content is the easy part.
Human connection is becoming premium. Every day that passes, in-person dialogue, unfiltered conversation, and real relationships feel more valuable. The more AI mediates, the more human presence differentiates.
Taste cannot be automated. If you can’t produce premium output as a human, AI just gives you tools to do mediocre work at scale. Craft comes first. Technology serves it.
The 5 Topics Worth Your Attention
1.
The Real AI Use Case for Agencies Is Research, Not Output
The conversation revealed something most agency leaders won’t say publicly: AI’s biggest impact on high-level brand work isn’t in what it creates. It’s in what it helps you find.
Kara’s agency works with publicly traded companies on brand strategy and portfolio architecture. The kind of work that requires deep market intelligence, perception analysis, and competitive mapping. Before AI, that research phase alone consumed a minimum of eight weeks.
Now, Kara can scrape earnings call transcripts, feed them into AI, and extract specific market signals in a fraction of the time. Combined with social listening tools and CRM data, the synthesis happens faster. Not the thinking. The synthesis.
“We use AI to do anything that it can do better than we could, and we still do the things that we do best.”
Kara Redman
The distinction matters. AI isn’t replacing the strategist. It’s replacing the analyst grunt work that used to bottleneck the strategist. A three-month engagement can now become six to eight weeks, not because the thinking got cheaper, but because the data preparation got faster.
For marketing leaders evaluating how AI fits into their agency relationships or internal teams, the takeaway is clear:
Ask where AI accelerates research, not where it replaces judgment. The agencies that are honest about this distinction are the ones worth working with.
2.
Without Proprietary IP, You’re Just a Middleman Between a Client and a Robot
This might be the single most important idea from the conversation. Kara has spent 12 years building a proprietary brand diagnostic framework with 10 dimensions. When client data goes into AI, it passes through that lens. The framework tells the AI what to look for. Not the other way around.
“You need a lens through which to pass that information. And for us, it’s our IP.”
Kara Redman
Without your own methodology, AI becomes a random answer generator. It will analyze anything and give you something back. But without guardrails, without a diagnostic model, without years of pattern recognition baked into a framework, you get generic consulting output that could apply to any company.
The AI also surfaces nuances that human reviewers might miss. It picks up on recurring language patterns in client transcripts. It identifies words and phrases a client uses frequently that might indicate deeper brand positioning opportunities. But it only does this because someone designed the right questions to ask.
This has direct implications for any marketing leader building their team’s AI capabilities. The question isn’t “which AI tools should we buy?” The question is “what’s our proprietary lens for interpreting what AI tells us?” If you don’t have one, you’re not using AI. AI is using you.
3.
The Human-AI Contradiction: You Need Fewer People, But You Can’t Lose the Human Point of View
One of the most honest moments in the conversation was about a CMO who cut a 15-person team down to two or three, but deliberately kept a few people whose work could technically be done by AI. Not for the execution. For the point of view.
“You have to have humans at the workplace to get that point of view, even though most of the work that they do still could be done by AI.”
Kara Redman
This is the contradiction nobody’s solving cleanly yet. AI can eliminate most of the repetitive execution work in a marketing department. Social posts, research summaries, competitive audits, and first-draft content, all automatable.
But the moment you strip out the humans who produce that work, you lose the diversity of perspective that makes strategy meaningful.
Kara hasn’t reduced her team size due to AI. And she’s clear-eyed about why: the level of work they do is high-level strategic thinking. That’s hard to replace. But she also acknowledged the tension, particularly for larger organizations where the temptation to cut headcount is stronger.
The practical implication for CMOs:
Be very careful about what you optimize away. If you eliminate every role that touches execution, you also eliminate the people who develop the judgment and taste that eventually make them great strategists.
The junior marketer writing social posts today is developing the pattern recognition that makes them a valuable brand strategist in five years. Cut that pipeline, and you’ve saved payroll this quarter while destroying your talent bench for the next decade.
4.
AI Transparency Isn’t Optional Anymore, It’s the Next Privacy Policy
The conversation surfaced an emerging ethical obligation that most agencies are ignoring: AI transparency with clients. Not just “we use AI sometimes,” but a structured disclosure of where AI fits in the workflow, what data gets fed into it, and what the human oversight layer looks like.
Kara treats every client’s data as if it were in a highly regulated industry. That’s the baseline. But the transparency question goes further than data handling.
“I think it can become very easy to go, I’m not going to think about this. I’m just going to upload everything and then see what it spits out.”
Kara Redman
I pushed this further, arguing that agencies need something like an AI transparency policy, similar to a privacy policy, that explicitly states how AI is used in client work.
Not because clients are demanding it yet, but because the trust foundation matters. Especially for agencies doing high-level brand work where the CMO has a deeply personal relationship with their brand.
Some people don’t know that the AI-generated content they received isn’t good. They think it’s good. A transparent framework protects both parties. The agency from accusations of cutting corners, and the client from unknowingly receiving outputs that lack the human judgment they’re paying for.
For marketing leaders on both sides, whether you’re hiring an agency or running one, this is worth getting ahead of. The companies that build AI transparency into their client relationships now will have a structural trust advantage when this becomes a standard expectation.
5.
Taste Is the Last Premium, and AI Can’t Manufacture It
The final and perhaps most resonant theme: discerning creative taste, real style, the ability to feel whether a brand direction is right, these remain stubbornly human. And they’re becoming more valuable, not less.
“I think that discerning taste, creative taste, and style have always been premium. And no tool’s really going to be able to replicate that well, at least in the short term.”
Kara Redman
Kara’s argument isn’t anti-AI. It’s anti-shortcut. If you can’t produce premium creative output as a human, AI doesn’t fix that. It just lets you produce mediocre work faster. She said it plainly: AI is not a shortcut. If you already can’t create anything meaningful, it isn’t going to fix that for you. It’s just going to give you tools to do ugly work at scale.
The market analogy is instructive. There’s always been a market for mediocrity. But premium has always coexisted with it, and premium has always commanded outsized returns. AI doesn’t change that dynamic. It amplifies it.
For senior marketers, this reframes the AI conversation entirely. Stop asking “how can AI make our content faster?” Start asking, “Do we have the taste and judgment to know when AI output is actually good?” Because the answer to the first question is obvious. The answer to the second one is what separates brands that win from brands that just produce more.
The cream will rise to the top. It always has. The question is whether your team has the palate to tell the difference.
Found value in this conversation?
Share it with a marketing leader who needs to hear this.
Subscribe to AI-Ready CMO to catch future live episodes and get daily AI marketing intelligence that actually matters.
