How to Make Early Technology Decisions for your Enterprise SaaS

Spencer Jones Saturday, June 24, 2023

When you are first starting your Enterprise SaaS product, there are a myriad of decisions that can slow you down. It’s hard to know what’s trivial and what is not.

What to optimize for?

Here’s a framework for how to make decisions.

Choose boring technology.

In almost every case with very few exceptions, you should choose stable and well-known technologies rather than newer and hotter tools.

New technology comes with many unknowns. Invariably, issues will come up that will take the focus from building your product. It’s best to avoid these types of delays if you can.

Technology that is stable, well-known, and has broad adoption brings a lot of advantages:

  • If there are issues, it’s likely many others face the same issues and you’ll find lots of good resources and debugging help.
  • A broad ecosystem of tools, plugins, and extensions can save you tons of time from building in-house tools.
  • Readily available learning and training materials will make onboarding easier.

Hiring

Some technologies are harder to hire for, and talent is therefore more expensive. In today’s climate, hiring developers is hard enough and many companies are competing for top-tier developers.

Choosing languages or tools that only a few developers know will likely make it harder to find and hire developers.

Languages like JavaScript/Node.js, PHP, and Python have vast communities all over the world.

Optionality

In the early days of product development, you make a lot of assumptions about the needs of the users and what functionality you will need.

Many of these assumptions will likely prove to be wrong.

Therefore, your technology should allow flexibility to change as new requirements come in.

Slow Decisions, Fast Decisions

When to Consider Carefully

Some decisions will be harder to undo, and will be very costly.

Things like your database, programming language, and frameworks will be extremely difficult to swap later. The good news is that these pieces have a myriad of good options today.

For your database, default to a SQL database like Postgres or MySQL. Not only is SQL widely known and understood, it has stood the test of time for many products.

Any programming framework that has broad adoption is likely safe. For instance, on the frontend, React is a very safe bet for Single-Page Applications. On the backend, there’s are usually a few dominant frameworks in each language community - Laravel for PHP, Django or Flask for Python, express or Nest.

You should carefully consider how much of your business you are risking on a vendor. If a vendor is new to the market, it’s likely that their service will be less stable or the company may disappear altogether. It’s safer to pick incumbents with at least a few years in the market.

When to decide, and move on

If you have multiple reasonable options with no obvious flaws, pick one and keep moving. It’s not worth the time to debate these.

Low-level libraries that provide specific functionality should not require extensive consideration.

The Good News

In the world of software, any early technical decision that you make can be changed given enough effort.

Invariably, something will not match the future needs of the business and you will not to invest time to undo that.

The trick is to thread the needle between technology that enables you to move quickly to market, adapt as you learn, and grow with your business.

Want help setting up your Enterprise SaaS’s technology?

Take a look at Ghost CTOs services. We can help you pick the right tools and technology for your business.