3 Reasons Enterprise Architects Need a Strategic Low-Code Platform Now
It’s 2023. Low-Code Development Platforms are now mainstream. However, adoption is relatively low in the UK and Ireland markets where Version 1 operates. This article explores why that is understandable, but potentially limiting IT’s ability to keep up with the growing demand.
Large organisations have typically walked a tightrope between off-the-shelf software for common functions that can be quickly implemented and deployed, and custom software where it can drive competitive advantage.
However, there are challenges with both approaches — customising on top of packaged software can make the update cycle hard in the medium term, and traditionally coded software can present a range of programming languages and models if there is no strong supported pattern and governance around it.
Most companies are also now considering the advantages of cloud hosting versus on-premises data centres, and balancing scalability and efficiency with the absolute control that owning your hardware can provide.
And everyone is looking for a modular, composable, integrated architecture, where capabilities are exposed as services, integration is available across internal solutions and a growing SaaS landscape, and good governance can ensure effective planning and change control.
In reality, delivering against that vision is hard. Businesses expect better customer experiences, streamlined and automated operations, AI assistance, and agility to changing market demands. The backlog is getting bigger, but with more technology and skills required, it’s hard to keep up.
So how can another platform help?
Put simply — in accepting that bespoke development is needed, and to avoid technology sprawl, a low code platform can be the starting point for every “how do we build that” question — we orchestrate and assemble the user-facing systems and processes, and then we connect with other composable functions where specialist software is needed.
For example, if an organisation needed an answer to the question “How do we improve our customer-facing applications?” — there could be capabilities like web, mobile, voice integration, scalability, security, and license growth that would be interesting.
Or, if the question is “How do we combine data from multiple back-end applications to streamline internal operations and improve workflow visibility” — there may be more questions about integrations, reusability, and role-based access.
By defining the criteria for what a good solution looks like, a proof-of-concept and first “real” project can be tested and measured against traditional approaches.
Once there is an outline of the shape of the strategic solution required, it becomes easier to evaluate options, work through a license purchase and internal/external resourcing options, and start to build the team to execute.
So — Why OutSystems?
A High-Performance Low Code Platform like OutSystems can deliver against many of those important criteria for the CIO.
First, it’s quick.
Quick to learn, as a new tool for experienced developers to pick up, or as a first tool for more junior employees — the first app can be built in a few days, with plenty of support as the complexity grows. And quick to deliver, with visual and AI-assisted development enabling the team to work through up to six times the speed of traditional coding.
That speed means that you can deliver more of the backlog, faster — and therefore cheaper, than the alternatives. Indeed we see that low code app development is often just as quick as implementing a COTS solution, and delivers on the unique DNA of the organisation far more than a generic industry solution can — and with a modern, clean UX as well.
But it’s also quick to pick up — complex logic is coded as flowcharts, so it’s easy to watch through the debug process and see where an unexpected result occurs. That makes it far easier to support than working through someone else’s code, no matter how clean.
OutSystems investment in AI since 2018 also provides natural language queries, suggested next action, real-time code reviews and static analysis for antipatterns, and they’ve just doubled down the investment in prompt-based app generation, with a handoff back and forward to the developer to manage it over its lifetime. All of that investment speeds up the process but keeps the developer in control.
Next, it’s modern.
OutSystems provides an enterprise-grade platform that delivers on the need for security and scalability.
In our recommended cloud configuration, the platform is fully managed and all of the hosting is included in the license, so a growing database or increased demand on an app server is not a problem. There are ISO-grade support processes, routine maintenance and vulnerability patching, and optional add-ons for intrusion detection.
Beyond that, the platform is actually generating traditional code, predominantly in React.JS and .Net, and in doing so is using best practices across the board. Of course, developers need to look at role-based access as a functional requirement and ensure checks are made on both the client and server side, but the platform is also helping out — with warnings and errors raised inside the development environment if issues are created, and a nightly code scan for known anti-patterns also helping keep technical debt under control.
By abstracting the code creation, updates in technology choices can be made — recently we’ve seen the standard shift from traditional web server apps to reactive single-page apps — following the trend for more processing on the end user device and opening up options for performance in the browser. We’ve seen PWAs and native mobile apps generated from the same configuration, and we’ll no doubt see a shift from Cordova-based apps to Ionic as that acquisition starts to bed in.
But most impressive is the shift to cloud-native, from cloud-hosted. OutSystems have spent the last few years working with AWS to take the same fast, visual development experience, and build it into a containerised microservice pattern, that will be deployable to any host environment. Companies will get all of the advantages of speed, scale, and truly global deployment, but abstracted from the 80+ components that make it happen.
Last, it’s adaptable.
Software has never had a shorter lifecycle than it does now — the days of a big bang project and handover to sweat the assets are gone. Business is changing, customers are more demanding, and new market capabilities are presenting opportunities at a faster pace than ever.
With a low code platform, the application lifecycle management has to be a critical part of any decision. The OutSystems approach is to build in a modular architecture, following simple rules about ownership, to enable decision-making. Applications can be loosely coupled and deployed independently, while the platform keeps an eye out for breaking changes between dependencies. Regression testing can be baked into the release process to check for negative impacts.
Integrations are critical in a modular architecture, whether between low-code components, large functional system services, or cloud SaaS offerings. Updates can come unexpectedly for a variety of reasons, from the cost or performance of external solutions to mandatory updates forced by an internal risk analysis. Fortunately, these are all straightforward, and with good version control, we can strangle consumers of deprecated components until fully replaced. This also means we can pull in the latest capability like GPT as an integrated offering, as quickly as you choose the provider.
This modular, integrated architecture is really driven towards active re-use. With proper version control, developer roles and access defined into teams, and an oversight tool for dependencies, the platform truly enables governance of change — allowing new projects to go even faster by providing common data models and connections, but slowing down the pace of change for well-used components.
How do companies get started?
Whilst the decision to embark on a low code journey will always start with the first project, it’s worth looking at that initiative as a test. When success is proven, the platform should have a purpose within the architecture — it can be the customer experience engine, the internal operations orchestrator, and the home for custom development in those spaces.
That strategic purpose for the platform should come with some capability. The most successful companies grow their talent internally, and supplement with partners for larger projects or ongoing support. Organising that into a product model, where the product owner owns the prioritisation of scope on the platform, and the team take on projects according to business value, can prove much more effective than individual project procurement. As that capability grows, a centre of expertise emerges, and teams divide into specialities and governance as well as rapid business-facing development.
As confidence grows in the tools and processes, the applications can grow in scale, complexity, and value. Companies such as Vopak, who have replaced their ERP landscape, and Western Union, who provides a global money transfer app, show how much is possible.
In Northern Ireland, we’re proud to have built solutions in OutSystems for processing school applications for 75,000 kids a year, processing maintenance grants and loans for benefit claimants in emergencies, and supporting the jobs and careers services to get people into work. And we’re excited to see who we can help next.
About the author
Andy Borthwick is a Solution Architect here at Version 1.
