The Death of "I'm Not Technical" - How AI is Demolishing Your Biggest Excuse

In honor of 100 episodes of the Sidecar Sync, I wanted to analyze all our episodes with Claude. Simple enough goal, except our transcripts were scattered everywhere—SharePoint, our website, various folders. I could spend hours hunting them down and copy-pasting each one, or...

The easier solution would be to just transcribe all the audio files fresh in one place. Write a program to bulk process them all at once. But I don't know how to write a program. Unless you count basic HTML from my MySpace days, which I don't. Making glitter graphics work on a 2005 social media profile doesn't exactly qualify as software development.

Or does it?

I went to Claude. "How would one write a program and run it?" I asked.

Thirty minutes of setup later, a few hours of waiting, and I had processed all 99 episodes. I had built software. Me! The person whose last "coding" experience involved choosing whether my profile song should autoplay.

This capability isn't entirely new—non-technical folks have been able to build programs since ChatGPT launched in late 2022. But back then it was clunky. Now? It's excellent. And it's only getting better. The shift happened so gradually, then suddenly, that most association professionals haven't noticed the door is now wide open.

The Surreal Reality Check

Imagine waking up tomorrow able to play Chopin on the piano, or having a fluent conversation in Mandarin, or performing surgery. That's essentially what's happened with technical capability for association professionals. People who couldn't imagine building software a few years ago are now doing it—not just thinking about it or planning for it, but actually building working solutions to their problems.

For decades, associations operated within ironclad constraints. They watched for-profit companies build custom tools while they exported another CSV from their database, manipulated it in Excel, and called it a day. They sat in board meetings explaining why that "simple" request from the membership committee would require a $50,000 investment and a six-month timeline. They accepted their role as perpetual customers, never creators.

This was realism based on legitimate constraints. Building software required years of training, specialized knowledge, and ongoing maintenance expertise associations didn't have.

Until suddenly, impossibly, it didn't.

The vertigo hits when you realize you now possess abilities that defined other people's entire careers. Computer science graduates spent four years learning what you can now accomplish by describing your problem clearly to an AI.

The Disadvantage of Not Even Trying

Every time you say "I'm not technical" and stop there, you're operating at a competitive disadvantage. Not because you lack some innate ability, but because you're not even considering solution pathways that are now available to you.

While you're collecting three vendor quotes for that custom integration, another association just built it themselves. While you're explaining to your board why the member portal enhancement isn't in this year's budget, someone else spent their Saturday morning creating exactly what their members asked for.

The constraint disappeared, but the mindset remains. Most association professionals haven't even attempted to build something. They're still operating with 2019 assumptions about what things cost and who can create them. Your biggest barrier isn't ability or budget—it's your willingness to feel temporarily incompetent, to sit in that uncomfortable space where you don't know what you're doing for approximately 20 to 30 minutes before something clicks.

What Building Actually Looks Like

Let me demystify what "building software" means in 2025. You start with a conversation. Not with a developer. With AI.

"I have 99 podcast audio files. I need to transcribe them all in bulk and save them as text files. Can you write code to do this?"

The AI responds with code. You don't understand it. That's fine—you don't need to.

Here's the meta moment that breaks people's brains: You can ask the AI where to put the code.

"I have no idea where to run this code. Can you walk me through it?"

And it does. Step by step. Click here. Create a new notebook. Paste this here. Run this cell. Install this library.

Error messages appear. You copy them back to the AI. "I got this error. What does it mean?"

The AI explains and provides fixed code. You try again. This iterative dance continues. "That's close, but can you make it also rename the files based on episode number?" More code. More attempts. More refinement.

Twenty minutes of confusion. Thirty minutes of mild frustration. Then—it works. You just built software.

Not "kind of" built software. Not "used a template." You solved a specific problem with a custom solution you created. The fact that AI wrote the actual code is irrelevant. You identified the problem, designed the solution, tested it, refined it, and deployed it.

The New Competitive Landscape

Let's be clear about boundaries: You're not building production software. You're not replacing your AMS or creating mission-critical systems that thousands of members depend on. Leave that to professionals.

But quick programs? Prototypes? Tools to solve your immediate problems? That's exactly what you should be building.

The challenge for non-technical folks isn't just learning this is possible—it's understanding WHAT is possible. Here are real things you could build this week:

Bulk file processor: Have 200 PDFs that need renaming according to a specific convention? Build a program that handles it in seconds instead of spending your entire afternoon dragging and clicking.

Email list cleaner: Multiple exports from different systems that need deduplication and formatting? A quick script can merge, clean, and standardize your lists faster than you can open Excel.

Report automator: Pull data from a Google Sheet, format it, create charts, and generate a PDF report—automatically, every Monday, while you're having your coffee.

Event registration analyzer: Extract registration data, identify patterns, segment attendees, and create targeted follow-up lists without the manual sorting you've been doing for years.

Content migrator: Moving blog posts from one platform to another? Build a tool that preserves formatting and metadata instead of copy-pasting until your eyes glaze over.

Survey response processor: Transform raw survey exports into actionable insights with automatic categorization and sentiment analysis—no more reading 500 responses manually.

Member data matcher: Reconcile records between your AMS and email platform to identify gaps and inconsistencies that have been plaguing you for months.

This is a shift in capability, but it's also a shift in economics. "We need to find budget" becomes "Give me an afternoon." "Let's put that in next year's plan" becomes "I'll have a prototype by Monday."

Your Uncomfortable Assignment

This week, you're going to build something. No more theory. Pick your most annoying recurring task:

• The Monday morning report where you pull numbers from three different places
• The event follow-up shuffle with registration exports and segmented emails
• The duplicate detective work comparing member lists between systems
• The PDF naming nightmare with dozens of documents needing consistent conventions
• The monthly spreadsheet merger combining multiple Excel files

Now use the process I just described. Open Claude or ChatGPT and describe your specific problem. For example: "I have 50 PDF certificates that I need to rename to include the member's ID number at the beginning of each filename. Can you write code to do this?"

Follow the steps. Ask where to run the code. Copy your error messages back. Iterate until it works.

The discomfort lasts about 20 minutes. The capability lasts forever. 

The Choice You're Making

"I'm not technical" used to be a fact. Now it's a choice.

A choice that carries consequences. When your board asks why the member portal enhancement will cost $50,000, and you don't even explore building it yourself, you've made a choice. When new staff suggest creating a custom tool and you dismiss it as impossible, you've made a choice.

This doesn't mean abandoning all vendor relationships. You'll still buy QuickBooks. You'll keep your AMS. Enterprise software serves a purpose, and maintaining mission-critical systems isn't where you should start experimenting.

But the connective tissue between systems? The custom workflows? The specific tools your members request? The analysis you need but can't justify hiring a consultant for? These are now within reach.

The technical divide in associations isn't about JavaScript versus Python versus whatever other language you've never heard of. It's not about computer science degrees or IT certifications. It's about courage. The courage to be confused. The courage to ask embarrassingly basic questions. The courage to copy error messages you don't understand and try again.

People are discovering they can solve problems they've worked around for years. They're building member engagement tools over lunch breaks. They're automating workflows during conference calls. They're creating data visualizations that make their boards actually understand the numbers.

Others are still saying, "I'm not technical."

Your members don't care if you're technical. They care if their needs are met. Your board doesn't care if you coded something yourself. They care if problems get solved efficiently. And increasingly, the most efficient solution is the one you build yourself, in an afternoon, starting with "I don't know how to code, but..."

Welcome to the strange new world where your biggest limitation isn't your technical ability. It's your willingness to admit you don't know something—and then do it anyway.