Low-code/No-code development platforms – Application development platforms of the future

This topic, by its very nature is debatable and disruptive as well. These upcoming development frameworks are already challenging the way, in which decades-old software development has been happening and it is about to transform the entire computing industry – at least a substantial chunk of it – for the better with unprecedented disruption!

In simple terms, low-code is a radical approach to building custom business applications – seriously FAST! You are not required to build application by writing code line-by-line – you draw it like a flowchart. Forrester Research defines low-code as “…Application development platforms that enable rapid application delivery with minimal hand-coding, and quick setup and deployment…”

Low-code/no-code already has the potential to be the future of application development, and its very ethos is based on shifting the power of custom application development and tailor-fit programmable components being available for citizen developers – a term that was invented and has evolved as a direct outcome of the low-code engineering paradigms. This has been one of the, if not the, most revolutionary developments in software engineering domain. Despite this fact, many legacy organizations are yet apprehensive for adoption of low-code – that fear largely exists due to two factors, viz., their lack of clarity and visibility on the capabilities of low-code platforms and lack of wide-spread availability of low-code framework developers community yet – and rightfully so, since everybody wants to invest in a framework that will sustain the enterprise requirements for a decade (if not more), and that the framework does not get outdated on the technical competence yardstick.

Low-code engineering platforms allow creating enterprise-grade applications with crazy amount of automation and ease of engineering, without needing the micro-level involvement of the core engineering teams/programmers. In fact, average tech-savvy individuals and even business users can create business applications by defining business-process workflows and decision-making logic, via simple drag-and-drop flow-chart lookalike controls within most low-code platforms. Yes, there will still be some degree of customization needed, especially on the processes where your application is about to consume external API’s or is about to connect with any of the Legacy Tech frameworks. But by and large, the overall average engineering time can get reduced by up to 60 percent – which is a substantial saving of time, efforts, and cost, especially when we consider the magnitude of time and money most enterprises invest/spend in custom-development of their products.

Recent market studies and research by Gartner and other eminent setups indicate that the low-code development market will grow to USD 28 billion by the year 2022 (just a few months prior to the Covid-19 outbreak, that number was USD 4 billion). With the entire global economy in doldrums, the post-pandemic world will be drastically different from what it seemed to be in November 2019. Profit margins in sumps, cash reserves at their lowest ever, teething need of rebounding with vigor to regain market share, and dire need to optimize every penny spent, will be the new norm of the post-pandemic world. Low-code application development is about to prove itself as a blessing in disguise in this new world order!

Most enterprise application development activities can be largely compartmentalized:

User interface/front-end development takes approximately 20~25 percent of the development efforts depending on the complexity of the flow;
Middleware and backend takes approximately 40~50 percent of development efforts. This includes design and development of business logic layer, data architecture layer, CRUD operations, workflows, and schedulers;
Testing. Automated and manual testing takes about 15~20 percent of the entire engineering efforts; and
Deployment. Configuration and release management takes about 10~15 percent of the overall efforts.

Most low-code/no-code platforms of the present offer various degrees of support and inclusion of the above four processes.

Some of the differentiating factors of the low-code/no-code development framework:

(a) Low-code fully supports crazy bleeding-edge tech. All low-code environments have, literally, hundreds of ready-for-consumption API options. Almost all these apps are future-tech-ready for up-and-coming technologies that can solve our day-to-day business problems with a few clicks. For example, machine learning and artificial intelligence is one of the hottest evolving technology these days, and everybody has huge curiosity toward its real-world applicability and its infinite possibilities to derive results. Most low-code platforms these days already offer click-to-activate toolkits for ML/AI and these can be easily included within our workflows with near-zero programming efforts. For example, you can easily activate an AI-powered Chatbot that spawns through your existing knowledge-base and derives its own probabilistic model for responding to the customers’ requests, will allow you to tweak or custom-feed the knowledge-base with your domain-specific questions and appropriate responses, and will be ready for an early release within a week, if not days. The ability to handle vernacular responses or features like interpretation of text to speech or speech to text, speech recognition, voice authentication, etc., can be effortlessly implemented for a quick deployment, as against the superhuman efforts these would have taken in conventional programming practices.

The same holds true for working with IoT and Blockchain-related innovation tech-enablement too. Most power-user grade individuals can create business process apps for tracking of shipments, logistics, and orders in real time, while keeping a tag on their entire supply-chain management. In the recent lock-down times, most home-essentials-delivering setups have been exploiting the power of low-code and IoT to manage, optimize, and fulfil their entire logistics and freight management. The logic behind determining what needs to get shipped, in which locale, by what time, and with what level of urgency, along with ground-level real-time tracking of the shipments as they happen, is now handled by ML and IoT running over low-code platforms.

(b) Low-code improves agility. Adoption of Agile+Kanban frameworks has become essential in these ever-changing and hard times. Low-code, by its very nature, not only helps but also encourages organizations to adopt agility. Infinite ability to refactor the frontend and UI/UX elements without any need to code speeds up development exponentially. Built-in automated testing tools and plug-ins help save on the laborious QA/QC as well – resulting in far lesser frustration for product and project owners. You can build applications faster than you ever imagined to, before. Simply put, less code means fewer bugs.

(c) Low-code helps directly reduce costs. Quick iterative modeling, quicker application frontend refactoring, super-fast minimum viable product release time-windows and early-to-market schedule, combines in reduction of time and costs – both. Additionally, since you can do more of modeling and lesser of coding, your team headcount gets reduced dramatically – further savings! And since the cross-functional teams are now empowered to actively participate in modeling business processes, low-code makes all the departments involved – not only Tech – even more productive!

(d) Low-code helps attain greater sprint velocity. Application process definition, frontend development, several frontend and server-side validations, automatic adoption of best practices, and automatic optimized rendered code, all results in your ability to build/deliver more applications in shorter time. The application delivery time gets reduced from months to days and, in some cases, some components of the application literally take few minutes to build. All these paradigm shifts result in tech team investing more time on fresh thinking and innovations, instead of doing mundane tasks. tt low-code platforms have automated CI/CD in-built into their deployment process, which almost eliminates the pain of configuration and release management

(e) Low code nurtures better customer support and experience. Since low-code has great ability for quick delivery of MVP and ability to adapt/adopt repetitive refactoring and change requests, the downstream effect is dramatically impacted and enterprises can thus speedily adapt to ever-changing market and customer demands/needs.

(f) Low-code helps in mitigating risks and offers better governance. Since there is great amount of consistency and uniformity in the codebase and adoption of by-default best-practices/guidelines, which is the very nature of low-code, a large part related to governance and risk management is automatically adhered to, during the entire development cycle. The famed segmented multi-tier architecture, most low-code frameworks offer, allows unparalleled levels of consistency and discipline in the manner every process is defined. This takes a massive load off, of the developers, for risk and governance adherence.

(g) Low-code is happy to accommodate frequent changes. Low-code allows extreme level of flexibility in handling frequent changes to frontend, business logic, data access layer, etc., thereby encouraging a harmonious culture between business and technology teams. Frequent changes have little to insignificant impact on the overall MVP dates and hence low-code excels in addressing one of the most perennial pains of software engineering!

(h) Low-code allows device- and operating-system-agnostic development. Building a common/single code-repository for all-devices-and-operating-systems-agnostic product, is possible ground-up in most low-code platforms. This helps organizations in reducing the additional headache of maintaining multiple codebases and spending harrowing time, identifying and fixing defects over multiple devices. The products you build from scratch are all-encompassing functional from day zero till their dev-completion.

Some of the popular low-code development platforms are:

Microsoft PowerApps, Google AppMaker, Betty Blocks, Mendix, QuickBase, OutSystems, Appian, Visual Lansa, Kissflow, and Salesforce Lightening.

So, what’s the catch…?

(a) Most organizations are largely unaware or ignorant or both, when it comes to adoption of dramatic/exponential technology transformations, which has been one of the primary reasons for low adoption of low-code/no-code platforms in mainstream corporate establishments in India.

(b) Availability of experienced/skilled developers on LCDP has been another challenge. Almost 90 percent developers or development partners functioning in LCDP space cater to development and support of ultra-large-size applications for companies abroad, thereby being able to demand dollar-based billing and frequent long-term off-site appointments for their developer staff. The swanky lifestyle abroad, combined with exponential pay-packages, has made it tough for SME enterprises in India to identify and partner with suitable and affordable set of developers/dev-partners locally.

(c) LCDP being new in the overall engineering space, has caused a recent surge in technology training institutes, focusing merely on churning out LCDP engineers. Freshers from engineering colleges are hired, trained in LCDP, and deployed on large projects at meagre hourly billable, while the companies who hire them, are ripping-off clients by raising exorbitant bills.

(d) This model also seems to be a risky one for the developer, as if he is separated from the LCDP environment, there is little-to-no engineering expertise available with him, for him to code on alternative non-LCDP frameworks. (This is like when you want to learn to drive a car, instead of learning how to drive a manual transmission car, you directly learn to drive an automatic transmission one – which, to my mind, is not the right way to determine whether one can drive or not!).

Outlook

LCDP is here. It is here to stay and its capability to deliver large-scale mainstream rugged enterprise-class applications is proven already. If the scarcity of seasoned resources part is addressed and a significant level of normalization of the expenses incurred in the adoption of LCDP is addressed, it is, as on date, the best development framework to be working on, since it is ultra-portable, ultra-fast, bleeding-edge tech-ready, and extremely sustainable!