Dentsu, the giant agency holding company, released a report about large brands CMOs and their expectations around AI. A hugely telling statistic was that 70% of CMOs are more interested in cultural velocity than cost savings.

“Cultural velocity” harkens back precisely to Chime (they are a great bank - seriously, my kids love them) CMO/CGO Vineet Mehra’s LI post that we dissected in AIC #103 where we talked about the need for agencies to change from “campaign factories” to “culture engines.

I talked about this on a recent podcast, and the clear take away for all you agency owners out there, isn't AI a great tool for making your margin better? Rather, since AI is making the execution layer easier, the goal is to create advertising & results that are better.

The Dell Pro powered by Intel® Core™ Ultra with Intel vPro® adjusts power usage based on how you work, so your battery life never slows you down.

Explore More

Many years ago, I joined a fantasy baseball league with a bunch of serious nerds that I went to high school with…I was a really good baseball player in high school and college, I wasn’t nearly at the level of quantitive analysis these guys inhabited, but I thought my understanding of the game as a player would level the playing field, so to speak. Fantasy baseball was a new thing for me.

I wasn’t very good at the draft, my team kind of sucked. I offered my friend, Dave, what I thought was a really good trade. I was giving up a lot of pretty good players for one really good player. Dave smacked me down hard:

When using AI, agencies so often fall into the 50 pounds of shit thinking: "If we can make one ad that is kind of good. We can make 1000 variations that are in the neighborhood of good." And since the incremental cost of developing those 999 other ads is incredibly low, we keep on shoveling the shit. We start delivering variations of ads that are further and further AI-pilled, and we forget that the job is to create an ad that connects brand and audience. I’ve seen multiple versions of these two ads, and they clearly came from the "“we will just make more ads…one of them is gonna work” school of thought. These ads have the thoughts that live inside of a good ad, but they've been abstracted into this vague, fluffy, round-sounding AI-speak:

I don’t understand what either of these ads are trying to say. But I am sure that, between all of the context, the creator of the ad had the obsequiousness of the LLM and their desire to get something out into the world allowed them to create ads that don't serve the brand - they are somewhere between 10 and 50 pounds of AI-percolated shit. Compare those ads to these American Marketing Association’s Top Ads of All Time:

and sure, the two marketers I showed above didn't have the budget or the audience of Apple and Nike. That's not the point. The point is that these ads were created with the intention to communicate brand values to a targeted brand audience, and production was hard, and expensive.

Those two constraints forced those agencies to create better ads. The two unnamed marketers above did not have those same constraints. They dumped some notes into a chatbot and burned a few tokens.

When we remove constraints, we can't also remove quality.

The DTC beauty category is crowded. To break through, Jennifer Aniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.

Learn more

in a lot of my agency coaching, we are often talking about this theory of constraints, where you identify a constraint and remove it to increase your capacity. For instance, the constraint might be team availability. So, we double down on making processes efficient, and we remove the constraints of capacity.

But when it comes to creative, or messaging, or even search ad copy, create the constraints that ensures your output isn't a 50-pound bag of shit. Generating bag after bag of shit might be really great for your margins. It’s not going to be great for your impact.

The constraints that you have got to hold in place throughout the creation process is a firm grasp of brand promise and audience experience. I don't mean their experience with the ad, I mean the experience that the audience has every single day:

By holding these constraints in place, you are forcing the brand promise through a narrow opening because your client only has a narrow opening to its audience. Unfortunately, AI in its state today is not especially good at narrow. It is marvelously good at patterns, generalization, and globalization. That predilection can be amazing for efficiency.

But remember that 70% of CMOs value cultural velocity over cost efficiency. That number needs to be a wake-up call for you. I've been in this industry since 1997, and our true focus for the last three decades has been getting more efficient. As luck would have it, we were pushing efficiency into a market that was growing exponentially, so all sorts of good stuff happened.

But today, even though digital marketing is increasing in its amount of spend and there are more participants in the market than ever before, digital engagement is not growing fast enough so that being the best goddamn efficiency optimizer on the planet is worth what it once was. Now AI tools can actually create significantly more efficiency through large-scale data analysis than ever before. Previously, marketing was held back by a data constraint and a deliverable constraint. The data constraint is being removed, and it feels like the deliverables constraint is being removed, too.

But in fact, it's just the difficulty of deliverable creation that has been removed, and since the deliverables are so easy to manufacture, there is much less quality control. I implore you: as a marketer, as an agency owner, as a writer, as a creator, create the constraint of quality, because quality will be the only differentiator between you and any other agency or a $100/month Claude subscription in the hands of an in-house marketer.

Agency Field Guide Stat of the Day: Agencies that have 4 or more defensible numeric results in their case studies and have 5 or more named case studies are 11x as likely to be a top tier platform partner like HubSpot Elite Partner or Klaviyo Master Elite. See how your agency stacks up against the 10,097 agencies in The Agency Field Guide.

Parallel, Turbopuffer, and Wordsmith run their entire GTM motion on Attio, with agents that chase every buying signal, build pipeline, and move deals forward, 24/7. Try Attio now.

Our sponsorships are ads (obviously - duh). But some of the companies or services linked in our content may pay us a minimal commission if you buy something from them. Potential payment is never a factor in what we link to, and we have direct experience with every company or service where you can buy something that we mention in our content.