Notes from a CTO #12: Prioritization, Sewing Machine, and Best Way to Learn
All ideas look good, but successful people are those who can prioritize the best idea.
1. Thought of the Month
This month, let's talk about prioritization.
I have been thinking about what differentiates a company that can achieve $1 million in 4 years and $1 million in 2 years & keep growing 3x-4x each year. One major reason is the founder's ability to prioritize and focus on important tasks. This is a topic about which I feel somewhat hesitant to write because I, myself, feel like I haven't learned to prioritize effectively. Otherwise, Docsumo would be on a clear path to $10 million in ARR in the next two years. We have a path, but it's not very clear. We will need to get better at prioritization. This is the most critical aspect.
It's very easy to come up with good ideas, but finding the best idea, focusing on it, and making changes based on more data you gather along the way is a superpower that will make things more predictable.
I apologize if you expected me to provide a framework on how to do it; I am still learning this. I will write more useful blogs as I discover more. For today, this is just a dump of my ideas on what I have learned so far and what I will do.
One step to prioritizing better is to have a clear focus on what you want to do and what problem you want to solve. The more focused you can be, the better it will help you make decisions. One of the most difficult tasks is saying no to clients when they want to pay money. You will make mistakes time and again. The way I see it, we get greedy. I am trying my best not to, but when you see MRR (Monthly Recurring Revenue), your animal instinct kicks in. To counter this, we have a client framework to determine whom we should say yes to. It helps, but sometimes we deviate from this framework as we can bend anyone to fit the framework, saying this and that. We are trying to do less of this and are still learning.
One other thing you can do is establish clear metrics that you want to track and have a clear idea of how this initiative will increase those metrics. Avoid misleading yourself with fake metrics such as signups, but instead focus on retention or focus on SQL not leads. It's important to seek genuine improvements rather than relying on false metrics.
The last thing is, you have limited resources. I wish I could accommodate all the good ideas that come to mind, but we only have 12 hours per day for work. Figuring out what you can afford to lose by taking risks and creating a simple RICE scoring with all your ideas can help. I used to rely on my intuition for a lot of things, but lately, I have come to respect simple frameworks like RICE. Don't try to read into more complex frameworks/scoring methods; just keep it simple and add things as you need. Don't go after every shiny new thing all the time. Rice for prioritization and BANT for sales.
In conclusion, prioritization is the cornerstone of success, especially in the startup world. Our experience has taught us the importance of focusing on the best ideas, using data for refinement, and declining client money if it doesn't align with our opportunities. Maintaining clarity in our objectives and avoiding misleading metrics is crucial, as is effectively managing resources within the limitations you are surrounded by.
2. Podcasts/Essays
No one thinks about sewing these days. My mom had a sewing machine way back, and I used to be mesmerized by it. The way a sewing machine works always fascinated me. Below is a video about history and how it work. You will notice how revolutionary the product was at the time. At one point, sewing machines were powered by steam engines. Today, very few people give it a second thought. Looking back at history brings back all the memories from childhood.
3. Interesting links
Repos:
Openstatus: The status page for any website is the first thing you check when things go haywire. Here is an open-source alternative to setup for your website.
JsonGenius: I am a fan of using LLM for website data extraction, but getting the data in a structured JSON format that is useful is the next battle. I hope this helps a little bit.
localsend: Sharing files between devices can be a nightmare sometimes, and I hate using online solutions because I don't trust when files have to go through cloud services.
localpilot: Simple but very interesting idea. I want to try a 1B parameter LLM codes models on my personal server.
For more, follow me on Github: bkrmdahal
Articles:
Misunderstanding About Education: What is the Best Way to Learn? I often contemplate this question. Is it through the use of new AI technology for personalized learning, or is the old-school approach still the most effective? In reality, it depends on the individual; what works best can vary from person to person. Finding the right approach that suits you is crucial, but not everyone can make that decision successfully, and the consequences can be disastrous.
So, should you always grant individuals the freedom to make their own choice in this matter? This seemingly simple question can be a contentious one, as there is no universally correct answer, and people often hold strong opinions on the matter.
Retries: An interactive study of common retry methods What makes a good programmer can be summarized into a few qualities: no code smell, proficient debugger, effective monitoring and error handling, ability to write good retry mechanisms, and comprehensive comments. It is this simple.
God Help Us, Let's Try To Understand AI Monosemanticity: How does AI work? This question is the next billion-dollar question, along with AGI. Here is a good article on the same. HN discussion on same.
A100 Cost: I always wonder how much running an A100 costs, and how it compares to cloud costs. Here's a breakdown.
4. Quotes/ Books
You have power over your mind - not outside events. Realize this, and you will find strength.
- Meditation by Marcus Aurelius
I have been bed-bound for the past four weeks due to a minor surgery. The wound is taking time to heal, causing me to feel restless and stressed. Trying to calm my mind is a challenge, especially with the daily fires I have to put out in startup. It's easy to say "be calm," but difficult to practice. Reading "Meditation" by Marcus is always a good idea. It helps me realize that the source of each problem is within myself, and so is the solution. It brings my thoughts back to the fact that I am in control of my day and my mind. As long as I remain calm, things will fall into place. I shouldn't overthink, but instead let things settle.
In memory of Charles Munger, who always inspired me. Rest in peace.
Our Slack channels have a great collection of memes.
That’s it for this edition. I hope you find it useful.
Best,
Bikram Dahal
P.S If you learned something new today, please share “Notes from a CTO” with your friends and spread the love. ✌🏻