#16: Today’s Breakthrough, Tomorrow’s Baseline

Hey friends, and a special welcome to all new subscribers!

There’s currently a big gap between what most people assume AI tools and workflows are capable of, and what’s actually possible.

People fall into two camps;

• those that patiently wait for the tech to be diffused and “easy to use”

• those that roll up their sleeves, get their hands dirty and starts experimenting

Either is fine by me, but the focus of The Inside Track is to power the second group. Why? Because there’s a huge competitive advantage to be at the forefront here (both at a company level and individuals that wants to build their career).

PS: I’m still shaping the concept for a bespoke consulting firm designed to help small and midsized businesses cross this chasm.

If you run a biz and have a task/workflow/idea you’ve thought “there must be a better way” – hit reply and let me know. I’m offering free consulting for a select few as I test the framework.

PS2: feel free to connect with me on LinkedIn and say hi! It’s always fun to connect with new readers.

– Martin

The Idea

Remember when GPT-4 launched and everyone lost their minds? That was March 2023. By December, the same teams were grumbling that responses took "too long."

Sam Altman calls this the "gentle singularity"—exponential AI gains get absorbed so fast they feel ordinary. What seemed impossible Tuesday becomes baseline by Thursday.

Your Advantage Has an Expiration Date

The pattern repeats everywhere:

• AI writes code → GitHub Copilot normalizes it → junior developers who can't prompt get filtered out

• AI analyzes spreadsheets → Excel adds Copilot → manual pivot tables feel primitive

• AI generates content → Claude handles research → human-only workflows look like fax machines

Revolutionary becomes routine in quarters, not years.

Most Founders Are Moving Too Slow

Here's the problem: most operators are still treating this like a gradual upgrade cycle. "We'll add some AI features to next quarter's roadmap."

Meanwhile, three startups just launched with AI-first approaches that make your current workflows look ancient.

The gentle singularity creates a brutal competitive reality. That manual process you're proud of? Someone's automating it right now. That expertise your team spent years building? AI democratized it last week.

Your current moat has an expiration date measured in months, not years.

The Only Real Edge Left

In a world where breakthrough capabilities become commodity tools in quarters, there's only one sustainable advantage: continuously riding the bleeding edge.

The space moves so fast that being 3 months behind feels like 3 years. The companies winning aren't the ones with the most sophisticated AI—they're the ones constantly experimenting with whatever dropped last week.

How This Plays Out

• For individuals: Your skill isn't prompt engineering. It's pattern recognition—spotting which new AI capability kills your current bottleneck, then rebuilding your workflow around it before everyone else catches on.

• For companies: Competitive advantage now lives in iteration speed, not feature depth. The team that can test, implement, and optimize new AI tools in days beats the team that spends months perfecting last quarter's breakthrough.

The real edge: You're not building on stable ground anymore. You're surfing an avalanche. The moment you stop moving forward, you get buried.

The Tactic: How to Create Self-Running Research Pipelines

One of the most valuable workflows I've built is automated information gathering from sources like niche websites, industry publications, Reddit, and LinkedIn.

The concept is dead simple: set up regular searches across your source list, use a carefully designed prompt to filter for relevant content, then get it summarized. That's step one. Step two is whatever you want to do with that intel—email it, turn it into content, feed it into other workflows.

What makes this worth building is that research becomes the foundation that powers everything else you're working on.

This Relay workflow is your starting point—think of it as the research engine that feeds bigger automations. You configure it once with your sources and search terms, then it delivers curated summaries on whatever schedule makes sense for you.

Once you've got that base running, you can build whatever comes next:

• Email Digest Route: Just send the summaries to your inbox for daily updates
  

• Content Creation Route: Filter the research through your brand voice to spark new content ideas
  

• Competitive Intelligence Route: Auto-flag when competitors make moves or change pricing
  

• Market Research Route: Pull pain points from niche subreddits and forums where your audience hangs out

The template does all the heavy lifting—crawling sources, removing duplicates, summarizing everything—so you can focus on what to do with the insights.

Here's how to get this running:

1. Import the template: Click here to copy the Relay workflow into your workspace
   

2. Add your sources: Add the URLs to sites you want to watch
   

3. Customize the search prompt: Replace the prompt with your specific focus.
   

4. Set your schedule: Daily works for fast-moving stuff, weekly for deeper research. Don't go crazy with frequency—you want signal, not noise
   

5. Test it first: Run it manually to see what quality you're getting, then tweak your sources or prompts until the summaries are actually useful
   

6. Build what comes next: This is where it gets fun. Add email delivery, content generation, or whatever workflow you need the research to feed into

Start simple with just the research collection, then layer on the workflows that turn intel into leverage.

What do you want to add as the next step to this flow? Hit reply to let me know, and I’ll build a custom flow for the first few replies that make sense.