With AI Commoditizing Content, Strategic Judgment Is The New Agency Advantage

Push a button, get a hundred blog posts. As AI upends traditional marketing channels, many businesses are drowning in automated content. The marketing leaders responsible for their strategy are trying to figure out what value they bring now that the rest of the industry has access to the same generative tools. With massive multi-agent runs making basic copywriting the smallest use case, producing more content is no longer a differentiator. The agencies staying relevant today invest in the operational details algorithms can't replicate on their own.

Bryan Fikes is the founder of Bonsai Marketing and an industry veteran who has watched search strategies develop over the last 27 years. He has worked across nearly 300 distinct industries and helped more than 6,000 businesses find their footing online. Building on that foundation, he's currently building a proprietary AI ecosystem, including platforms like Bonsai Brain and BonsaiX, to execute AI search optimization for local service businesses.

"Even with all of the fizz, bang, and pop, the story, uniqueness, deliverability and execution still matter. If you can tell the story correctly, manage the middle, and execute on the back end, it's still the same recipe," says Fikes. His argument is that discovery and engagement have to connect through the full customer journey, from the initial signal, to the click, to the phone call, to the completed service, to the review that brings the next customer in. The signals AI models now use to identify a legitimate business run through the same chain, because both search engines and LLMs are checking whether a real entity exists and delivers on what it promises.

The "so what" test

The specific advantage AI creates varies by operator. Without a strategist directing the work, generative tools produce more of whatever was already being produced, and the businesses cutting costs at the expense of creative quality end up with faster mediocre output at lower unit economics. The dynamic is pushing the market toward operators with specialized taste and strategic judgment, because their work is what AI makes more valuable while commodity output gets cheaper. "It took me 15 or 20 people to execute what I can do myself today, and I was happy if I got 40% efficacy of that deliverable," Fikes says. "Now, you take the knowledge that I have, put it through the AI pipeline, and you end up with 99% of that efficacy."

The same principle applies to how Fikes evaluates whether the marketing is working. His diagnostic is the "so what" test, and he runs it as a structural audit across the full funnel. The metrics Fikes rejects on this basis measure activity without connecting to a business outcome, which is why a busy dashboard can coexist with a flat sales pipeline. "Did we get a thousand people to the website? Did 1.5% of that traffic actually do something? What did we do once we actually captured their focus? Focusing on the top of the funnel and how they got there is obviously very vital," notes Fikes. "But we also can't forget the whole rest of the engine that's got to work well in order to actually make money."

Signals that build AI trust

The way businesses show up in AI-driven discovery has moved past the traditional blue link ranking model. Websites now function as canonical databases that LLMs use to identify legitimate operators, and the signals that carry the most weight in this new environment come from real-world operational history. Local chamber memberships, Better Business Bureau records, Dun & Bradstreet listings, and other institutional relationships that a business has built over years all feed the discovery layer AI models use to decide which businesses to cite. "If you've been a part of a local chamber for 10 years and they are taking a website that's 20 years old and directing traffic to a business that's 10 years old, that's the signal that you're trusted," Fikes explains. "Everyone's got to find the authoritative voice or platform in that space, manicure those relationships, and maintain them."

The other consequence of the discovery model changing is that agencies are transitioning from production shops into strategic advisors. Demand for AI-driven marketing strategy is climbing because businesses are looking for guidance on which technology to use, how to unify workflows across tools, and which lever to pull for the next stage of growth. Fikes' own practice illustrates the pattern, translating operational solutions built for enterprise dental implant clients into working plans for local roofers because the underlying business problems (lead flow, trust signals, measurement across the sales cycle) share enough structure to move between industries.

Fikes runs this consulting model as an external voice for agencies that have hit their operational ceiling on internal capacity. He maintains a tech stack deep enough to filter the flood of new AI tools coming to market, and draws on mastermind programs with Search Atlas partners to keep his bandwidth focused on strategic direction. The qualified traffic he drives through AI recommendations is the result of that focus. "One client is so overwhelmed with the day-to-days that the major stakeholders don't have resources to pull the true guts of what they need. They've hired me to be their voice for a year and basically I become the brevity report," Fikes says. "I'm like that advisor to the president explaining exactly what we need to be doing."

Where DIY hits its ceiling

The consulting model Fikes describes is also a response to the DIY momentum in the market, where free content on Reddit, code generation tools, and AI workflow templates are pushing more businesses to try running their own digital presence. The businesses attempting to replace an agency with an LLM tend to hit an operational ceiling once the surface work is done, because the technology can produce individual outputs without coordinating between them. Fikes's diagnostic move with clients considering the DIY route is to let them attempt it first, so the operational gap becomes obvious before they commit. "I always have them go do it, and then I'll ask the business owner what their next steps are," he explains. "It is that next phase that you just can't replace with a DIY solution."

The specific gap Fikes returns to is the difference between producing a piece of marketing and running the operational chain that turns it into revenue. Filling that gap is where experienced advisers earn their seat, evaluating which technology fits, unifying workflows across the tools already in use, and telling the client which lever to pull next. "On ChatGPT, if I give it the right prompt, I can produce one of the most magnificent home pages you've ever seen, but there are 99 more things to do to actually make that work and deliver a customer to you, and everyone is starting to forget that back piece," Fikes concludes. "A year from now when everyone's tried to do a little bit of it, they will realize they need to keep the professional in the seat that they belong in."