It was never about the tools.

I got the email this morning. You probably got it too. OpenAI announcing Codex, their new coding agent, plus a bunch of updates to how ChatGPT handles code. Another tool. Another announcement. Another thing to evaluate, test, maybe adopt, probably forget about in two weeks. And my honest reaction, sitting in my inbox at 7 AM with a coffee getting cold, was: I am so tired of new tools.

The tool economy is a distraction

Here is what I think is going on. The AI world learned a trick. News brings eyes. Eyes bring signups. Signups bring cash. So each week there is a new thing. A new agent. A new model. A new IDE add-on. A new "copilot for X." And each week, millions of people open those emails. They click through. They sign up for the wait list. They add one more half-set-up tool to the pile.

No one needs tool number 50,001.

I'm not saying Codex is bad. I haven't used it yet. Maybe it's great. But the pattern is the problem, not the product. The pattern is: announce, hype, adopt, half-configure, forget, repeat. And the result is that most businesses. Most people . Are sitting on top of 10 or 15 tools that could genuinely transform how they work, and none of them are wired together. None of them talk to each other. None of them actually do the thing they were supposed to do when you signed up, because you never finished the implementation.

The gap isn't "lack of tools"

I talk to small business owners constantly. Chamber events. BA5 meetings. Coffee one-on-ones. And the conversation is always the same. "I've got this CRM but I'm not really using it." "I signed up for that scheduling tool but it doesn't connect to my calendar the way I need." "I tried that AI writing thing but it doesn't sound like me." "I have Zapier but I only set up one zap and it broke."

The gap is shipping it. Every single time.

They don't need one more tool. They need someone to sit down and wire what they have into something that runs on its own. They need the CRM to update when they meet a new person. They need the calendar to block time when a client books. They need email to get sorted without them touching it. They need invoices to go out when the job is done.

All of this works today. It has worked for over a year. The tech is here. The APIs are here. The hooks are here. What's missing is the build. The dull, quiet, no-applause work of wiring point A to point B and keeping the wire from snapping.

Why this keeps happening

New tools are fun. Building them out is dull. That sums it up.

A new AI coding tool feels great to sign up for. You get a buzz. Wiring your old project tool to turn emails into tasks? Not fun. But one of these moves the business. It's not the one on Product Hunt.

There's a deeper thing here. New tools feel like progress without making you change. You can sign up for 10 new things and feel busy without changing one workflow. Doing the work means looking at how you really work, owning what's broken, and fixing it. That stings. So folks keep stacking tools instead of using the ones they own.

The friction problem

I've built systems for my own work. And now for others too. Here is what I learned. The real problem is not what they can do. The real problem is friction. Just friction.

You have an idea. Between that idea and it being real, there are 47 tiny steps. Each step requires a context switch. Each context switch costs momentum. By step 20, you've forgotten why you started. By step 30, you've moved on to something else. The idea dies not because it was bad, but because the friction between "I want this" and "this exists" was too high.

New tools don't cut that friction. They add to it. Now you have one more login. One more dashboard. One more set of docs. One more setup step. One more thing that can break.

What cuts friction is doing the work. Take the tools you trust. Take the flows you know. Pull out the by-hand steps in between. Make it so when you say a thing should happen, it happens. Not "it happens after you open three tabs and copy and paste a thing." It just happens.

What I'd actually tell you to do

If you're reading this and you feel called out. Good. I'm calling myself out too. I've been guilty of the same thing. Early in this journey I tried every new AI tool that launched. Evaluated dozens. Signed up for hundreds. Most of them are gathering dust in browser bookmarks I'll never open again.

Here is what worked. I stopped looking out. I started looking in. I asked myself a boring question. What are the ten things I do every day that I wish I did not have to do by hand? Then I started fixing them. One at a time. With tools I already had.

Email triage. Contacts. Meeting prep. Invoice chasing. Content scheduling. Status updates. Each one was a 5 to 15 minute daily task I did by hand because "it's only five minutes." But five minutes times ten tasks times 250 work days is over 200 hours a year. That's five full work weeks. Spent on tasks a good system can run in the background while you think about other things.

Not one of these used a new tool. Each one ran on stuff I already pay for.

The real competitive advantage

Here is something that sounds basic but most folks miss it. The edge in AI right now is not better tools. We all have the same tools. The edge is how fast you ship. How fast you go from "this should be automatic" to "this is automatic." How fast you cut friction from your real daily work. Not how many AI apps you pay for.

The businesses that are pulling ahead right now. And I mean right now, not "in the future when AI gets better" . Are the ones that took the tools available 12 months ago and actually implemented them. Fully. End to end. Connected the systems. Removed the manual steps. Let the boring plumbing do its boring work.

We are still reading the emails.

So here's my ask

Next time you get that email. The new tool. The new agent. The new copilot for it all. Before you click, ask one thing. Have I set up the last three tools I got?

If the answer is no, close the email. Open the tool you already have. Spend 30 minutes actually wiring it into your workflow. Connect it to the thing it was supposed to connect to when you signed up six months ago. Make it do the job you hired it for.

That 30 minutes will do more for your shop than any new tool news.

Stop adding tools. Start cutting friction.

← Back to all posts