I Could Have Done It Myself. That’s the Point.

I could have done it myself. That’s the point.

I was trying to explain to my husband what it feels like to work with AI on a real project.

I told him: I provide the detailed end game, and the AI helps with all the manual cutting, pasting, clicking, reading directions, running commands, testing things, and figuring out why a button that should work suddenly does not.

That is not a perfect technical description, but it is close.

The important part is this: I could have done most of those things myself. AI did not give me permission to do technical work, and it did not suddenly make the internet accessible to me. What it changed was the amount of friction between an idea and a finished result.

Being capable is not the same as having unlimited time

I am comfortable learning new tools. I can read instructions, build a website, organize information, connect systems, and work through a problem.

But when I do an unfamiliar task manually, the task itself is often the smallest part of the job.

First I have to find current instructions. Then I have to learn the platform’s language, figure out which settings apply to my account, move information between several places, test the result, discover what did not work, search again, and try a different route.

None of that is impossible. It is just friction.

And friction adds up.

• Which menu is the setting hiding under?
• Is this article still current?
• What does this error actually mean?
• Which version of the command applies here?
• Did the change save, publish, or only appear to save?
• How do I document this so I can do it again next month?

I can answer those questions one at a time. AI can often read and digest the directions faster, keep track of the steps, and help me apply them without losing the larger goal.

That does not remove my ability from the process. It removes some of the barriers around using it.

My part of the work is still the part that makes it matter

I still have to know what I am trying to build.

I bring the relationship, the context, the taste, the story, and the reason for doing the work in the first place.

I know when something is technically correct but does not feel right. I know when the words sound like a person and when they sound like filler. I know the difference between putting someone’s photographs on a page and actually presenting a life’s work with care.

I also make the decisions:

• What are we trying to accomplish?
• Who is this for?
• What information matters?
• What feels honest?
• What needs to change?
• When is it ready to publish?

On a recent client website, I supplied the story behind the work, the personality the website needed to reflect, the visual direction, the corrections, and the definition of done.

The AI helped turn those decisions into research, organized content, website sections, image preparation, technical configuration, search setup, testing, and documentation.

I did not disappear from the work. I spent more of my time on the parts that actually needed me.

The difference is fewer barriers and fewer handoffs

Without an AI-assisted workflow, I would still have options.

I could do everything myself, moving from tutorial to tutorial and platform to platform. I could hire a writer, a designer, a developer, a bookkeeper, or a virtual assistant. I could divide the work into small pieces and manage every handoff.

All of those approaches can work.

But every handoff requires another explanation. Every new person needs the background, the files, the goal, the corrections, and another round of review.

What felt different this time was that the same working context could continue from the original idea through research, writing, design, technical setup, troubleshooting, publishing, billing, and documentation.

I did not have to start over every time the type of work changed.

This is not about pretending the work is effortless

AI still costs money. It still makes mistakes. Platforms still behave strangely. Sometimes the AI needs better instructions, and sometimes I need to take over a step myself.

There is still judgment, patience, and learning involved.

But the learning feels more available.

I do not have to remember every command or know the exact name of every setting before I begin. I can start with the outcome I understand, ask questions in normal language, read the explanation, make a decision, and keep moving.

That matters to me because so many people are capable of more than their current tools make easy.

A small-business owner may understand her customers perfectly but feel blocked by a website platform. An artist may have decades of work but no idea how to organize it online. A community organization may know exactly what people need but struggle to turn that knowledge into forms, systems, pages, and processes.

The barrier is not always talent.

Sometimes the barrier is time, unfamiliar language, scattered instructions, too many logins, or the fear that one wrong click will break something important.

Reducing that friction changes what people can realistically attempt.

How I think about the business value

The question is not simply, “Could I have done this without AI?”

Yes, I could have.

The better questions are:

• How many hours of searching, learning, copying, configuring, and troubleshooting did I avoid?
• What other work could I do with the time I recovered?
• Did the process create something I can reuse?
• Did I maintain the quality and judgment my client was paying for?
• Did the cost buy real capacity or just more activity?

I am learning to track AI the same way I would track software, equipment, contractors, or advertising. It is a real production expense, and I want to understand what that expense gives back to the business.

The answer will not be the same for every project.

But on the right project, the value is not that AI does something no human could do. The value is that the path becomes shorter, the instructions become easier to use, and one person can carry an idea across more of the distance without getting stuck at every technical border.

The model I am building at Tinker

I think of this as an owner-directed, AI-assisted way of working.

I hold the relationship, the intention, the creative direction, the business judgment, and the final approval. AI helps reduce the procedural friction around research, organization, technical implementation, testing, and documentation.

It does not make human skill irrelevant.

It makes skill easier to put into motion.

That is the part I find exciting—not replacing what people know, but lowering the barriers between what they know and what they can build.

If your work, story, photographs, or years of experience are scattered across social media and old folders, you can learn more about Tinker’s Website and Digital Presence Launch.