When I joined my current company, I walked into a system that worked just enough to hide its flaws. Several processes were being handled manually, and others were simply unsafe. Passwords were shared over email. Some pages in WordPress that should have been posts were being published using hardcoded solutions.
The first decision I made was simple: I set up a password manager. We moved away from sharing credentials through email and started storing them in a centralized, secure place with added features like two-factor authentication. I chose Proton Pass because I already used it and trusted it. I led the implementation, and now everyone who joins the company gets access to it as part of onboarding.
This repeated itself across the company
We were using the free version of Power Automate, which didn’t meet our business needs. At first, it worked, but “working” is not the same as scaling. Eventually, we transitioned to Zapier to support more robust automation.
The same thing happened with uptime monitoring. We started with cron jobs, which required manual setup and maintenance, and later moved to UptimeRobot for a more reliable and hands-off solution. It gave us better visibility, notifications, and less operational friction.
Infrastructure was another decision point. The website was running on an EC2 instance that cost about $65 per month. It wasn’t wrong, but it wasn’t aligned with our actual needs. The traffic didn’t justify the setup. I made the decision to migrate to Hostinger, saving the company roughly $700 a year. That was a straightforward case of resource reallocation.
Payment processing followed a similar pattern. We moved from WooCommerce to SureCart and then back to WooCommerce. From the outside, that might look inconsistent. In reality, it was a trade-off. We prioritized speed over a perfectly clean implementation. The only reason we moved back was due to FSA/HSA payment requirements, where our current processor integrates directly with WooCommerce. That made the decision practical, not ideological.
What ties all of this together is ownership
Most of these changes happened because one person was responsible for making decisions in the technology side of the business.
My CTO had strong ideas and instincts, but uncertainty slowed execution. Over time, through experience, we both learned what ownership really means: someone has to decide, test, adjust, and move forward.
After about a year of operating with disorganized workflows, we shifted toward structure.
We began standardizing processes by creating documentation for each department. We introduced clear requirements for presenting projects and business decisions. Instead of long, unfocused meetings, we now rely on written context.
Conversations often start with: “What does the documentation say?” or “Let’s review what we’ve already defined and stick to it.”
This reduced communication overhead significantly. Decisions became clearer, faster, and more predictable.
At the same time, we started building systems that are increasingly self-service. As we acquire more clients, this becomes essential. The goal is simple: reduce dependency on constant human intervention. We are still working on this, but the direction is clear.
Looking back, none of this came from a single moment of insight. It came from exposure, experimentation, and pattern recognition.
Sometimes the solution was direct: “What if we use a password manager? It works like this.”
Other times it came from recalling something I had seen elsewhere: “I’ve seen teams use tools like Asana for this. We can probably leverage something similar.”
Those ideas only mattered because we implemented them and tested them in our own context.
Not every tool worked. Some had to be replaced. Others were unnecessarily complex, and we reverted to simpler solutions like Excel or Google Sheets, depending on the task.
Over time, the gaps in our processes became more visible. And once you can see them clearly, strategy becomes unavoidable.
It was never really about the tools or the software. It was about identifying business needs and building the processes required to support them as the company grows.