A Brief Survey of Low-Code and No-Code Solutions
We will take a quick tour of the low-code and no-code landscape. After a quick review of a few tools that have been around for a while, we’ll look at some of the new, web-age tools and examine a couple applications built with them. We’ll finish by talking about where the tools breakdown and how this might be addressed in the future.
Who is this for? This is for anyone who has heard about low-code or no-code but hasn’t had time to investigate on their own. Though not required, some level of technical know-how is assumed: you might be a developer, designer, PM, etc. Skeptics will not be convinced by the end of this but I hope some reconsider.
Nucode and Makerpad were extremely helpful in this research. They are low-code communities with listings of tools, templates, tutorials, news, and apps people have built. If you’re interested in learning more, they have a lot more information than you’ll find here!
A quick note: I am not affiliated with any company mentioned in this post but simply excited by potential this sector promises.
Low-Code vs No-Code #︎
I’ve mentioned both low-code and no-code as descriptions of tools. What’s the difference? That’s not entirely clear yet. No-code tends to emphasize more point and click styles with predefined blocks (for example: on click → signup user). Low-code tools also have point and click interfaces but their blocks are more low-level (for example: on click → capture email and password fields from form → validate → hash → insert into database). They sometimes allow writing textual code in mainstream or proprietary languages, as we’ll see.
Many people consider no-code to be a marketing term and no-code tools to be low-code tools in disguise. My opinion is that this is largely correct. I consider low-code to be more accurate in general however I think no-code can draw in more potential users who might otherwise be intimidated by any-code. That’s great! I’m in favor of no-code as a marketing position to bring more people to software.
The Tools #︎
This is simply a highlight of tools I’ve enjoyed researching and is by no means a comprehensive list. The tools listing at Nucode and tools listing at Makerpad have many more to look at for the curious.
Old Guard #︎
Low-code tools have existed for a long time. Software from Appian, SAP, Salesforce, and others has long enabled technical users to improve business processes with less code than might otherwise have been needed.1 There’s also LabVIEW to program scientific devices and instruments, MaxMSP to program audio and visual synthesizers, and game engines like Unity. All of this software is specialist software for specific domains.
Let’s not forget Microsoft Excel, still responsible for many billions of dollars of business activity, and it’s younger cousin, Google Sheets. Spreadsheets are a classic low-code example. Microsoft Access also deserves a mention here. These are more general purpose tools where any business or home user might find value.
New Wave #︎
Those tools serve their purpose and now low/no code is expanding heavily into the web with social, multiuser, and multiplatform experiences. Let’s take a tour through a small variety of these new tools.
Airtable is one of the most exciting tools around at the moment. It takes Excel and Google Spreadsheets further by adding a rich set of widgets and a multiuser experience on top. It’s like Microsoft Access but on the web with a better UI and better widgets. They have an extensive template library to get started and a big catalog of real examples.
Almost any company can get value out of Airtable. The only question is when it will become Microsoft Airtable.
Webflow is focused on web page design and content management. It’s a good fit for static informational sites, portfolios, blogs, surveys, and ecommerce. It’s probably the most powerful WYSIWYG tool on the market.2 The sites you build are responsive by default – important in today’s world.
Take a look at their showcase for some examples. If you’re curious to see what the tool is like, signing up is quick and easy but they also have great landing pages for designing, animating, managing content, and running an ecommerce store.
Glide lets you design and publish mobile apps directly from your browser. The apps pull their data from Google Sheets so the new project experience is great: point Glide at a spreadsheet and it will derive a default application based on the data in the spreadsheet. Neat!
I’m a fan of this approach for two reasons. One, people can use data they already have in Google Sheets and don’t need to learn a new data entry tool or upload a CSV. Two, the company is focused on building what makes it unique and not building yet another spreadsheet tool.
Pipedream stood out because it fills a need I have often felt as a developer: I want to make some quick data pipeline or workflow with good visibility into errors. In general, it’s not hard to make a workflow but finding a place to run it can be a challenge in some environments. I don’t want to maintain a daemon or a Jenkins pipeline. That’s the problem Pipedream solves.
You get a place to run workflows, visibility and error handling, and prebuilt integrations to a good variety of services. The downside is you can only write your code in Node.js. They’re a younger company and don’t yet have a paid tier.
Again, Pipedream has a gallery.
Bubble aims to encompass the entire application stack - from UI to backend to database. In that sense, it is the most ambitious in scope on this list and that’s why I’ve included it. You can build full web applications with Bubble. And people have! Check out Bubble’s showcase for some examples.
It’s very powerful but since it tries to do everything the interface is a little clunky in places. I’m sure there’s some hard work going on to improve this.
Divjoy is very new, having publically launched in late October 2019.
At first glance it’s a simple code generator to start new React projects. You
choose the libraries and integrations (including analytics, authentication, and
contact forms) you want to include and start a project. But you don’t
just download a
zip file and move on. You are dropped into an editor where you can
add and remove pages, change copy, and edit styling. Only once you’re done do
you download a
zip file containing your fresh new project.
This is useful enough, but if you go to their Divjoy for Teams page you’re met with the ability to signup for the beta and their vision for the future:
Divjoy for Teams is a visual editor for your existing React codebase. It understands your data, logic, and component structure, enabling anyone on your team to safely make changes and build new product experiences without code.
Product and marketing can try out new ideas with zero engineering cost. Your devs can stay focused on hard technical problems, with peace of mind that all changes made through Divjoy are tested and submitted as pull-requests for their approval.
Exciting! I like this a lot because it both integrates with existing code and allows you to drop down and work in code when the WYSIWYG editor fails. In other tools, you’re stuck.
That said, it’s not launched yet so we’ll have to wait and see how well the vision is executed. Divjoy is a project to keep your eye on. If you’re starting a new React project, it’s ready to use today.
The Artifacts #︎
Personal Finance App with Tiller and Glide #︎
I liked this because it shows how easy it can be to get started with app development and how powerful tools that can easily pull from other data sources can be. Dare I say it’s almost Unix-like?
I think it also hints at businesses that may be very successful in a low/no code landscape: those which harmonize different data sources into some lowest common denominator format for consumption by other tools.
Reddit Alerts #︎
A user going by
pravin built a tool to search Reddit and email matching
on Pipedream. This is nothing you couldn’t do with a
short script but that is sort of why I like it. It’s not always
straightforward to run that script consistently and reliably. Pipedream lets
you do that. This workflow is a nice mixture of custom code and off-the-shelf
Clone of AirPods Pro Product Page #︎
To show off the power of Webflow, Moritz Petersen made a clone of the AirPods Pro product page (warning: bandwidth and processing heavy page). Webflow has also published their own tutorial on creating an animate on scroll effect. This is some impressive stuff!
The Gaps and The Future #︎
If low/no code tools solved all problems that could be nice. But they don’t. Highly complex and interactive applications and applications with unique scaling concerns might never be solved (entirely) by the sorts of low/no code tools we’re seeing now. Beyond that, here are a few problems I see that can be addressed.
Security and Auditability #︎
If aggregating your banking information into a Google Sheet and pulling it down with some random app didn’t give you pause, you might not be paranoid!
On the one hand, by lowering the barrier to entry of software many people will inadvertently expose data or trigger actions they did not intend.3 The implications of building software are not always obvious. As tooling develops, I expect tools that help users understand what’s happening to their data and when will be important.
On the other hand, if you’re security concious in the first place and do understand how data is flowing, low/no code platforms can provide better security than alternate solutions under some threat models in much the same way hosting in the cloud over hosting your own rack can. Patches and updates are centralized and platforms will be hiring more security engineers than you will (and paying them more).
These tools don’t always play well together. They excel in different areas. No tool will ever be a silver bullet. It’s still the early days for the movement so there’s time to figure this out, but there has to be a way to integrate the best bits of different tools into your software project.
Open Source Software #︎
All of these tools are commerical and the artifacts you build can only run on their platform. As a long time Linux user, fan of free things, and proponent of owning your own tooling, I worry the low/no code (lonoco?) future will be systematically detrimental to development accessibility and independence. I acknowledge these tools, on the surface, make development more accessible and independent by lowering the barrier to entry. However, they are all out there on the web, in the cloud, and you can be shut out at any time for any reason.
Offline Editing #︎
Not all of these tools work well or at all offline which can be annoying in some circumstances. Sometimes there isn’t high-quality (or any) Internet available. Sometimes you just want to unplug and get some work done without distraction.
This is not the biggest problem but it’s irksome.
This is just a small sampling of some tools available, what can be built with them, and some of the problems that exist in the low/no code movement. There’s a lot more out there so if you’re interested let me again call out Nucode and Makerpad.
Low/no code tools are already enabling entrepeneurs to launch businesses they were not able to before. The rate of entrepeneurship will only increase from here and we should see an explosion of small businesses, some of which will be happy to stay small and some of which will grow to be large. We will see businesses bring software to niches few, if any, developers have thought about before. I would not be surprised to find the founder of the next LinkedIn or Facebook to be building products with low/no code tools already.
You can direct any questions, complaints, or ideas for future articles to me on Twitter.
- I’m sure there are plenty of readers with horror stories about implementations based on offerings from these companies and I’m sure many of them are true. I’m also sure many companies have gotten real value from these sorts of solutions and that there are plenty of horror stories across Java, AWS, Kubernetes, etc: bad implementations are not unique to low-code platforms.
- Webflow bills itself as a no-code product heavily but it is a technical product - to use it well you do have to have good design sense and technique. It’s easy to get lost in the powerful interface. Again, I think this is a good marketing perspective but it does reflect why I think low-code is a more accurate term: you still have to encode some technical information in order to use Webflow effectively. It’s coding but not coding as we typically mean (textual or low-level block diagrams). It’s better! It’s a direct encoding rather than an indirect encoding. This applies to all no-code solutions – I’m not just picking on Webflow ;) [return]
- They already are [return]