Let's say every time you walk through a certain door, you hit your head. Not hard enough to knock you out, but enough that you have to stop, go to the sink, grab a towel, clean things up, and put on a bandage.
So naturally, you start thinking about how to fix it. You could put up a sign. Keep towels nearby. Automate the whole thing — a sensor detects the impact and triggers an alert.
All of that would make the situation easier to deal with. None of it would solve the problem. Because the problem is not the bandaging process. The problem is the door.
I have been thinking about this as I watch what is happening in go-to-market right now.
AI and automation have made it incredibly easy to generate messaging, build workflows, trigger outreach, and scale activity. So we optimize the system. We build better sequences, personalize at scale, automate follow-ups. There is a real sense of progress in this.
But from the outside, something different is happening. Buyers are getting flooded with outreach that looks personalized, sounds thoughtful, and increasingly feels identical. It is not hard to see why — we are all using the same tools, training on the same patterns, optimizing toward the same metrics.
None of this is irrational.
If you are operating under real constraints — small teams, limited budget, aggressive targets — when tools promise to automate the pipeline, it is entirely reasonable to lean in. And early on, it often works. More messages lead to more responses. More coverage leads to more meetings.
That is what makes this so compelling, and also what makes it dangerous. The same forces that make automation effective early are the ones that begin to break it later. From the inside, it still feels like optimization. From the outside, it starts to feel like noise.
This should feel familiar.
SOAR promised to automate the SOC. In theory, it worked well if you had clean, well-defined playbooks. In practice, most teams did not. What ended up getting automated was whatever processes already existed. When those processes were weak, automation scaled the weakness. AI did not fix that. It made it easier to do more of it.
GTM is starting to follow a similar path. We are getting very good at automating execution without asking whether the underlying approach is sound.
What actually matters.
Automation does not fix unclear thinking. If your positioning is weak or your understanding of the customer is shallow, automation will scale that problem rather than solve it.
Activity becomes a less reliable proxy for effectiveness. It is easy to measure emails sent or sequences launched, but much harder to determine whether you actually created something meaningful for the person on the receiving end.
Automation is a force multiplier. It does not improve the underlying system. It amplifies whatever is already there.
So before building the workflow, writing the prompt, or launching the next sequence, it is worth asking a simple question. If this works exactly as designed, are we actually solving the problem? Or are we just getting better at handing out bandages?