Why doesn’t Windows have a proper wake word like “Hey Siri”?
I use voice tools for almost everything now, but I still need the keyboard to start them.
I’m looking for an open source option. If I can’t find one, I may build it myself.
Sometimes systems should adapt.
If you have someone really talented who is struggling with the way the company works, I don’t think the answer is always asking them to adapt or leave.
Sometimes I would rather change part of the system around them.
Of course, only when it makes sense for the company.
Good systems should help good people perform better, not become the reason you lose them.
Give your team room for trial and error.
Mistakes are normal. You can’t expect people to take ownership, try new things, and improve if they are afraid of getting something wrong.
Sometimes you need to leave space for the team to fail, learn, and try again.
I think that is an important part of building a company that can actually grow.
Sometimes I work for reasons other than money.
I really enjoy building something from scratch, fixing things, and seeing it grow.
That’s probably why I’ve always liked startups.
Sometimes seeing something succeed is enough motivation for me.
Weekends are underrated....
I work less and think more!
I review what went wrong, what wasted my time, what needs to change, and whether I am still moving in the right direction.
Sometimes one quiet day fixes problems I was too busy to notice all week.
I don’t think every workload problem needs another hire.
Before adding someone to the team, I like to understand where the time is actually going. Sometimes the team genuinely needs more capacity. But sometimes people are spending hours on reporting, data entry, follow ups, or repetitive admin work that should have been simplified or automated.
Hiring is expensive. Fixing the workflow first can sometimes remove the need completely.
AI makes it much easier to produce more work
That also makes quality control more important
Producing 10 times more content, reports, or research means nothing if someone still needs hours to review, fix, and rewrite everything
For me, the value of AI is not how much output it can generate
It is how much finished work it can help produce with less time and less effort
If I’m still doing the same manual task every week, I usually see it as a system problem.
Reporting, invoices, follow ups, data updates....
Each one looks small, but together they can waste hours every month.
That’s usually the point where I stop doing the task the same way and start looking at automation or a simple internal tool
Not because everything needs to be automated
But repeated manual work is usually a good signal that the process can be better
Delegation is not giving someone a task.....
It is giving them enough ownership to make decisions around it.
I have seen managers become responsible for results while every small decision still needed approval from above
At that point, you are not delegating. You are just distributing tasks while keeping all the thinking centralized.
I finally finished an internal invoice management tool I’ve been working on for Marketingx1.
It manages invoices and the full archive in one place, exports every invoice with the brand identity, tracks invoice values, and gives me a better view of the numbers.
I also automated several steps in the process, which saves me a lot of time when creating and managing invoices.
It’s completely free and not a commercial product. I built it only for internal use.
I’m really happy with the final result, especially the design and layout. Modern, simple, and fully built around the way we actually work :))
I’m obsessed with anything related to AI and automation that can increase productivity and make work smarter.
But the more I use these tools, the more I notice something.
A lot of the old school business lessons still work.
- Delegate properly.
- Build clear systems.
- Control unnecessary costs.
- Give people ownership.
- Focus on what actually makes money.
AI can make all of this faster.
But it cannot replace it.
A bad process with AI is still a bad process.
A manager who cannot delegate will not suddenly become better because of automation.
The tools keep changing.
The fundamentals really don’t.
One of the worst CEO mistakes is making decisions inside a manager’s area without really consulting them, then blaming the team when the result is bad.
I remember this from one startup. The CEO wanted to hire a new media buyer. I was against it because the issue was not paid ads or the current media buyer. The real issue was that we did not have enough strong content for organic and paid campaigns.
But the decision was made anyway.
A new media buyer was hired, the result did not improve, and the company lost around $3K.
The lesson for me was simple.
If you hired managers to own their functions, involve them in the decisions that affect those functions.
A CEO can still make the final call. But making the decision alone and assuming it is automatically the right one usually creates expensive mistakes.
I’ve started hitting ChatGPT’s maximum conversation length in some of my long chats
And it’s more annoying than it sounds!
I had multiple tabs with ongoing work, full project context, previous decisions, and weeks of history inside them
Now I can’t continue in the same chat.
My current workaround is moving the work into a Project and starting a new chat there
I hope OpenAI adds a proper way to continue the same conversation without losing all that context.
I refuse to read long AI outputs.
And raw output from my team.
I do a lot of research every day, and I don’t want to spend my focus filtering pages of information.
Reading, filtering, and deciding what matters adds cognitive load. Do it all day and it kills your concentration.
As a CEO, don’t only track Media CAC
Media CAC tells you how efficient your ads are.
But Fully Loaded CAC tells you if your business actually makes money.
Big difference.
Media CAC = ad spend / new customers
Fully Loaded CAC = ads + sales team + commissions + creatives + tools + agency + marketing salaries / new customers
If you only track Media CAC, you might think you’re scaling profitably…
while your real acquisition cost is quietly killing your margin
Track both!
Free..!
One word can completely change a buying decision.
Dan Ariely tested this with a simple chocolate experiment.
People were given two options:
Lindt Truffle for 15 cents
Hershey’s Kiss for 1 cent
Most people chose Lindt.
It was more expensive, but it felt like the better value.
Around 73% chose Lindt, while only 27% chose Hershey’s Kiss.
Then he changed one small thing.
Both prices were reduced by just 1 cent.
Lindt became 14 cents.
Hershey’s Kiss became free.
Logically, the difference was almost the same.
But the result changed completely.
Now around 69% chose the free Hershey’s Kiss, and only 31% chose Lindt.
This is why “free” is not just a discount.
Free removes risk.
Free makes the decision feel easier.
Free makes people stop comparing value the same way.
And sometimes, free can make a lower-value option feel more attractive than a better paid option.
That’s why pricing is not only about numbers.
It is also about psychology