The Best Marketing Tool Isn't One Tool

By James Arnold
Rooster Chief Digital Officer

Ask a group of marketers what tools they can't live without and you might expect to get a list. ChatGPT. Claude. Gemini. Waldo. Fireflies. Gamma. Sprout Social. Meltwater. Google Ads. Zapier. Copilot. We did get a list.

But after asking that question throughout AMiDA 2026, I think the list may be the least interesting part of the answer. What stood out was how differently marketers are putting these tools to work. They aren't necessarily looking for one platform that does everything. They're combining tools, matching them to particular jobs and increasingly connecting them into workflows that can take work from one step to another.

That might be the more important lesson for marketers trying to make sense of a technology landscape that seems to change every week.

Stop looking for the one AI tool

There was certainly no consensus among our AMiDA guests about which large language model marketers should use. And that's probably a good thing.

“We use a combination of Gemini, Claude, ChatGPT and a few others because we realize every LLM has their strengths and weaknesses,” one of our guests told us.

Another described ChatGPT as a daily tool, particularly for research. Others preferred Claude for deeper research and analysis. Copilot came up for working with PowerPoint. The point wasn't which one won. It was that different tools were finding different jobs.

Bethany Andrew of State of Marketing described using ChatGPT for writing ad copy or simply dumping thoughts from a meeting into it and asking for “a nice polished, buttoned-up summary” with the action items highlighted. She turns to Claude for deeper research and industry understanding.

That's a pretty good model for marketers right now. You don't necessarily need to pick a team. You need to figure out which tools make you better at the work you actually do.

Research is becoming a team sport between tools

Waldo came up several times in our conversations, specifically with Goodway Group, and the way they described using it was particularly interesting. Waldo can aggregate research from multiple sources and allow marketers to interact with that information conversationally. But our guests weren't necessarily stopping there. One described using Waldo to gather information and then plugging that information into ChatGPT “to further bake that out” and explore how it could shape messaging, content or additional research. Another uses Waldo to build insights reports, understand market activity and competitors, and help explain not only what happened in marketing measurement, but why.

The payoff is time.

“What it's really doing is saving us a lot of time from just searching all those research databases that have existed for 100 years,” one speaker explained.

That's an important distinction in the AI conversation. AI hasn't suddenly made every specialized research platform irrelevant. Instead, we're beginning to see specialized tools and general-purpose AI working together. One finds and organizes the evidence. Another helps interrogate it, synthesize it and figure out what to do with it.

That is a considerably more powerful workflow than simply asking ChatGPT a question.

The really interesting part is when the tools start talking to each other

Ryan Green took this idea considerably further. His meetings can be captured by Fireflies, with the transcript then becoming part of what he described as his “central nervous system” in Claude. From there, he can ask Claude to identify the things that need to happen next and begin doing the work.

“There were three to-dos that we needed to do out of that,” he explained of one meeting. “One of them was to build this landing page. Here's the creative for it. The second one was to create a content calendar for the next six months.”

Instead of going back through notes, finding the relevant conversation and rebuilding the context, the context is already there. Add tools like Zapier and that idea gets bigger. Paid search. Creative. Meetings. Monday.com. Slack. Research. Project management. They don't have to remain separate islands of information that require a person to continually carry work from one system into another.

Ryan's advice was essentially to connect the plumbing first. That might be one of the most immediately useful ideas from this entire episode.

For years, marketers have built technology stacks by adding platforms. The next phase may be less about adding another box to the stack and more about connecting the boxes we already have.

Specialized tools aren't going away

With all the attention going toward AI, it would be easy to assume that general-purpose AI will eventually swallow the rest of the marketing technology stack. Our conversations didn't suggest that at all.

Sara Wade talked about the creator platforms her teams use and the value of Sprout Social and other specialized influencer tools. These platforms have access to data and functionality that marketers need to manage creator programs. AI's role isn't necessarily to replace that.

“It's not that they are doing the work for you,” Sara said. “The AI is helping you collect the data and analyze it in a way that would take ... days and weeks to go through every creator post.”

That's a very different proposition. The specialized platform understands the creator ecosystem and houses the data. AI helps marketers work through that information at a scale that wasn't previously practical.

We heard a similar idea around media intelligence platforms such as Meltwater. These are incredibly capable systems, but that doesn't mean every organization needs every capability. The question becomes much simpler: What do I actually need this platform to do?

As one speaker put it, marketers should figure out “what am I going to use, what's really going to help me,” and then buy the tool that matches those needs.

That sounds obvious. In a world overflowing with marketing technology, it probably isn't.

Sometimes the breakthrough is ridiculously simple

Not every useful change requires an elaborate AI workflow. One of my favorite examples from these conversations involved a microphone. David Maineiro from AI Digital bought one despite not being a podcaster and having little need to record anything. The reason was much simpler. He created a keyboard shortcut, presses a button and talks to the AI running on another screen.

“And I talk and talk and talk, and it's not necessarily that well structured,” he said. “And I sometimes just say, like, organize this. Organize these thoughts. What should I do next?”

That's it. The innovation isn't the microphone. It's recognizing that talking through a complicated idea may be faster than carefully constructing a prompt. AI can take the messy thinking and help create the structure afterward.

Sometimes improving your workflow doesn't require changing the technology at all. It requires changing how you interact with it.

Start where you know what good looks like

The easier these tools become to use, the easier it also becomes to produce something that looks pretty good without actually being very good. Ryan offered what may be the most practical advice for marketers who are still figuring out where AI fits into their work.

“Start using AI in your field of specialty first because you will know what is [bad] and what's not.”

He uses AI across the paid search campaigns he manages, but he also brings roughly 15 years of experience with the platform to that work. He knows how match types have changed. He knows where the problems can hide. He understands what the output is supposed to accomplish. AI gives him leverage because he can judge what it produces. That's an important part of this conversation that can get lost when we focus entirely on efficiency.

AI can accelerate research that is incomplete. It can beautifully summarize the wrong idea. It can generate code that doesn't work. It can create an advertising campaign without understanding the business behind it.

The tool can produce the work. Someone still has to know whether the work is good.

And some things still belong to people

That matters even more when the work depends on human experience. We heard it clearly in our conversation about AI-generated influencers in agriculture.

“You're never going to be able to build true authenticity with an AI creator because they've never been where you've been,” Wade said.

An AI-generated farmer may look right. It may wear the boots and drive the truck. But it doesn't know about “the generational scrap metal pile out behind the barn.” It hasn't sacrificed a pair of shoes to the dairy barn. It hasn't lived the experiences that create the connection between one producer and another.

Technology can replicate the appearance of that experience. That's different from having it. And for marketers, knowing the difference is going to matter.

Keep testing

Maybe that's why one comment from Bethany Andrew is a good place to end this entire AMiDA series.

“I am still testing. I'm always learning, and that's, I think, really important for marketers.”

There isn't going to be a final marketing technology stack. There isn't going to be one AI model everyone settles on. The tools we are talking about today will improve, merge, disappear and be replaced by things we haven't seen yet. So test them.

Use different tools for different tasks. Connect them when it makes sense. Learn where they save you time. Learn where they make you better. And be equally clear about the places where your expertise, judgment and experience still need to lead.

The marketers who get the most from this generation of technology probably won't be the ones with the longest list of tools.

They'll be the ones who understand what work belongs to the technology and what work still belongs to them.