There are many businesses that are not yet fully exploiting the capabilities in data and analytics that are available to them: as a result, they are being left behind.
The ‘next big thing’ in tech and in data is always just around the corner – but businesses should concentrate their efforts on solving the challenges that face them. Whether they happen to exploit basic analytics, advanced AI capabilities, or true data-native capabilities such as the data mesh, the focus needs to be on adding capabilities where they deliver value, rather than adding capabilities simply for the sake of it.
The challenges faced by data laggers
Established companies face legacy obstacles that more agile and tech-driven startups and younger businesses might not. Here are a number of reasons why certain enterprises will be falling behind others in the world of data and analytics:
1) Organisational structure
Businesses need to be good at tech, but also have the organisational structure to support it and exploit it. Startups are often set up in this way from day one – Facebook’s IT capabilities, for example, are spread and shared across departments in a horizontal fashion. Technology is in every function of the business.
By contrast, older or waterfall businesses are verticalised and are likely to have siloed IT teams. Whilst tech teams need deep, technical skill sets, they also need to be able to work closely with others – so that they can ensure the tools and solutions they are creating also match up with the overall needs of the business.
2) Lack of data literacy
In order for IT teams to be able to work alongside and engage with teams in the wider company, a general level of good data literacy across the company is critical.
This also includes at the board level – do decision-makers and stakeholders understand the requirements and business benefits of data analytics? For many traditional organisations, even building basic data literacy is a significant journey to embark upon.
3) Migrating from legacy tech
Legacy technology has layers and layers of functional business requirements built on top – a legacy database such as Oracle or Teradata might have thousands of scripts that depend upon it.
Whilst migrating the tech and the data is reasonably straightforward – although labour intensive – the real challenge is ensuring backwards compatibility of the scripts, and re-training the teams who generated the scripts in the first place.
Businesses tend to work with older databases on a licensed basis – a cycle of say three to five years. Moving from legacy tools to a modern tech stack requires very careful timing: where in the license cycle should businesses look to make their move.
Teams need to be confident that they can migrate everything out of the existing stack into a new one before the license to the legacy database expires. If a business decides to shut down a data warehouse, they need to be sure that everyone using it can still do their job within a new data warehouse environment.
These challenges make it harder for organisations to move on from legacy tech – and thus hold them back from fully making the most of their data, and the analytical capabilities available.
Data exploitation is key to business
Plenty of studies show that organisations capable of exploiting data grow faster, acquire customers more cheaply, and retain customers better.
At Ciklum, we regularly hear from clients that are looking to utilise data analytics in order to offer personalised experiences or product recommendations for their customers. Personalisation offers enormous business potential.
Data can also help to optimise cost – at Ciklum, we’ve been able to substantially reduce our clients’ infrastructure costs by as much as 30%, providing them with more scalable and functional platforms that replace complicated and monolithic systems.
Data can often be the biggest differentiator between industry disruptors and established businesses – and whilst the former may have the ability to better exploit data, the latter is likely to have a greater depth of industry-specific data. As a result, data can give established businesses the much-needed competitive edge against startups, who won’t have the same scale of data available to them. The ability to exploit existing and historical data is critical to the survival of businesses that are being disrupted in their sectors.
Start small, think product, not project
Businesses looking to utilise data and analytics capabilities for the first time should start small and experiment; setting up an isolated team that, if possible, are unhindered by the governance and challenges of the wider organisation. The focus for any experimental data project should be on medium value, unpolitical use cases with clear success metrics – such as product recommendation and demand forecasting.
These use cases may be appropriate for the simple onboarding and testing of new, analytics capabilities, whilst also rapidly delivering value to the wider business. As this value is delivered, it becomes easier to evangelise about how to, and why, exploit data to the wider business.
As a result, some (or much) of the foundational work will be done – helping to reduce cost barriers to entry and create enthusiasm amongst stakeholders.
Data and analytics solutions are constantly active and require ongoing development – rather than being projects to be completed and left alone, they must be regularly reviewed in order to best serve the end-user and the business.
Shifting from this project mentality to a product mentality is key if businesses want to fully compete in the market. Once a project is delivered, it can immediately start occurring technical debt – which in future might make even small, iterative changes to a system a massive replatforming challenge. By moving to a product mentality, in which teams always work on a product and improve it, it becomes easier to continually deliver incrementally better value – and execute major change projects if required.
Help data culture to spread across the organisation
Once some value has been proven with a small experimental project, the focus must then be on removing any barriers so that the value can go live, and really begin delivering to the business. As your experimental team starts to deliver on a number of solutions, businesses can then go a step further and build platforms to sustain and scale the value – through MLOps, ML Delivery platforms or cloud modern data platforms.
Proving value makes conversations with the business stakeholders about the virtues of data far easier.
A great way to extoll the business benefits of data is via regular outreach sessions or show and tells – make sure as many people are invited to the table as possible. By showing the inroads your experimental team has made, other areas of the business will also want to adopt these capabilities into their own teams.
With value, enthusiasm, and opportunity, the data culture will spread in the organisation, and as it does, so too will the time and opportunity to build and deliver new operational platforms and capabilities grow.
Discover how Ciklum can help your business to better exploit data.