January 25, 2022

How To Build A No-Code App — Learn From My Mistakes

Written by

I love the idea of using no-code platforms to speed up app development within a business, but the “anyone can build an app” marketing line is dangerously misleading. 

At TechnologyAdvice, the content team uses Google Sheets to plan and track our content production for over 20 technology publications. Admittedly, Google Sheets worked just fine when it was me and one other writer planning and tracking our blog posts for a single publication. But after three acquisitions and the addition of over 30 team members, managing and updating all of the tabs, formulas, and charts on that sheet takes up two-thirds of one team member’s working hours. And she rebuilds the sheet every quarter. 

We’re currently looking for a new project management software that fits our needs. We’ve narrowed it down to a couple of options, but the thought keeps coming back to me: We have a pretty good system built out in a spreadsheet, which is basically an app. Couldn’t we just turn our spreadsheet into an app?

Short answer: No. You cannot “just” turn a behemoth into a fully-functioning app by uploading the spreadsheet to the app tool. I’m as disappointed as you are. But because I have spent many frustrating hours trying to build this Ikea bookcase without ever looking at the directions, I have some advice on how to approach your first no-code app.

1. Learn the tool first

“It’s no-code! I’m not a coder! This was built for me!” she said, regretting every word. 

Just because I can do a vlookup with a YouTube assist doesn’t mean I have the technical knowledge to build a complicated app right away. In the world of no-code, you gotta start simple.

My first mistake was to jump in with a very complicated set of spreadsheets built with impressive formulas that run burndown charts, track budgets, and count a custom production points system. I thought that since all of this work is already in a spreadsheet, I could easily translate it to a no-code app. 

But I forgot that I actually had to learn how the tool works before I can build a complicated app. After playing around with a sample of my publication plan data for too frustratingly long, I found a beginner tutorial on YouTube with step by step instructions to build a task management app. The tools suddenly made sense.

While building a task management app doesn’t have all of the functions I want from my final content production app, it’s a good place to start learning the app. Turns out it’s a lot easier to work with 5 columns of data instead of 16.

Takeaway: Build a simple sample project or two to learn the tools before attempting to dive into that tough project. 

2. Start with simple data and build an MVP

Our content planning sheet has gone through many iterations, but because it works as both a planning and tracking document, the data is not consistent across tabs or even within a single sheet. Inconsistent data and missing fields and column quirks that work in a spreadsheet make for a frustrating app build. 

My “clean” spreadsheet included extra columns that were helpful for formatting but annoying in the app. I had also included lots of rows where I had planned blog posts but didn’t fill in any of the other columns. If I had started with fewer columns and rows, I might have been better equipped to manipulate that data when I got into the app.

But you’re building your app for your ideal user and your ideal state, so build the ideal data set. And make that data set fit the minimally viable product (MVP) that will suit your needs. What does the ideal behavior look like for the most basic version of your app? Start there. You can build in exceptions later. Consider:

  • Does your most basic app need to track budgets, or can you add that in later?
  • Does your most basic app need to include alerts?

Takeaway: Unless your team is hyper-organized, I would suggest building your first app from pared-down, clean, and as-close-to perfect data as possible. Fill out all the fields you can, or get rid of any columns you don’t immediately need.

3. You still need to know spreadsheet-style formulas

I can clean up a spreadsheet and import the data into a tool, but I invariably get stymied when the no-code app builder needs me to build a formula. I consider myself pretty tech-savvy (I know what an API is! I have a GitHub account!), but I cannot write regular expressions or formulas without the help of documentation and probably several video demonstrations. 

This might be completely obvious to most people who build software, but if you’re new to this, you’re going to need to invest some time learning to speak the language of the software you’re using. The magic of WordPress and WYSIWYG editors mean that most people who publish on the web don’t need to know HTML to do it. But if you want to customize your blog post, you might need to learn what an HTML tag is and how it works. The same is true with a no-code app. 

Takeaway: You will need to invest some time learning how to speak the app’s language in order to make it work. Most of these tools have communities, videos, and ample support documents to help you out.

4. Don’t fall victim to the marketing

This is my major gripe with the promises of the no-code world. The marketing says “Anyone can build an app!” “No coding necessary!” “Fast and easy!”

And sure, building an app on a no-code platform is faster and easier than building a web or mobile application from the ground up. There’s a reason that platform-as-a-service (PaaS) tools exist: Learning to build an app infrastructure is hard, so you should pay someone else to build it for you. 

Also More: IaaS vs. PaaS: 2022 Comparison

No-code applications are an extension of the PaaS movement, where companies use existing databases, servers, and operating systems to speed up their app builds. You don’t have to learn a whole new coding language to understand how to use a no-code tool. You don’t have to understand how to make database calls or even how to code the design elements of your app. This makes them incredibly useful for startups and individuals that are strapped for resources. It even makes them useful for care-mad employees at larger corporations who know that their current systems could be improved. 

Takeaway: Don’t be fooled that no-code means no learning curve. Set aside the time to learn your tool of choice and outline the right projects to build. Once you get the basics down, you’ll be building apps for everything — but that’s an entirely different problem.