Rob Watts

I've been so AI pilled, that I've done a complete 180. AI is a money incinerator. It's a collective hallucination. Everyone sees everyone else shoveling money into the incinerator and starts doing the same. Everyone just keeps shoveling money hoping that one day the engine starts running. And maybe it will. But it isn’t. AI is expensive. Especially when you have to actually pay for tokens. And the real systemic work, the stuff that replaces people because it is autonomous, runs on tokens. The question isn’t whether your AI systems are cheaper than Americans. It’s whether they’re cheaper than people in Latam, Southeast Asia, or Africa. They aren’t. AI also suffers from broken clock syndrome. When we task AI with doing a specific task, it can often give a quality or correct output. But the problems arise when we ask an AI to produce that type of output with different inputs. If you don't know the process it used to give you the initial correct output, there's no way to know if it can take your input type and generate a quality output repeatedly or if it is just a broken clock that happened to be right. So in order to build durable results with AI, you need to build the process the AI is going to use to generate the output. You need to build in oversight. And the more varied the inputs, the more complex the output, the more granularly you need to control the functions within the black box. Otherwise, your fail rate is going to be high. And this failure rate increases whenever there is nested processing and thinking. Then there is the cost of building these systems in the first place. If you actually want something that works, you need to invest considerable resources into the build. Senior talent to develop it, token costs during calibration, ongoing maintenance when it breaks, etc. The common argument against this is to “let AI do the building on its own.” The problem with this is AI doesn’t think. It doesn’t reason. It guesses the next most likely thing in a series. So it will build something for you that will look like a house, but you go in the basement and the foundation is upside down. It’s got wires running through water pipes. The hot and cold water taps are reversed. It looks right from a distance, but any deeper level of scrutiny the illusion falls apart. Nobody is making money from the “work” that AI is doing. If they were, we would have such a rapid consolidation of wealth nobody would know what is happening. Every industry would be consolidated overnight into unassailable monopolies. I hear from lots of people who say they are making money from using AI. And my question to them is how much did it cost to build and what’s the payback period. Large companies are pulling back on AI investment. These are the companies that should be most able to realize a positive ROI on AI. So the question is whether they are uniquely incompetent at development and deployment or are smaller businesses bad at calculating cost and ROI? This results in our current environment: The only people making money in AI are the people who are selling it. It remains to be seen when using AI is profitable and how much work it can profitably replace.

Go use this tool. Check to see if your Meta credit limit is high enough to support your BFCM spend growth this year. No email gating. No signups. Just a free tool to check if your limit is high enough.

Maybe a hot take, but I think many brands are too precious about what ads get launched. I'm not saying go blast 1000 AI slop ads into an account, but I've found that some brands are so worried about getting an ad 100% the way they want it instead of just launching more ads. The imagined worst case scenario where a really bad ad that doesn't perform gobbles up all the spend basically never happens. The most likely situation for any ad that gets launched is that it gets a little bit of spend and average performance. It's a background noise ad. The next most likely is that Meta determines it's awful and it gets no spend. Finally, you get the outlier winners which carry the account. None of these are bad outcomes. And you don't actually know which of the three buckets the ad will fall in until it actually gets spend. "We tried that." "This doesn't work for our audience." "I don't like this." You don't actually know. And even if it were true before, it's always good to retest your prior assumptions because things change. Ecom didn't work in 1995 because nobody buys things on the internet. I think personal taste should play almost no role in the decision to launch or hold ads. In the time it takes to do 3 rounds of revisions so the kerning on the letters is perfect, you could have launched 2 rounds of ads and gotten real world feedback from the auction on what was or was not working.

We took a Google account from $22M to $37M in ad spend this year, all while lowering CAC and improving ROAS. That's 68% growth, an extra $15M in efficiently scaled spend, and it's only accelerating from here. Most agencies call an account this size maxed out. Nine times out of ten it's the agency that ran out of ideas. The challenge with large accounts is simplicity becomes the enemy. The automated systems are not designed to squeeze every last ounce out of the available inventory. They're not creative about finding new inventory either. If you want to scale well past a certain size, you need an incredible depth of platform expertise. You need to understand how everything runs under the hood. That means knowing how the auction, delivery, and inventory systems actually interact across the whole ecosystem, not how Google tells you they work. Each small inefficiency is worth hundreds of thousands of dollars. 99.9% of Google Ads practitioners lack the depth of knowledge to know where these inefficiencies are and fix them. I've been running Google Ads for 16 years. There are probably only a handful of people on the planet who know the platform this in-depth, inside and out. If your Google account has plateaued, DM me. We'll tell you if it's actually saturation or just a knowledge gap nobody's closed yet