For Agencies

How $100M+ Leaders Use AI to Hit Milestones?

Atharva Padhye
Atharva Padhye
October 10, 2024
How $100M+ Leaders Use AI to Hit Milestones?

Rise in Ad Spend

In 2023, the ad spend on Google crossed $160 Billion and is growing 8% YoY. Adding Meta, Amazon, and other major ad platforms, this number goes up to $400 Billion.

As John Wanamaker famously put it - over 50% of ad spend is wasted i.e. only half of the advertising actually works. The science is to figure out which half actually works.

This is one of the most well-known but not-so-well-understood problems in advertising.

Despite the existence of thousands of tools be it MTA (multi-touch attribution), MMM (marketing mix modeling), or PPC optimization (pay-per-click), agencies and experts, there is still a huge gap in figuring out the best ways to optimize the $ ad spend.

However, over the last 2 years, most marketers have been able to get better at what they do by using AI for some of their tasks. The most common use cases are:

  1. Creative: Using AI to come up with first drafts of creatives & ad copies.
  1. Analytics: Using AI for calculations, pattern finding & recommendations.
  1. Automation: Using AI to automate repetitive, back-office, grunt-work tasks.

We decided to do our own research by interviewing leaders from all walks of the advertising industry i.e. ad agency founders, CMOs, mar-tech SaaS providers, and hands-on performance marketers.

Here are the top 3 patterns we uncovered:

1. AI is not a magic wand

It’s important to set expectations right

Using this Reddit thread as an example, the most common perception of people is that AI should magically take up all work and help increase revenue instantly. They’re also finding a replacement for the work marketers are supposed to do.

2. AI for Analytics is an underrated use case

Most focus from marketing leaders has been on creative & content generation while analytics & insights use cases are actually proving to be high ROI

An average ad agency takes over 10 years (sometimes much longer) to cross the $100M mark. But new age marketing agencies like NP Digital crossed the $100M mark in 2022 in just 5 years i.e. twice as fast as the rest. 

“But we’re also using AI, how are they growing faster than us?” This question would be most critical to a founder of an agency in business for more than 7-8 years and still struggling to cross 7 figures in revenue.

In this video, Neil Patel is having a chat with Eric Siu about how AI is being used in the marketing space. What they agreed on is that while most agencies are using AI for creative work which is giving sub-par results while the bigger opportunity is to use AI for analytics & insights which should be freeing up time for marketers to do more creative work. 

In fact, when AI is made to do creative work, it’s not creative at all. TED Speaker & Marketer Jessica Apotheker highlights the overuse of AI in this video in the marketing field.

“Generative AI has been trained on existing content and data. Because of that, it reduced the divergence of outcome. And that great equalization of marketing is certainly not a productive outcome.”

~ says Jessica

3. AI will complement, not replace.

The most effective application is to complement a marketer rather than follow the frenzy of replacing marketers with AI. This has worsened the delivery quality to clients as the service lost the human touch which made them personal.

Most leaders look at performance marketing as a combination of Art + Science + Operations.

While the art and science parts can only be partly solved with AI, the operations aspect can be close to 100% automated. This just takes out the repetitive, manual labor that sucks out the creativity of a performance marketer. 

If we’re able to pull this off, a performance marketer would be able to take on more meaningful work, be able to manage more accounts, and run more experiments. 

For an ad agency, this directly translates into better performance thus better retention while reducing the pressure on the bottom line.

In the video with Eric Siu, Neil Patel further discusses the implementations of AI in performance data analytics, you can see this LinkedIn post we published a few days ago to get the context. 

When we were first told about this video, we thought they were talking about GetCrux, because this is what we are also building for performance marketers.

So, if you’re an agency owner who wants to explore performance analytics with AI, schedule a call and talk to us!