Is AI doing to marketing fees what the internet did in 2000?

I remember the meeting in the fall of 2000 vividly. I was sitting at the large meeting table in the glass and steel San Francisco office of a prominent web agency. As the "Experience Lead," I was in these management sessions, and the rest of the leadership team was present. Our usually upbeat Managing Director entered with a serious face and spoke as he walked: "We need to talk."

From the front of the room, he turned and delivered the news. "I've learned that some other web agencies are now selling consulting engagements for less than $1 million. And worse, another competitor is offering these engagements with no fees to build up their client list." On that day, he was the messenger of doom, announcing the collapse of a high-dollar phase of web consulting. The .com era was at an end.

A different pricing model

Back then, we didn't typically bill by the hour. We used value-based pricing: what did we feel a service was worth in the marketplace? That's what we quoted, and usually billed. We'll come back to this concept.

Fast forward to today. Like many marketing consultants, I am onboarding new AI tools at a rapid pace. Last week, I had another (possibly crazy) business concept, and once I had a brand name, I wrote, designed, and launched a new brand and website within 3 hours. I used it as an experiment to build a brand and site fully with AI, and it worked remarkably well. (This is also a message of doom for consumer web CMS products.)

With client work, I now write estimates that often end up being too high once the work is actually built. This math issue occurs when projects estimated with an hourly model. Today, work timeframes are collapsing, and that's creating a potential revenue crisis for these types of engagements. I don't want to race to the bottom with fees cut in half by these tools. And then there is the overhead we all now carry: monthly tool subscriptions, API usage fees, token costs. These add up fast and are invisible to clients.

The numbers tell the story

Hourly billing carries a potential efficiency penalty for any consultant. As consultants become faster and more skilled, they bill fewer hours and earn less money. That's the trap. The same analysis of the 2026 consulting market notes that AI has made basic execution cheaper than ever, while digital marketing strategy has gotten more expensive because the complexity has skyrocketed. 

The broader consulting industry is already restructuring around this reality. The work that used to take a team of six now takes two people with the right martech stack. But the rates clients pay have not moved in step. Of the consultants who bill hourly, 79% say they want to raise their fees. The tension between those two facts is where most of us are living right now.

Back to the early internet years

Value-based pricing worked well in the .com era, thanks to healthy demand and clients who had no idea how to benchmark what good web work should cost. Can it work again in today's AI reality?

I think the answer is yes, but with some modifications. The most significant trend in 2026 is the shift toward value-based pricing tied directly to measurable business outcomes, with fees typically structured as a percentage of cost savings or revenue increases attributable to AI initiatives. According to recent research, 73% of consulting clients now prefer pricing models tied to measurable business outcomes rather than time spent. Leanware

For agencies, the consulting formula targets roughly 10 to 20% of the value created, with a 5 to 6x client ROI as the sweet spot, and among $150K+ earners it is the primary model 62% of the time. Flat fees tied to deliverables rather than hours are how the math works when AI has compressed delivery time without reducing the strategic value of the output. A Taskip analysis of AI agency pricing put it plainly: "If AI can handle the execution, then execution is no longer where your value sits." 

Can we demonstrate an edge?

Another key question: can we demonstrate higher value not only in speed, but in how we use these tools? There will be a lot of mediocre AI output to rise above, and it's already out there. The difference in our AI-derived products will usually come down to quality, creativity, and judgment, and we will need to demonstrate that clearly to clients.

For a full-time in-house person, their edge might be commanding so many tools that they function as a one-person marketing team. For independent contractors, this is a more serious reset. Possibly even a moment to think about different business models, alternative career paths, or other changes.

At Idea Ovation, we've responded to this challenge by shifting the value proposition away from execution hours and toward something more durable: AI-powered brand systems and automations that make the client's entire team faster and more consistent. Our Brand AI Systems approach embeds intelligence into branded templates so that anyone inside an organization can create new assets that stay on-brand, reducing their ongoing dependency on agency hours entirely. That's a value you can price against ROI, not a stopwatch.

The joy in all of this

I don't mean to paint a doom-and-gloom story. The creative possibilities for any creative person, and especially marketing professionals, are genuinely extraordinary right now. It's such a joy to easily realize and deploy new ideas like my 3-hour website. The tools are remarkable.

But how do we get paid to keep up with and pay for all of these new tools, when the outputs happen at lightning speed? The same way the best .com era agencies survived the 2000 pricing collapse: by doing work that couldn't be easily commoditized, by charging for judgment rather than hours, and by making sure clients understood the difference. The firms that survived that era weren't the fastest or the cheapest. They were the ones who could clearly explain why what they did mattered.

That's still the answer. It's just a harder argument to make when anyone with a laptop and a few subscriptions can spin up a brand in an afternoon.

more i/o posts  |  about the author

Brand AI is built in partnership with Synaxis