Whither data and data analytics in 2019?

• Focus will shift from simply putting data in one place, to having one view of the data
• Data literacy will become a key performance indicator for the enterprise

POWER was once tied to land ownership – but that has changed dramatically these past two decades, as it now resides in information. We see how this may lead to misuse as companies can feed off their huge hyperscale data centres and participate in the data and AI arms race, affecting industry after industry.

As such, the question on whether data and analytics will be centralised or de-centralised may represent one of the most decisive issues for how society evolves in the 21st century.

However, organisations need to get on top of the trends that are emerging to understand what can be done with data and analytics before they can answer that question. Here are four key ones to expect in 2019:

Data literacy will become a key performance indicator (KPI) for the enterprise

While everyone understands that reading, working with, analysing and arguing with data is important, it is hard to know where to start.  Some will need to learn the basics, while a data scientist requires a gamut of challenges to overcome boredom.

The classic adage says that “you can’t manage what you can’t measure.” To improve data literacy in organisations, there first needs to be a diagnosis to determine where the company is on a scale.

New methods of measuring and indexing data literacy are emerging. Using these, people and organisations can go about raising skills in a more precise and targeted way.

What is even more interesting is that a corporate data literacy score can now be derived. This is particularly exciting because early data seems to indicate that there is a correlation between an organisation’s data literacy, and an enterprise’s value of corporate performance. These include gross margin, return-on-assets, return-on-equity and return-on-sales. Will that correlation be the watershed moment for making data literacy a mainstream imperative?

Data literacy is all about raising skills from the bottom up, and data literacy becoming a KPI can help a CDO or other executive steer corporate performance from the top-down, as a strategic and differentiating initiative. In the future, having a high data-literacy score will likely become a criterion for hiring talent.

Performance and scalability will take centre-stage in enterprise selection criteria

Performance is undervalued when it comes to tool selection – and too often an afterthought. Where query performance is good, and latency is low, is where analytic workloads run.

If a query takes longer than a few seconds, users lose interest and stop interacting with the data. If it takes more than a few hundred milliseconds, users may not leverage it in a business process or an augmented reality experience.

In organisations, as the self-service trend was in its nascence, perhaps performance was overlooked by many because building visualisations on a flat file does not take that much horsepower.

However, many self-service BI solutions (often referred to as “modern BI”), that seem so cheap up-front fail when it comes time to scaling more data, workloads and people across the enterprise.

Breakthroughs have only recently been achieved through indexing, caching and pre-preparing very large and distributed datasets. Now, as companies of all sizes are increasing their adoption of hyperscale data centres, performance will rise in the selection criterion.

Some organisations have moved their data back through “re-patriation” because they have not seen strong enough performance.

This becomes even more important in an IoT application. More and more workloads will run locally or at the edge to avoid latency. In short, efficient performance will be a deciding factor for how architectures will look – centralised or distributed.

There will be a convergence between visual, conversational and presentation technologies, facilitating persuasive storytelling

There is no such thing as unbiased decision making. Emotion is considerably more powerful than logic when making decisions.

Most people perceive data to be boring, but love stories backed by data. What can be done to introduce a higher proportion of logic into the mix, i.e. letting the data “speak”?

The emergence of machine-driven data storytelling has offered narrations through natural language generation. Conversational analytics will make this approach much more interactive and accepted.

These two approaches will need to be augmented by even more user-friendly ways of telling data stories, where visual artefacts can augment findings.

As a result, data storytelling and presentation technologies will gradually merge. This will be the next set of converged technologies that will be persuasive in the boardroom.

It will also greatly help with the broader movement around data literacy, by explaining and expressing data and analytics so that a wider business audience understands it. This also means that the data analyst and graphic artist role will increasingly overlap.

It will be more important to have a single view of all data 

In 2019, focus will shift from simply putting data in one place, to having one view of the data. The problem is that historically, it has taken a lot of effort to make that happen.

Cumbersome efforts to put all the data in one place failed to achieve the goal, and this is now happening again in the cloud.

It is a seemingly impossible feat because there will always be new data coming in and being able to combine as well as analyse data at the source is what enables the crucial agility needed in a fast-moving world.

This will now become possible thanks to the different vendors coming together to standardise data models. There is also an emergence of enterprise data catalogues, which makes it possible to audit the entire distributed data estate – delivering a shop-for-data marketplace experience.

The more users share, collaborate and use the hub, the more valuable it becomes to the business.

Ultimately…

The responsibility lies on everyone to put regulations in place that benefit the many, to keep data secure, private and de-centralised.

Also, if information is to be democratised, it hinges on an enlightened constituency and exemplary leadership. Above all, performant technology infrastructure, agnostic of one data centre, that thrives on networked distribution is key.

Dan Sommer is the senior director of Global Market Intelligence Lead at Qlik.