The Funnel Builder Who Said No to AI for a Year — And What Changed His Mind
James Curran spent a year resisting AI for funnel work. Then he caved — and changed how I think about adopting it too.
I just read [the ClickFunnels writeup](https://www.clickfunnels.com/blog/james-curran-ai-[funnel]( https://www.onecommaclub.com/?aff=39183cc3-2122-42a8-9eb8-b295ed7d8554 "Try One Comma Club")-building/) on James Curran's pivot to AI-driven funnel building, and I had to break it down for you. Because his story is essentially the story of every smart marketer I know in 2026 — including a version of mine.
The people who win the next five years of online business aren't the ones who adopted AI first. They're the ones who knew enough about the work AI was replacing to use it correctly when they finally caved.
Who James Curran Is
Curran is a 20-year software developer who pivoted into funnel building around 2022. He's the guy a lot of high-end clients call when they need to migrate to ClickFunnels 2.0 — not because he's the loudest in the space, but because he actually understands the technical layer underneath.
In July 2024, his mentor Dave Woodward suggested he start exploring AI in his workflow. Curran initially refused. His reason wasn't anti-technology — it was that he'd looked at most AI-marketing tools and decided they produced surface-level output without real strategy underneath.
He held that position until late 2024. Then the coding models got dramatically better. And his entire stance reversed.
What Changed (His Words, My Take)
In the article, Curran says AI didn't replace his expertise — it bridged the parts where he was rusty. That's the line I want to sit with.
I've said something similar in basically every ClickFunnels show I've been on over the last 18 months. When I built the SaaS my agency runs with a major health institution, I shipped it on ClickFunnels with custom code and AI doing the gluework — not because AI is magic, but because the strategic and creative thinking still had to come from me. AI just let me do the parts I could already do, 5x faster.
That's the actual AI advantage for marketers right now. Not "AI builds your funnel." It's: AI removes the tax on the tasks you already understood.
What Curran Built (Sentinel)
The most useful concrete thing in the writeup is the software Curran built using AI — he calls it Sentinel. It does three things:
- Monitors funnel pages for load speed
- Tests that buttons actually fire the events they're supposed to fire
- Checks that internal and external links still work
Sounds boring. Run a funnel for six months and you'll realize boring is the entire game. Half the conversion drops I get hired to diagnose come down to "the button didn't fire the right pixel for three weeks and nobody noticed."
Without AI, building Sentinel meant hiring developers — which most marketers can't justify. With AI, Curran built it himself in weeks, using his existing dev background to spot when the AI was hallucinating.
The Other Thing AI Unlocks: Parallel Testing
The other piece of his approach that I want to put a spotlight on: Curran uses AI to generate multiple landing page variations simultaneously for split testing.
Before AI, you'd test one variant against control. Maybe two if you were really committed. Each variant took hours to mock up, write copy for, build, and configure. The cost of being wrong was high enough that you played it safe.
With AI generating the first drafts? You can ship five variants in the same time it used to take to build one. You get much more data, faster. The conversion math compounds.
I'd add my own spin here: the trap is letting AI also pick which variant to keep. AI generates great drafts. Humans still need to read the data and decide what to learn from it. If you outsource the analysis, you'll end up scaling the wrong winner.
What This Actually Means for Your Business
Here's how I'd translate Curran's story into a 2026 action plan if you're running funnels:
- Stop trying to find "the AI tool that does it all." It doesn't exist, and the ones that claim to are mostly wrappers around the same APIs you'd hit yourself.
- Find the bottlenecks in your workflow first. Where do you waste time? Where are you rusty? Where do you keep meaning to test something but don't?
- Use AI as a force multiplier on those exact tasks. Not as a replacement for your judgment.
- Keep the strategic layer human. The offer, the audience, the angle, the why — that's still your job.
If you don't already have a ClickFunnels account to test this kind of multi-variant approach in, that's the platform I run everything on — including the SaaS work I mentioned above.
The Bottom Line
Curran's nut graf in the piece is the one I'd put on a wall: "Your clients don't want to manage AI agents. They want you to use AI to solve their problems faster." That's the entire 2026 service-provider playbook. The marketers who get rich in the next 24 months aren't the ones running AI on autopilot. They're the ones who got 10x faster at the parts of the job they already knew how to do — and still have a brain on top of it.