96% - Threshold for Perfection

A project that was supposed to be six months has now turned into a two and a half year nightmare.

Documents signed on December 2023.
First working on it on January 2024 with a fragile, disoriented, departed team.
Terms of payments: 500 million IDR first payment + 200 million IDR for every phase in the next 5 consecutive phases.
Today is June 2, 2026. The project has been completed two and a half years late.
Jesus Christ.

Honestly, I'm still confused who to blame and why it failed. I have several ideas

1. Lack of accountability and leadership
   My oversight and overconfidence led me to trust the future of the project to a product manager I had hired before. Out of laziness, I pushed this project to a person who showed a clear lack of leadership and communication. He failed to schedule organized meetings, coordinate with the client, or create a task structure that the team could follow.
2. Failure to define the scope of work
   We were unable to define the exact tasks for the client, including the number of iterations allowed or how much feature development could occur before calling a task "done." My inability to say no or establish clear boundaries cost us significant time, exhaustion, and fatigue.
3. Lack of communication
   This wasn't just poor communication; it was a total lack of it. The client never communicated their priorities or their unhappiness to us. Meanwhile, internal team members were often unaware that the project even existed. As people left, there were no handovers regarding project status or the new PIC (person in charge). This breakdown in basic communication led to a total failure, as the client eventually stopped acknowledging us as a partner they needed to respond to.

This project did not start with good intentions. It began because I wanted a way out of my company; we were in significant debt, and I needed to offload that debt by generating revenue somehow.

By selling our IP—which in this case was the software code—to another client as an intangible asset sale, we made close to $100,000. At the time, the total project was worth 1.5 billion IDR. Honestly, I don't even know why this client said yes to us. I don't think it was out of desire; I think it was just our common investors nudging them.

Thank God for them, though, because they actually paid off one of my biggest debts at the time, which was to SAP (the logistics company). We also had to pay a bunch of employees while shutting down, as there was no proper transition plan.

I think I knew after one year that this project was not going to work out. It raises the question: how far are we willing to take this, and how do we define whether a project is a success? I always felt like the handover and integration were failures because, in the end, the client dumped all the technology we created for them. Honestly, I salute my team, who endured a year of unpaid work just to make sure the project was finished. I think they deserve to be mentioned: Abiel, Adi, Hafidz, Aziz (who I actually never met or spoke to. He never joined any calls, but there was some work output, so yeah)

Now I understand that technology is not truly an intangible asset because code has an expiration date. After a couple of months, code is no longer usable; libraries become outdated, frameworks are updated, and if your code is deprecated, it must be updated. For example:

1. Our Node.js version was beyond the current stable version.

2. Many of our libraries were deprecated because our goal was to maintain the project with as few changes as possible.

If you think about it, how can a codebase from 2023 be used in 2026? There is no way if it is not maintained properly. Since we did not maintain it at all, the codebase became absolutely outdated to the point where even the client didn't want it. They told us to update it, but we said no because it was out of scope.

One thing we learned is that software is not an intangible asset. I would not categorize technology itself as an asset. In fact:

(a) A user base can be an asset.

(b) Monthly recurring revenue can be an asset.

(c) B2B contracts can be an asset.

The deal could only work in 2023. By 2026, you can write code for the whole 100k software in under three weeks. That makes you question how valuable the software actually was if someone can copy and clone it in three weeks.

I am lucky that the deal happened, but one strong learning outcome is that we shouldn't aim for 100% perfection or the 100% best outcome. I feel like 96% is good enough, and in this case, 96% is what we achieved. We secured 1.44 billion IDR out of the total 1.5 billion IDR payment amount. The client ditched the software at the end, but I don't care; we got paid.