I recently participated in a low-code conference at the invitation of Ninox, a low-code vendor. My colleague John Rymer gave the keynote, and we then had a lively discussion largely focused on citizen developers. Conference attendees hailed from (literally) all over the world; many were young entrepreneurs building businesses on low-code platforms or businesspeople solving problems within their groups.
This group was very different than our usual audience at conferences (corporate IT leaders) and suggests an exciting trend: that developing and doing business through low-code apps is becoming simply “how we work” at a grassroots level in many businesses. I cheerfully speculate that some of the great business leaders of the future were in the audience, building their startups digitally, from the ground up, using low-code platforms.
After the sessions, I stuck around for a few minutes and answered some of the written questions via the conference chat. I’ve cleaned up these exchanges and reproduced them here — these questions are relevant to everyone using or thinking about using low-code platforms.
Q: How does low code/no code differ from last-century 3GL/4GL development environments?
A: It’s the same value proposition but much better executed. Low-code products tend to provide visual development, software development lifecycle capabilities, cloud deployment, etc. The time, place, and mindset are the even bigger differences. Most people are now accustomed to running their personal lives on their phones, so digital work is much more natural for them. A lot more people have software skills. Also, spreadsheets have taught millions of people the basics of data and logic, and many others are comfortable building web pages and sites.
Q: What domains are low-code solutions targeting now, and what should they be looking to solve (near-future)?
A: Well, nobody’s using low-code platforms to make video games. Rather, the platforms are generally used to tackle many variants of “business applications.” Customer relationship management, order management, logistics management, marketing automation, many kinds of inspections, tracking pretty much everything under the sun, project management, online marketplaces, field services, financial processes, many forms of operations management . . . almost anything!
Citizen developers usually start with simple workflow or tracking apps — tracking defects, inventory management, logging hours, varied approval processes, whatever. If the platform is powerful enough and the developer is ambitious, apps grow in sophistication and scale from there.
Q: So where is the best place to store data via a low-code platform in your opinion? Would you agree Excel is a versatile data storage option to build a dashboard to monitor data inputs once populated with low-code apps?
A: I like platforms that include their own database builders. Excel is beautiful, but it ain’t a database. Learning relational data models (at least before that need is abstracted away) is generally part of the journey to building powerful apps.
Q: [In order to create citizen developers], how do we help team members who don’t currently have a “logical mindset” to start adopting that? Is this something not worth doing?
A: Teach them a “digital-first” mindset by digitizing simple things (like storing data online with some simple security and reports rather than keeping a personal spreadsheet or a paper binder). Remember, the baseline is spreadsheets. If the person can use spreadsheets, he or she can handle doing at least some development using low-code platforms. Also, start introducing concepts from design thinking or other structured problem-solving methods. “What problem are we trying to solve?” is a very powerful question!
All that being said, some colleagues will be better at larger, more complex projects than other colleagues — as with any other kind of work.
Q: Is there not a worry of universality of code? What about quality of code? Are we going to see a bloating of code due to the inexperience of novice programmers?
A: I interpret this question to mean that code is “universal” in that it is portable (avoiding vendor lock-in) and there is an established workforce that can write in the standard languages. The “worry” is that you’d lose those benefits by adopting a low-code platform where you can’t see or edit the underlying code.
To answer, some platforms are code generators where you can work with and edit underlying code (they aren’t so-called “black boxes”), so this addresses at least some of the concern.
Regarding quality, a good platform aimed at citizen developers obviates the need for “code quality” to a great degree. This is mostly because they write very little actual code. Instead, a developer might write simple business logic (like Excel formulas), or create screens with click-and-drag canvas, or draw a process map. The concern is less about code quality and more about design quality: Are these the rules and calculations we want? Is this screen attractive? Is this process effective? and so on. The underlying code isn’t something you should have to worry about.
And, finally, yes, there will be some bloating in general as the pool of novice developers explodes. My opinion: “Where no oxen are, the manger is clean, but much increase comes by the strength of the ox.” That is, there’s always a bit of mess to go with great gains. We’ll figure it out.
Q: Is the low-code market growing? Also, you mention the industry size was about $7 billion. Have you got any data on where this will evolve in the next few years?
A: There are about 100 products now (lots of supply), but the market will continue to grow for many years, especially products designed for citizen developers.
On the pro-developer side of the market (overlapping with traditional business process management), those vendors will sometimes fight over big deals, so we do think consolidation is starting there.
On market sizing: We estimate the market will be around $14 billion by the year 2024. It’s still growing, even during COVID-19.
Q: How likely is it that an existing IT department perceives low code as the ultimate form of shadow IT, threatening its existence?
A: Extremely likely — pretty normal, actually. But, in my opinion, this attitude is shortsighted and does not serve the employees or the business as a whole. There will always be a need for a technology team, but viewing IT as the sole provider of technology solutions in a business is truly last-century thinking.
The fact is that citizen developers are in your organization already (whether you know it or not), and much of their work is excellent (we’ve seen many examples), and that these people are only “shadow IT” if they are maligned, ignored, or neglected. Modern low-code platforms represent an opportunity for IT to empower these developers, collaborate with them, and act as mentors, all in a governed and visible environment. In short, low code isn’t a threat to IT — it’s likely the best solution to shadow IT we’ve ever had.