productivity on

The 7 AI Tools I Use Every Single Morning to Run Two Companies

Thu, 14 May 2026 00:00:00 +0000

These are the seven tools I open every single morning to run two companies with a team of three. No sponsored picks — just what actually works.

I break down exactly how I use each one, what it costs, and why it earned a spot in my daily stack. Total cost: about $125/month. That’s less than one freelancer for one day.

When you’re running multiple businesses with a tiny team, your stack has to cover four things: thinking, creating, automating, and communicating. Every tool below earns its keep by doing at least one of those better than I can on my own.

1. Claude — Thinking & Strategy

Cost: $20–$200/month (Pro or Max)

If ChatGPT is the Swiss Army knife, Claude is the scalpel.

Claude is where I do my thinking work. Strategy sessions. Business planning. Analyzing documents. Working through complex decisions. Anthropic recently dropped Claude Opus 4.7 and honestly, it’s genuinely scary good at understanding long context — we’re talking a million-token context window.

The reason Claude wins over ChatGPT for this specific use case is its long-context accuracy. When I’m working with big documents — contracts, financial models, scopes of work — Claude holds onto the details better. I can drop in an entire project folder and have a real strategic conversation about it.

This is my “sit down and think” tool. If I’m making a business decision, writing a strategy doc, or trying to understand something complex, Claude is where I start.

2. ChatGPT — Execution & Speed

Cost: $20/month (Plus)

Yes, I pay for both. Because they actually do different things.

ChatGPT is my execution tool. If Claude is where I think, ChatGPT is where I get things done. Quick tasks, content drafts, email responses, data formatting, image generation — anything where I need speed and versatility.

The ecosystem around ChatGPT is also more mature. Custom GPTs, plugins, integrations — it connects to more of my existing workflow. When I need to crank through a bunch of tasks fast, this is where I live.

Think of it this way: Claude is the advisor in the corner office. ChatGPT is the senior associate who cranks out deliverables.

3. AI Agents — Hermes + pi.dev

Cost: Free to low cost (open source)

This is where things get really interesting. Your business starts running itself, getting smarter every day.

I use two different agent frameworks. Hermes is a fully open-source agent that handles automation workflows. Pi.dev is my daily working agent that I’ve customized with skills, plugins, and extensions tailored to my specific businesses.

The key difference between agents and just using ChatGPT is persistence and autonomy. An agent doesn’t just answer your question and forget about it. It remembers context, learns your preferences, and can execute multi-step workflows without you babysitting every step.

I’ve built custom skills that handle everything from support ticket triage to content scheduling. The agents are connected to my other tools, so they can pull data from meetings, check project status, and proactively surface things I need to know about.

If you’re not using AI agents yet, this is where the biggest unlock is waiting for you. It’s the difference between having AI as a tool and having AI as a teammate.

4. Granola — AI Meeting Notes

Cost: ~$15/month

I’m on calls constantly. Client calls, team syncs, partner meetings, sales conversations. Before Granola, I was either taking notes and missing half the conversation, or paying attention and forgetting the details later.

Now every single call gets auto-transcribed, summarized, and broken directly into action items. Perfect recall on every conversation. No more “what did we agree on in that call last Tuesday?” — it’s all there.

But here’s what makes Granola different from other meeting tools: it has a fantastic MCP and API. So all of those agents I just talked about? They’re connected directly to Granola. My AI agents can pull meeting context, reference past conversations, and follow up on action items — automatically.

That integration is what turns a simple transcription tool into an operational backbone.

5. Codex — Full Coding Environment

Cost: Included with ChatGPT Plus

I was all in on Claude Code. Then I tried Codex. Codex wins now, hands down.

Codex is OpenAI’s dedicated coding agent, running on the full power of GPT-5.5. The difference is in how it works: you give it a task, and it goes and does the entire thing. It doesn’t assume you want to sit there pair-programming side by side. It just executes.

For me, building features on LGL Tools, probably 80% of the time it can one-shot what I’m working on. And the fact that I never have to leave the environment makes debugging 20–30% faster.

As a founder who codes but isn’t a full-time developer, this is massive. I can ship features that used to take a full day in a fraction of the time. The quality of the output is high enough that I’m reviewing and tweaking, not rewriting.

6. Notion AI — Operational Brain

Cost: Included with Notion plan

Notion is my operational brain — where I store things and how I get things done. Project trackers, client databases, SOPs, content calendars — it all lives in Notion.

With Notion AI layered on top, I can ask questions across my entire workspace. “What’s the status of the SafetyCulture project?” “When is the next MGA event?” “What did we decide about the pricing change?” Instead of digging through pages, I just ask.

I also use Obsidian for offline and local notes — things I want completely private and stored on my machine. The combination of Notion (collaborative, cloud, integrated) plus Obsidian (private, local, markdown) covers every knowledge management need I have.

7. AI Design — Claude Design, GPT Images 2.0, Impeccable Design

Cost: Included with existing subscriptions

The design landscape in AI has exploded. I’m using three tools depending on the job:

Claude Design for UI/UX mockups and structured design work
GPT Images 2.0 for quick graphics, social content, and visual concepts
Impeccable Design for polished, production-ready design work

As someone who runs an event production company, visuals matter. A lot. Being able to generate client-ready graphics, social content, and presentation materials without waiting for a designer has changed how fast we can move.

This isn’t about replacing designers for high-stakes creative work. It’s about eliminating the 24–48 hour delay for the everyday visual assets that keep your business looking professional.

The Full Stack Recap

Tool	Use Case	Monthly Cost
Claude	Thinking & strategy	$20–$200
ChatGPT	Execution & speed	$20
AI Agents (Hermes + pi.dev)	Automation that learns	Free
Granola	Meeting notes & recall	~$15
Codex	Coding environment	Included w/ ChatGPT
Notion AI	Operational brain	Included w/ Notion
AI Design	Visual content	Included w/ Claude/ChatGPT

Total: ~$125/month for what would otherwise require multiple hires.

These tools run ²⁴⁄₇. They don’t take sick days. They don’t need onboarding. And they get better the more you use them.

If you’re a founder running lean, you don’t need to adopt all seven at once. Start with one. Pick the tool that addresses your biggest daily pain point, and build from there. The compound effect kicks in fast.

I’m Cole Morgan. I run Let’s Go Live and LGL Tools. I post AI productivity content for business owners on YouTube every Tuesday and Thursday.

How I Run Two Companies by Myself (AI Changed Everything)

Tue, 12 May 2026 00:00:00 +0000

Last week, I ran a livestream event for a few thousand people. I shipped a feature on my SaaS product. I managed client deliverables. And I still made it home for dinner with my wife and two-year-old.

A year ago, that literally never would have been possible. It would have buried me.

The difference isn’t that I’m working harder. The difference is that I stopped doing all of the work myself.

Who I Am and What I Run

I’m Cole Morgan, and I’m the founder of Let’s Go Live — a small event production company. We do live streaming and event production for clients like Tony Robbins, Monday.com, and Twilio.

I also run a SaaS software company called LGL Tools. I build stage timers, prompters, and production tools — everything that AV professionals need to make their show better, all running in a browser.

I have a small team. Three people. The reality is, as a founder, everything flows through me — the product, client management, creative decisions, support fires during live events. There’s no “delegate that to a department” when the entire company is just three people.

Why I Rebuilt My Workflow Around AI

About a year ago, I made the decision that instead of hiring more people, I was going to see how far I could push AI into my daily, active workflow.

Not as a toy. Not to write tweets. But as a legitimate operational layer in my business.

Here’s the reality most people miss: the work that eats you alive as a small business owner isn’t the creative, strategic stuff. It’s the operational middle layer — triaging support tickets, writing proposals, managing timelines, shipping code, handling client communication. That’s where the hours disappear. And that’s exactly where AI fits.

The AI Adoption Gap

Goldman Sachs published some data in April 2026 showing that workers who use AI save roughly 40 to 60 minutes per day. That’s real enterprise data from OpenAI’s customers.

But here’s the number that matters more: only 19% of U.S. companies have adopted AI in any meaningful way.

That means over 80% of businesses are still doing everything manually. If you’re a founder, freelancer, or small business owner trying to figure out where AI actually fits — you’re not late. You’re early. And that gap is your competitive advantage right now.

What a Real AI-Powered Workday Looks Like

Let me walk you through what this actually looks like. Not theory, not demos. Real operations.

Morning triage. I wake up, and my support inbox already has new tickets. Instead of reading through every one manually, I take all that context and paste it directly into AI — Claude, ChatGPT — and say: Prioritize these. Tell me what’s urgent. Tell me what can wait. Can someone else handle these? Can you handle these?

Within 30 seconds, I have a triage list. That used to take me 20–30 minutes of just reading and sorting.

Daily priorities. Once triage is done, AI tells me my top three priorities for the day. Not a to-do list of 47 things. Three things that actually move the needle.

Proposals and client work. When I need to write a scope of work or proposal, I give AI the context — the client’s needs, our capabilities, the budget — and get a first draft in minutes. I edit and refine, but the blank page problem is gone.

Code and product. When I’m building features for LGL Tools, AI handles the grunt work — writing boilerplate, debugging, suggesting implementations. I stay focused on the architecture and decisions.

Event support. During live events, when 40 support tickets come in within an hour (and they do), AI helps me triage, draft responses, and identify patterns so I’m not drowning.

The Mindset Shift Founders Miss

The biggest thing I’ve learned isn’t about the tools. It’s about the mindset.

Most people try AI once, get a mediocre result, and say “it’s not ready.” But that’s like hiring an employee on day one and expecting them to know your business. You have to teach it. You have to give it context. You have to build workflows around it.

The founders who win with AI aren’t the ones with the best tools. They’re the ones who stop thinking of AI as a search engine and start thinking of it as a team member.

The Gap Between AI Hype and Execution

There’s a massive gap right now between the AI hype you see on Twitter and the actual execution happening inside businesses. Everyone’s talking about AI. Almost nobody is showing what it looks like to actually run a company on it.

That’s what I’m going to do here. Real workflows. Real tools. Real results — not from a massive company with unlimited budget, but from a three-person team doing production, SaaS, and client services simultaneously.

If you’re a business owner trying to figure out where to start, here’s my advice: pick one operational pain point — the thing that eats the most time in your day — and see if AI can take 80% of that off your plate. Not 100%. Just 80%. That’s where the magic is.

I’m Cole Morgan. I post AI productivity content for business owners on YouTube every Tuesday and Thursday. If you want to see the seven specific tools I use every day, check out my next video: AI Tools Crushing It in My Business Right Now.