Are there any products which have been successful and made with no-code tools? Seems like I'm always seeing no-code tools, but never any proper products made with them.
Sure there are website builders like wix which are used for personal blogs and such - but that's not really what I'm asking about. Or else there's wordpress: but I'm loathe to call wordpress no-code because of the amount of complexity in managing plugins - we've got whole "engineering" roles dedicated to tweaking wp settings.
Surely this whole idea of "You don't have to write code" quickly turns into "You can't change/fix the product" and that's a whole other problem.
I work in low-code and make a tool so I know this area fairly well. I can't think of any proper product made with no-code tools now but I can think of many that started with low-code and I imagine you can too. Think of every internal application that started as a spreadsheet, or a single web page, maybe even static.
What I tell my potential customers is, that it's a pyramid of effort vs exploration. At the bottom you have a wide base of possibilities you want to explore, at the top of the pyramid you have a proper polished product. What low-code allows you to do is try many more experiments at the bottom and to maintain them at cheaper cost. Most will stay as internal tools and never become a real product but some will. For example my tool is letting data analysts in crypto firms built out internal trading controls, trade monitoring and alerts within days and weeks. The happiest customers I have are ones that were about to hire a UI guy. It let's them delay that hire. Then when they want to provide end client interfaces or seroiusly improve a low-code version to a separate polished app, they hire the UI expert and even better the UI person can see roughly what is needed.
The alternative isn't always between polished app and low-code, it's between apps that may never get built and low-code.
I suspect most of the low code products are internal-only products and proof of concept apps. Like spreadsheets, other teams find it empowering to be able to make automations without getting the engineering team involved.
Umm, what automatations can be done without engineering team? Even a readonly BI access needed a few meetings with engineers
The entire idea behind no code is that you develop a throwaway app in a few weeks. You could think of it as a spreadsheet with a more complicated UI.
This is for bootstrapping. After the code is generated you need to use it.
We used JHipster (mentioned in the post) a few years back. It let us get started with a lot of boilerplate and best practices. I don't know if it was the best approach but we had a fantastic demo in no time and the basics were pretty good.
We did fix a lot in the code after as the product kept growing.
More than once I was brought to fix some customers low-code applications, when they reached those edge cases no citizen developer even thinks about. Or just grow to handle more data than the initial thoughts. Did you know Power Apps can fetch a maximum of 500 records only? Which you can increase with an obscure setting to 2000, but that's the absolute maximum? Do you think the citizen developer reads those specifications?
I developed a what I would consider a pretty small application with Power Apps, and it was an incredibly impressive platform. Internal application, about 200 users, 2000ish records of data generated each day.
However, I wouldn’t want to build anything in it again. The amount of gymnastics required to work with and around those kinds of limitations (500 records) through pagination, error checking, retry logic. I ended up needing to implement things I wouldn’t want even my worst enemy to maintain. Definitely nothing wrong would expect a citizen developer create or maintain.
But it’s ability to easily bring together multiple data sources, consistent uni-directional dataflow, and lots of other things made it a rally nice platform to work on