OptiTax

OptiTax is available publicly starting today!

If you own a Swedish AB (aktiebolag) please try the MVP: https://opti.tax

It is free for 8 days if you sign up in October 2019. After that it will require a subscription at a price way below the potentially achievable tax savings.

It was an exciting new experience to click the button and finally deploy it to production. It includes the fruits of all the verification work by my accountant, and many rounds of fixes and improvements from my part to improve the user experience.

Two of my former colleagues who are consultants these days tried OptiTax as well while it was in closed beta. I'm grateful for all the feedback received from them, it made the final product even better.

Big part of the job was to translate the whole application to Swedish, since I originally developed it in English. I'm saying thank you to Anna for verifying all the translations and helping me to get them right.

I am still trying to keep it to the bare minimum (MVP, remember?) which unfortunately shows on the bit outdated Web design. I will address it later if the market receives the core product well enough.

It is time to concentrate on the marketing efforts. The primary audience are small business owners (AB, aktiebolag) and accountants in Sweden. A niche and relatively small market.

OptiTax as a product has measurable customer value, so the quantum of utility is there. I'm confident it can be sold to the target audience, it is "just" a matter of how to reach them economically.

Marketing is the next challenge for me, since I have literally zero experience in it. It is part of the learning experience I wanted so badly. I'm really excited to talk to real customers and see what results they can achieve with the product!

Contrary to the common startup advice I won't use the tools I'm the most familiar with. Why? I would like to turn the product development process into a learning experience, so I gain transferable knowledge from it even if the product won't succeed on the market. Therefore, I choose a tech stack I am not yet (deeply) familiar with and accept the hit expected on productivity.

Choosing simple and reliable tools for each task should result in an efficient workflow. After spending some time shopping around I settled with the following tech stack:

• Formal API definition: OpenAPI v3
• Middleware code generation: OpenAPI Generator
• Backend: Go (single static binary)
• Frontend: Elm + Bootstrap 4
• Persistence: JSON files
• Web server: Nginx
• Hosting: Linode
• Authentication: Auth0
• Monitoring: Datadog
• Email: Mailgun
• Payment: Signing agreements with BankID + manual invoicing

Yes, you've read it well. No fancy databases or containers. At least not at this stage. The solution need not scale just yet, but it must keep the possibility open. The backend is stateless, which allows for future scaling. The persistence layer has a well defined API internally, making it possible to switch to any scalable database solution later.

Basically, I deferred as many architectural decisions as possible according to Bob Martin's ideas from his great presentation: https://youtu.be/PWkeqFymteY

As with many other startups it all started with having a problem myself.

Back in 2017 I was lucky enough to get a great consulting opportunity because of a massive layoff. Being a consultant involves sending monthly invoices, so I created a small company (AB: aktiebolag) here in Sweden. I hired an accountant and started working hard. Revenue began to arrive at a steady rate.

As in most other countries the revenue of my company does not translate directly to net income. After a few emails with my accountant I had to realize how complex the Swedish tax rules are and how seriously they impact the net income I can achieve from my business.

It turned out that my biggest cost is the tax by far. Consultants have to pay both the income tax as an employee and the payroll tax as an employer. There is no family taxation, income tax is calculated individually. Dividend rules are complex and there is profit tax on all revenue not spent by the end of the financial year.

It turned out accountants have the software for the "forward" tax calculation, but cannot tell how to wisely choose the numbers we still have control over. Widespread practice is to just set an arbitrary salary and accept the tax hit later.

"Only Skatteverket knows the final tax to pay."

(Skatteverket is the tax office in Sweden.)

I did not accept this situation, since I would like to control my situation as much as possible, while still staying within the legal limits. I looked around for software solving my problem.

For income tax there are online calculators available, but my case was more complex with the small business involved. I needed a tax forecasting tool covering both the business and the personal income taxes with all the complex dividend rules considered.

What does a software developer do in such a situation?

Yes, your guess is right. I ended up implementing the whole tax calculation in a Python script. The first version was more of a hack giving a rough estimate, but still better than nothing.

I put the whole "forward" calculation into a few embedded loops to search the problem space for the global optimum of highest net income while keeping the total cost constant. This optimum is the same as paying the least on taxes, the two definitions are equivalent. Also plotted the results to visualize the solution space. Here we go, finally some usable results.

To my positive surprise I could save a significant amount this way. All I had to do is choosing the right salaries for me and my wife, then paying out the rest of the revenue according to the dividend rules after the financial year is closed.

It worked well for me in 2018, but what other consultants can do? Running my Python script? Not too straightforward solution.

This is how the idea of OptiTax, an online business income forecasting tool was born.