Cracking the code
We ask four experts: “How can CIOs best implement and execute a low-code/no-code strategy within their organisations?”
“Fusion teams are critical”
Victor Kuppers, vice president of strategy, Betty Blocks
By radically changing their approach to software development to include no-code/low-code (LCNC) development platforms, more and more organisations are becoming agile, improving competitive value and increasing employee engagement. Their newest secret is fusion development teams; teams that empower digitally-skilled and solutions-oriented employees to work side-by-side with the IT department to build apps using LCNC platforms.
A fusion team brings together people with diverse professional backgrounds who use data and technology to achieve shared business outcomes. Ideally, a fusion team combines professional developers with citizen developers. A citizen developer is a business person without coding experience who builds apps using a LCNC platform.
The citizen developers in these fusion teams are genuine problem solvers and easily recognised in the organisation. They are often found managing and analysing datasets, automating tasks, developing websites and much more. Trouble is, if you do not provide these innovative colleagues with the right tools, guidance and governance, you risk them creating shadow IT projects, ungoverned by the professional eyes of IT.
A fusion team is about empowering problem-owners to become problem-solvers without relying on IT, but having access to IT’s guidance and support.
A citizen development mindset will result in a new way of working. This change can ultimately benefit everyone in the organisation, as well as clients and external stakeholders. However, before that happens, your fusion team should learn key principles for good governance. Besides, you want to provide them with a platform in which they can safely develop applications, without the IT manager lying awake at night.
And that’s where LCNC platforms can really help. If you choose the right platform, with governance features and support, you have the perfect tool to help a fusion team achieve maximum returns. Solutions can be developed under the guidance of the IT department in a way that’s also compliant with the security standards of the business and which easily integrates into the company’s IT environment.
With fusion teams, what started as an innovative idea has become a pivotal moment for business-led tech. The use of LCNC combined with governance is an opportunity to make your workforce more productive, resilient, happier and better equipped for digital transformation.
“CIOs must understand teams and goals”
Florian Douetteau, co-founder and CEO of data science platform provider Dataiku
The rise of low- and no-code AI platforms isn’t just the proliferation of a new set of tools or solutions: it’s a new business model that will see business users find their own ways to add even more value with AI than data scientists have been able to add on their own. While data scientists will always be critical, a low- or no-code strategy represents a fundamental shift in mindset around data tooling — one where we’ll continue to see a bigger breadth of people access data and work with it on a day-to-day basis.
In order to see the immediate benefits out of a low- or no-code strategy, CIOs need to do their operational due diligence. There are any number of low- or no-code options available, but it’s essential to vet and evaluate them to determine whether they meet organisational, cost, compliance, and functionality requirements.
In implementing low- and no-code tools, CIOs also need to understand their users and resources upfront. They will need to dissect each part of the data journey to translate it into concrete organisational resources and needs, mapping those needs to the different data roles and profiles available.
A successful low- and no-code strategy is also about selecting the right internal business use cases. This means looking at the existing pain points that affect IT, tech, and business users to see how low- and no-code can best benefit each group. To start, businesses need to get data out of silos into a unified, analytics-ready environment. This is an essential step in using low- and no-code to scale AI deployment in a traceable, transparent, and collaborative manner.
Part of your strategy should also be identifying the tasks that are data science resource-heavy, and where a low- and no-code approach can help. Smart data ingestion, processing dates and times, clearing complex text fields, combining datasets — even creating new machine learning models — these are all examples of tasks that can be done code-free on many platforms. Users may also be able to venture into code-free data pipelining, data preparation, and model training in order to scale out models in production with complete transparency.
Empowering business and other non-data science roles with low- and no-code visual tools is one part of the opportunity: however, your data scientists may also be able to leverage low- and no-code solutions to operationalise more models quickly versus a code-only approach.
“Focus on time saved”
Ian Funnell, Head of Developer Relations, Matillion
It’s amazing how quickly new methodologies can gain traction in the technology industry. The use of low-code/no-code is a perfect example. For developers no longer content with the thankless tasks of data management, these tools have transformed their role and helped them drive innovation in a short space of time.
It’s easy to see why, too. Low-code and no-code solutions free up developers’ time in two crucial ways. First, they lessen the technical barriers to entry for non-IT workers across an organisation wanting to work with data, allowing data teams to spend more time on strategic data and technical initiatives that create impact and a competitive edge. Second, they turbocharge IT productivity by automating tedious development tasks.
But it’s too simplistic for CIOs to simply see the adoption of low-code and no-code tools as a binary decision to automate a portion of developers’ workloads. Yes, they will democratise software development, enable the rapid development of applications, automate some aspects of data integration and better support data visualizations. Those things are, of course, on every CIO’s wish list.
The key to transformation, though, is how CIOs direct developers to invest the time savings accrued by the introduction of automation. There’s a massive disconnect in organisations about how to approach analytics, with many stuck in a process-focused mentality. CIOs must push developers to dedicate greater resources to making analytics more useful, and focus on the collaborative aspects of data processing. Low-code/no-code tools clear the path for developers to put meaningful ideas into production in a more agile way. Developers become empowered to pursue projects that will further the goals of the organisation. By doing so, they become more fulfilled in their careers and want to stay; a worthwhile goal given the crazy demand for talent right now.
Of course, some tasks are easy enough for a machine to do: if they are, a machine should do them. In turn, we can give developers back the bandwidth for tasks that do require expertise, thought, and a human touch. Then we will be leveraging a data team’s full worth and value.
Low-code and no-code tools are undoubtedly here to stay, but are not a replacement for innovation: They are a catalyst. The skills and output of developers are incredibly precious. Without them, businesses can’t reap the truly valuable insights from their data, the key to ultimately differentiate and compete. So above all, when implementing a low-code/no-code strategy, the first priority on your list, as always in business, should be people.
“Data pipelines must be addressed”
Nathaniel Spohn, VP EMEA, Fivetran
As companies grow, they constantly integrate new platforms and tools to uncover fresh operational insights, fuel growth and act on their customers’ wants and needs. But investing in new technologies without the right data governance and infrastructure in place often creates complexities at the expense of decision-making. Most data engineers are playing a losing game against time as they build and rebuild underlying data pipelines when they inevitably break.
This is where the combination of no-code tools and data pipeline automation can be a game-changer for CIOs. Tools like Airtable, SendGrid and Stripe all offer no-code options to help businesses collaborate faster on spreadsheet data, create email marketing campaigns and monetise products and services without the overhead. But the advantages these tools bring in terms of speed and insight can be quickly diminished if the valuable data generated is then not made accessible to the wider business.
Decision-makers need comprehensive, accurate and up-to-date information from all areas of the business to understand where to prioritise investment and where customer experiences break down. What’s more, this data can help them get a real sense of their maturity and potential for implementing differentiating technologies such as machine learning. This is the next hurdle for data engineers.
Our research shows nearly all data engineers face challenges when building data pipelines. Every time a business integrates a new tool, it can take in-house data engineers over a week to write a bespoke code for the new data source (known as a connector), and when the flow of data is interrupted — for example due to changes in the source application — it takes a full day’s work to patch up the code. Shockingly, half of data engineers waste their time on manual data pipeline repairs on a monthly basis.
Data pipeline automation takes the coding and delay out of this process. Data engineers can deploy secure, pre-built connectors that automatically adapt when changes are made at the source and make it easy to plug clean and fresh data into Business Intelligence systems. For example, using these connectors for Stripe, data from one-off campaigns can be seen in the context of all other purchases and subscriptions as well as wider customer data. Ultimately, this not only alleviates the burden on data engineers — freeing them up to do more value-added work — but also empowers non-engineers across the business to drive growth and make better decisions based on data.
