Author: abhiyerra

It is time to depart from this multipart opsZero. We have one opsZero that can then create other brands. No need to build another brand.

This site is moving to being just a site for abhiyerra.com and the true YerraCo is OpsZero.

Quality is the next step for the company. The outcome of this is that every microservice we have will have to fulfill conditions of cost and quality. These are financial metrics as well as quality metrics. The approach we seek to take is to move more and more of our work to a decentralized microservices and then apply the SPC to each microservice.

Each microservice then should do the following:

• Generate KPIs for timestamps.
• Turning these KPIs into control charts
• Realtime view of sales and expenses
• Turning these into control charts

Each microservice should be relatively self contained..

The libraries we are building:

• Deming. KPI tracking and storage that is used within the app.
• Malone. Payment and billing system.

I have a fascination with functional programming languages like Lisp, OCaml, F#, Haskell as alternates to the tried and true like Python, Ruby, or Shell. But as a business owner I have to also make bets on technology that fit with the paradigms of the organizational structure that I have. I have largely bet on Python for everything. We don’t have really any code that uses any other language.

I’ve been exploring reasons to use F# and in many ways there are isolated scenarios where using it would actually make sense to use especially with the robust type system. But the types of programs I am building don’t require a robust type system. They need the ability to build fast and with access to a large ecosystem of libraries such as pandas and numpy that Python provides. Betting on an esoteric language just doesn’t feel like a good idea.

Insofar as that goes, I am committing all our efforts to using Python. It just doesn’t make sense to use anything else even a faster language like Go unless we absolutely need the performance that brings. We will be a Python shop and put all our effort in support the Python ecosystem and making that better.

As the company transforms into a HoldCo there are a few requirements that I realize I need to meet:

• Figure out the financial statements of the company including organizing the income statement and balance sheet with a focus on managerial accounting versus GAAP. I need to really understand the flow of work and the costs associated with them. This means both personal and business costs are modeled under a single ledger.
• Clean up the balance sheet. Currently the balance sheet for the HoldCo is a total mess. I want to clean it up to remove as many unnecessary pieces. This means divestment, reorganization of debt and general simplification. This will also mean reordering spending to appropriate accounts. The goal is to be aggressively asset light.

As a business you need to figure out what are your strengths and weaknesses and my strengths are not in asset heavy industries. I am just not knowledgeable about it to be useful. The last attempt I had of Startup Sonoma was in many ways a lesson I should have taken.

Assets are fickle. They need constant upkeep and the knowledge to upkeep them. If you don’t have that knowledge then it is pretty difficult to maintain those assets. Startup Sonoma wouldn’t have functioned without a continuous checklist of tasks that needed to be done and maintenance that involved multiple groups.

Yes, it was also an asset light business since we didn’t own the building or the upkeep, but we were beholden to an actual physical asset.

I suppose what I mean asset light is physical asset light.

It’s only been today that I released CashMoney.lol but it was still a learning experience. What I learned is that I released a product not knowing who the end customer was and it was a B2C tool in an industry notorious for fraud and shills selling easy wins.

It is not going to succeed. It just isn’t worth it for most people. Is it useful for me? Yes, totally. It makes it super easy to research companies, but it isn’t super useful as others can find the same information on Yahoo Finance.

Maybe a differentiator would have been to provide additional information such as yields, etc. However, again how to pitch this to others would be problematic.

I will continue building on this for other products, but it won’t be a primary focus per se for me and I will focus more on B2B.

As we get ready to launch more products I’m thinking the way to go forward is to have multiple domains.

• Each domain can be independent and have its own loose identity.
• We can build PHP functions to support our needs in WordPress.
• We can use third party tools for selling such as Gumroad, LemonSqueezy, etc.
• If we need to we can purchase WordPress for specific domains.

The pitfalls for this is that WordPress Subscriptions don’t work and we will need to come up with an alternative. Otherwise, not a huge deal.

At the core of what we do there is a set of things we do with our team & process. This is our model.

What we do for our customers is our controller.

Each marketing and sales and how we define what we do as a product is a view.

When I started opsZero I wanted everything within code. It made it easier for me to reason about and work on, but as we grow it doesn’t make sense for us to continue using this method as it is heavily dependent on engineering resources which can be used elsewhere. Instead it makes sense to push out as much of opsZero’s stuff into no code solutions. There are a few reasons behind this:

1. No-code tools like Notion are simply amazing. With AI and Automations they simply do more than we can do ourselves. However, because of their API we can accomplish what we need to if we need to augment anything more technical.
2. As we add more people using commodity tools is much easier than using specialized tools that we have built internally. Building internally built tools becomes a liability for our team as we end up spending time building things that we can piece together instead.
3. We will be able to move much faster. Our goal is to not have a ton of custom code except for where it makes sense, which are largely integrations.
4. Lastly, the stuff we are building are from a UX standpoint not complex. We do not need to build custom tooling around that and can get away with simpler UX built using simpler tools.

What we will be doing then is pushing out almost all things to Notion as our source of truth. This will then feed to WordPress, etc. The end result of this is that System6 will be going away. Most things that need to be done will happen in Notion with code largely being scripts that are run through Github and Github Actions. The end result of this is we can increase our security posture while reducing amount of custom code we have.

I think I’ve been thinking about products wrong. I don’t think a venture should encompass everything but it does get more valuable when there is more added to it. A single site selling widget x is less useful than a website selling x, y, z. However, some people may just want widget x and don’t need y and z.

The question is how do you build ventures that can do both. If we sell x, y, z then we need to segment but what if the thing you are selling addresses different markets?

I think it may make sense to build the larger sites then figure out which features are popular and build the unbundled versions. However, I haven’t figured out if this is something I want to do. Overall, having a larger bundled site provides a lot of diversification and the customer has a feature list they can use.

Likely a better approach.