Today I ran an entire AI workflow automation without opening a single traditional interface.
Here’s what I did:
- Checked Google Search Console
- Performed SERP research and compared it against GSC data
- Researched contacts for cold outreach on Apollo
- Set up an email sequence on Lemlist
- Fielded a qualitative research study on Ditto
- Created blog posts from the results
- Published them to Contentful
- Updated our website
Ten tools. Maybe more. But I never left Claude Code. The only exception was Obsidian, where I keep my markdown files organized.
I didn’t learn Apollo’s navigation. I didn’t master Lemlist’s interface. I didn’t think about Contentful’s content model. I told Claude what I was trying to do, worked with it to create an execution plan, approved the steps that mattered, and watched it execute.
Work that used to take a full day finished in two hours. I didn’t watch a single training video. I didn’t navigate a single overly complex menu tree (looking at you, Google).
Here’s what’s shifting:
DataForSEO is headless. API-only, rock-bottom cost compared to GUI-wrapped competitors. They charge pennies for the same quality results their competitors wrap in expensive dashboards. Same data. No interface tax.
When Claude Code spins up the software it needs, uses it, then tears it down afterward, what are traditional SaaS providers charging for? Access to functionality, or the friction of navigating carefully designed screens?
For SaaS companies, training users is both a massive pain (speaking from experience here) and a major differentiator. What happens when that becomes irrelevant?
Brands are becoming invisible. UX becomes irrelevant. The only thing that matters is whether I reach my outcome.
The intelligence layer is conversational. The tools beneath it are becoming commoditized infrastructure.
A pretty dashboard isn’t a differentiator anymore. The interface isn’t a moat.
And I’m never going back.
How to Build Your Own AI Workflow Automation
Start by writing down every SaaS tool you touched this week. Then check which ones have APIs. Most do. Many charge extra for the dashboard you’d be skipping anyway.
Look for multi-step sequences in your work. Research, then create, then publish. Pull data, analyze, report. Those chains are where automation actually saves time. Single-tool tasks usually aren’t worth the setup.
Pick tools with decent API documentation first. Google services, Contentful, Notion, Airtable. These have well-documented APIs and you’ll build confidence before hitting the ones with terrible docs (every company has at least one).
Your first automated workflow will be rough. Accept that. It won’t look as clean as clicking through a polished UI. But speed and consistency matter more than pretty for most business work. You can refine later.
Do the math on what you’re actually paying. Your SaaS subscription plus hours navigating menus versus API costs plus time describing what you want. For anything you do repeatedly, API access usually wins. Sometimes by a lot.
I’m not saying delete every app with a GUI. I’m saying stop paying an interface tax on work that doesn’t need one.
Not sure which workflows to tackle first? Book an AI Strategy Session and we’ll figure it out.