The rise of no-code platforms is revolutionizing the way digital solutions are created.
Traditionally, only professional developers had the expertise to build custom software, leaving non-technical users reliant on rigid off-the-shelf software, limited spreadsheet solutions or having to rely on costly experts and long development times.
No-code platforms now allow anyone to create digital solutions without programming, combining automatically generated or easy to build graphical interfaces, business process automations, and data creation and management.
In this article, we analyze what are the 3 main approaches to nocode, and why the data-first approach is the best suited for citizen-developers especially for creating internal digital tools.
Nocode is an approach that aims to allow anyone to create digital solutions without programming, with no computer "code".
As usual with digital revolutions, the solution comes from a real need.
The offering in digital solutions for business needs is really of three types:
Still the most popular tool since it's introduction 40 years ago (see below), because it's simple to understand and easy to use, and you can be autonomous - you don't need an army of consultants or the approval of your IT department to organize information. But of course very limited, of no use on mobile phones, with no user-right management, not GDPR compliant, no automation, no file generation capability, etc. In short : great to be free, but you cannot go very far.
Other great minds have spent time on designing a solution to meet a particular need, but unfortunately almost always rigid, complex, expensive, difficult or impossible to adapt to your exact needs and limited to their area of expertise - in short looks great on paper because you think you can save time but very soon you realize that you are stuck in chains and have to spend a fortune on experts to customize the solution to your needs and end up with dozens of tools that don't talk well to each other and are so complex that few people in the organization use them (except top management who are in danger of getting misleading reports because information is incorrect). So what happens ? Everyone goes back to... the good old spreadsheet!
You get exactly the software you want and you own the Intellectual Property Right, but very expensive to create and very long to develop, really months and years, during which you are stuck with the lack of productivity of not having the right solution - in short a very long and expensive path, that generates dependencies on others (expensive software developers) and ever growing costs (software maintenance, technology that quickly become obsolete that you need to replace, etc.)
Adding to the fact that only a fraction of one percent of the population knows how to create programs (and much less are really good developers), the promise to make it possible to create your own digital tools without programming, in days or weeks and on your own, instead of months or years and being dependent on expensive resources, is very attractive.
And indeed the offering, and the market, are really taking off. Gartner for example estimates that in 2024, 80% of technological product and services will be built outside the IT department and that there will be four times more Active Citizen Developers than the number of professional developers at large enterprises by 2023.
Let's dig a little more in the three main approaches to the nocode revolution: UX-first, process-first, and data-first.
Why are these the main three approaches? Well because they are really similar to the logical structure of all modern computers. A user interface, a Central Processing Unit (CPU), and memory to store information.
The UX-first (User Interface/User Experience) approach focuses on the user interface, by making it simple to drag and drop graphical components to organize your screens. They are already much used for building web sites, with tools such as Webflow, Wix or Strikingly. But these consumer facing solutions are really good when data does not change regularly and does not need to be modified by the user directly. Several companies have tried to go beyond and use this approach for business tools. Popular tools include Bubble, Retool, WeWeb, Ksaar or Power Apps.
Although these tools make it possible to create user interfaces that are very precise, they typically take a very long time to learn (several weeks or months), and need to be interfaced with other tools for automation and for databases, which require much more technical skills. For this reason, they are considered more LowCode solutions than really nocode platforms.
On the positive side, the savings can be important (typically 3 x less time to develop and maintain tools versus traditional software development But, because of the long learning curve, the complexity of having to add logic to each graphical element, the complexity of databases and because interfaces built using tools are still difficult to modify, you are still dependent on qualified resources and cannot really do things on your own or expect non qualified business users to create their own tools.
The Process-first approach (or automation-first) is another approach that focuses on the logic. What happens when something is done? A user fills a form, what should you do with it ? Create a new record in a database ? Send a request for validation ? Send an email confirming that the information is being processed ?
There two main sub-categories here:
The RPA (robotic process automation) path, which is the capability to replicate clicks and behaviors of a real human to automate tasks on computer programs. Nice because you don't need to change existing software or to access or to create complex APIs or interfaces between programs to make things work, but very dangerous because you are dependent on existing user screens and if one is changed at any given time, the complete processes fall down. And if the existing tools don't do what you want to do, you are stuck. So great to automate a few things and avoid repetitive tasks, but at the end pretty limited. Popular tools here include UIPath, plus many specific tools focused on automating one app, such as Waalaxy for example that will automate your Linkedin activities.
The second sub-category in the Process-first approach is the API (Application Programming Interface) path, which works well for modern tools that developed APIs to create / read / modify information stored in their internal databases. They are typically based on the concept of triggers and actions. When something happens (a trigger in any of the tool), do something (an action in another tool). The promise is very interesting. Instead of each tool having to build one interface for every other tool on the market, the software creator simply needs to develop an interface to the automation platform, and instantly the tool is compatible with hundreds or thousand of other applications. Very, very powerful.
This approach has been instrumental in the growth of the development, because you don't need to find a single tool that does everything, but instead you can focus on a few core platforms and then connect to other tools to fill the gaps , extend the capability, meet quickly user requirements. An important benefit is that you will almost always be capable of finding an alternative path (a 'contournement' as we say in french, a clin d'œil to our friends at contournement.io who have been so proactive in developing the nocode movement in France). Popular tools here include Zapier, Make, n8n or Power Automate. These tools, however, are not standalone, and need to be complemented with a UX and data platforms.
Finally, the third approach to #nocode is called the data-first (or information-first) approach, which focuses on managing data. Let's dig a little more in this approach and see why it is the most interesting for business applications and for citizen developers.
But first, let's start by looking at another revolution: the arrival of the personal computer and it's first killer app: the spreadsheet.
Although the terminology is relatively recent, solutions that allowed individuals with no programming skills to organize digital information have existed since the arrival of the personal computer in the late 1970s, and in particular with the first spreadsheet software. Steve Jobs explains very well the importance of the arrival of Visicalc in 1977 and how it was really one of the two most formidable "explosions" that propelled the personal computer industry forward and the reason behind the success of the Apple II.
And indeed the explosion of the personal computer industry thanks to the spreadsheet was really quite a revolution at the time. For example at least 400 000 accounting clerks lost their jobs in the US (or had to learn new skills) because of the arrival of the spreadsheet. And why was it so successful ? Because it allowed business people to manage information easily and on their own, without having to be dependent on external resources. The key was not to provide an easy way to create a computer interface, but an easy way to manage, organize, display and extract information.
When you think about it, 99% of all business tools are really simply ways to manage information.
A CRM (Customer Relationship Management) tool for example, really consists of four data tables : customers, contacts, opportunities, and activities, with the appropriate way to enter, edit, link and display information
Ideally with the right automation (e.g. send a reminder or an email), computations (does this lead meet certain marketing qualification criteria) and the user-right management capabilities (you don't necessarily want every one to have access to all the information).
Displaying information in the right form is important, e.g. a Kanban view for your opportunity pipe, a Form view to display information about one contact or a Calendar view for your activities (scheduled meetings etc).
But it's much easier if the views and forms are automatically generated in one click:
Similarly, a project management tool is mostly about making it possible to manage dates, statuses, tasks, comments and people assigned to tasks.
Of course it should also be possible to display information in the right way (Gantt, Kanban or Timeline views for example) and to create processes and automate certain actions (for example, sending an email requesting validation of an important milestone or changing the termination date to today when a status is changed to "Done").
A calendar is simply a visual way to manage events.
And emails simply the capability to exchange information with someone else by creating and sharing data from one database to another (with data fields such as subject, body, from, to, attachments, etc.)
So instead of buying an application, or of building an application, why not focus on what is common to all business applications ? Data !
TimeTonic calls this "Data is the application". The key is not the tool or the interface, but what you want to do with it. Information you want to extract, link, understand, manage, and the decisions you want to make based on this data.
A real paradigm shift: instead of buying or building a tool to solve a problem, start with the need, transform it into data (what information do I need to manage) and voilà: the application is ready !
The arrival of smart visual databases, as flexible as spreadsheets, aim to fulfill this promise.
They make it possible to focus on the core needs while keeping the promise of making it simple : you can learn to use them in hours or days, thus allowing anyone with no prior computer development knowledge to really become citizen developers.
The best platform will include built-in collaboration (because for perfect management you need both a good organization of information and a good way to share information, both internally and externally), smart user-right management (including mirror views filtered by connected users to create customer or partner portals), advanced document generation (to generate pdf for invoices, quotations or field reports), computational power for advanced calculations, instant views (Table aka spreadsheet, Kanban, Calendar, Gantt, Timeline, Map, Kanban, Pivot / grid views, etc.), dashboard capabilities and elegant external forms to capture (and ideally to share as well) information.
So what kind of #datafirst nocode platform exists ?
They are all based on this concept of a smart visual database as flexible as a spreadsheet and yet are as powerful as the most powerful tools like ERPs and CRMs.
Of course each one has its key advantages (this will be the subject of another article): like flexible UX interface, more APIs, or an native mobile application that works offline - ideal for field service personal (in factories, construction sites, building maintenance sites, etc.)
And with the power they provide in terms of data management, automation, calculations, user-right management, APIs, forms and views, you end-up with an ideal approach that meets all the key requirements of the perfect nocode platform:
You end up with a platform that gives you freedom (you are not dependent on external tools, software developers or external consultants), time (you can solve you business needs much faster) and agility (you can adapt to threats or new opportunities almost instantly).
As you see, all mighty competitive advantages.
And when you see the current hype about the other big revolution in the making, Artificial Intelligence (AI), and the rise of promising tools such as chatGPT, you understand that indeed, data is the application.
In summary, the "datafirst" approach makes it possible to focus on what is really important in all business tools: information.
The best datafirst nocode platform make it simple to :
Almost instantly, anyone in the organization, not just the IT department, can now create the digital tools they need and that are essential to their success.
Based on smart visual databases as flexible as a spreadsheet and as powerful as an ERP, these solutions are easy to learn, easy to use, and give a mighty competitive advantage to the organizations that deploy them at scale.