Marketing is changing fast, and the winners are the teams who build real feedback loops with AI. After watching Kipp Bodnar and Kieran Flanagan on Marketing Against the Grain, I’m convinced: the smartest move right now is using AI to audit your own work with brutal honesty. My view is simple. Let machines surface the gaps, then let people make the fixes that matter.
What Kipp and Kieran Got Right
Kipp walked through a practical playbook: teach an AI “what great product marketing looks like,” crawl your site, and have it grade you against that standard. That’s not theory. That’s a working loop. Their finding? Core product pages performed well, but the second-tier feature pages lagged—things like sales forecasting and analytics needed upgrades and could be automated.
“Our big time product pages are doing really well. It’s our feature kind of second-tier product pages where we could have a lot of improvement.”
“I asked it to build a ranking algorithm and score everything… Tell us what we’re doing well and… our weaknesses.”
This is the kind of clarity most teams never get because internal audits carry bias. People protect their own work. AI, when trained well, doesn’t care who wrote the page. It ranks. It flags. It pushes focus where revenue leaks.
The bold point: AI audits are not a gimmick; they are a management edge. They free leaders from weekend marathons in spreadsheets, and they give teams a ranked to-do list that ties to outcomes.
Why This Matters Now
In crypto, SaaS, and media, I’ve seen the same pattern for years: teams polish the flagship page and ignore the long tail. That long tail still sells. If your comparison pages, feature explainers, and integration write-ups are weak, you lose deals you never see. Kipp and Kieran are shining a light there.
“You can just give this to another agent who actually goes off and fixes them based upon this.”
That’s the second shift: audit-to-action. The next agent drafts the fix. Your team reviews, tunes, ships. The loop is: define quality → audit at scale → auto-draft fixes → human edit → ship → re-score.
Where I Add My Playbook
I’ve built and marketed products for decades across web, social, and blockchain. The lesson is the same everywhere: set the standard before you scale the audit. If you don’t define “great,” the AI will optimize for average.
Here’s how I’d run this, starting tomorrow:
- Write a short rubric for “great” product marketing: clarity, proof, outcomes, structure, CTAs.
- Train an AI with examples of your best pages and top external benchmarks.
- Crawl every product, feature, and comparison page—score each item against the rubric.
- Sort by impact and effort. Fix high-impact, low-effort pages first.
- Spin up an agent to draft upgrades; have humans edit for voice and accuracy.
- Re-score after publishing, and track lift in time on page, demo requests, and trials.
That list turns theory into motion. It also keeps teams from arguing over taste.
But What About AI Bias?
There’s a fair worry that AI brings its own bias. True. But the fix is simple: tune your rubric, include diverse examples, and keep a human in the loop. The bigger bias is pretending your current pages don’t need work. As Kipp hinted, teams often downplay lower-priority sections. That’s where revenue hides.
“If you’re running a team… you can have incredible insight into the work you’re doing… without having to spend… your entire weekend going through stuff manually.”
Leaders who adopt this loop will out-ship and out-learn their rivals. The rest will keep guessing while their funnel leaks through weak feature content.
The Signal Inside the Noise
Here’s what stood out most from Kipp and Kieran’s take: they didn’t chase hype. They used AI to do a boring, valuable job—find weak content, rank it, and fix it. That’s the move. No flashy demos. Just better pages that win more deals.
As someone who’s shipped products since dial-up days, I’ll take that any week. The internet still rewards clear value and steady execution. AI just makes the feedback loop faster.
My Call to Action
Build your “great” rubric. Run the audit. Sort the backlog. Ship weekly. Re-score. Repeat. Don’t wait for a quarterly review to tell you what an agent can surface in an hour.
Stop guessing. Let AI show you where the money is leaking, then fix it with speed. Your team’s best work is on the other side of that audit.
