By Peter Ellis

In a million ways our lives are being reshaped by new uses of data. From the risk-analysis calculator our banks use to decide whether to give us a loan to the optimised route our satellite navigation system suggests for our drive home to the diagnostic tool our doctor uses to review an x-ray, all sorts of activities are being influenced by the accumulation and analysis of terabytes of data.

Three things are driving the change: a revolution in statistical methods that started late last century and is still ongoing; cheap and powerful computing power; and new data sources generated as by-products of everyday activities.

The age of big data relies on the rapid growth of computing and communications horsepower – some 90 per cent of the world’s data has been generated in just the previous two years, according to one study.[1] But it would not be possible without the massive databases of information compiled by government and private organisations, and without the willingness of those organisations to act on the insights that emerge from analysis of those databases. Both of those require the organisation’s staff to have a high level of data skills.

The cumulative effect of leading-edge organisations intelligently exploiting data is that consumers expect the organisations they deal with to use data-driven insights to improve the user experience. By next year, there are estimated to be 50 billion devices collecting data as part of the Internet of Things.[2] Users expect the services that result to be highly personalised, lightning fast and available anywhere at any time on any device.

For organisations that do not think of themselves as technology organisations, this heightened expectation can be intimidating. Many executives find complex decisions about the use of technology to be at the outer edge of their comfort zones and are spooked by nightmarish stories of technology projects that made big claims about what they would achieve but left only a legacy of big costs and big headaches.

Exploiting big data opportunities requires the right skills

Several organisations whose primary capability is not technology have made innovative use of data to improve their performance.

For example, the Australian Taxation Office’s data-matching program electronically compiles, validates and analyses information from third-party sources. Its data-matching efforts are used to create pre-filled tax returns, thereby making it quicker and easier for taxpayers to lodge their return, and for the tax office to identify possible incorrect claims. The productivity and performance benefits of the ATO’s use of data are immense.[3]

But identifying and using opportunities like this requires careful preparation. The technology alone is not enough. Instead it relies on organisations understanding the data capabilities of their people and taking a strategic approach to developing those skills. Without data-skilled people – across all levels in all parts of the organisation – opportunities will be lost.

Recently NSW Health launched the Health System Performance App, which enables hospitals and other local health organisations to analyse performance against key performance indicators.[4] While the app obviously deploys data technology, crucial to its success is that users understand how to interpret and interrogate the data. Without people able to understand the information, the data risks being a blur of numbers and graphs, with no clear course of action resulting from it.

Data is more than a numbers game

When many people think about data literacy, their minds go to computer code, graphs and complex economic modelling. It is true these are important for getting the most from data. At Nous, we even have an “Analytics Team Maturity Model” we use to help specialist areas work through the challenges of IT platforms, source code version control, statistical skills and data governance.

But number-crunching is only one aspect of working with data. And just like war and generals, data is too important to leave to the data scientists. Complementing the specialist skills expected of statisticians and data professionals, data literacy for organisations also involves:

• sourcing the data or evidence most relevant to your work. There is a wealth of material from which an organisation can draw evidence, but much of it gets overlooked. A data-literate organisation needs to be comfortable identifying the internal data sources most appropriate for its work, as well as recognising external data sources. These could include academic studies, official government sources, international databases and private-sector sources, not all of which are necessarily easy to access. Sourcing the right data may also involve a conversation with other people in your organisation who hold more knowledge of what is available and how to access it.
• understanding the context surrounding data. This can be as simple as ensuring people have access to training that builds understanding of the organisation and the relevant market, sector and policies. Organisations that are aware of the broader context surrounding data make better-informed decisions about what data to use and what the data is telling them. Knowing the context helps people more efficiently analyse information and form intelligent insights.
• knowing how to assess the strengths and limitations of data sources. Not all data is 100 per cent impartial, and people can too often accept information without questioning its strengths or limitations. When an organisation can identify the strengths and limitations of data, they can assess how reliable the source is and decide whether and how to use the data. Without this skill, organisations risk using the wrong data in the wrong ways and making statements or decisions based on limited information.
• asking intelligent questions to interpret and make sense of information. Sometimes data is presented in an unclear way and there can be discrepancies between sources of data. In these instances, it is essential that organisations have the skills to ask the right questions to make sense of the information presented. A good understanding of the context and the strengths and limitations of data can assist in asking the right questions.
• communicating insights based on information. The key purpose of using data is to assist organisations to better understand the work they are doing (or should be doing) and the impact they are having. Robust evidence helps organisations make decisions and enhance their performance. To use data in this way, organisations need the skills of interpreting and communicating insights from data. The insights must be communicated in a way that has meaning for audiences – whether they be internal staff, board members or clients. Often this means concise and engaging, using a variety of media.

Essentially, data literacy is the ability to contextualise, critically appraise and communicate insights from data and information to support an organisation to achieve its purpose.

There are five steps you can take to improve the data skills of your staff

Every organisation can benefit from a capability framework to improve their data skills. From Nous Group’s work supporting organisations to make real progress in this regard, we have identified five fundamental steps. There are some powerful tools that can help to put these steps into practice.

1. Know your starting base. Your staff may experience certain challenges using data in their daily work. You need to ask yourself and staff: Are some staff not confident in their abilities? Are insights often generated without using the right information or data? Does the organisation rely on one person or one team for data-related issues or work? Are staff confused about where to find data, how to collect it or how to interrogate it? What do staff think of when they hear the words ‘data literacy’? Do staff want to increase their data literacy? Approaches to asking these questions include a team workshop, focus groups, interviews and a staff survey.
2. Define what capabilities you want to build. The capabilities must be tailored to meet the role requirements in your organisation. For example, the Australian Public Service Commission has grouped capabilities in five areas: Using data in the APS, Undertaking research, Using statistics, Visualising information and Providing evidence to decision makers. Organisations could define their capabilities by documenting on sticky labels all the ways they use data (including areas they would like to develop) and then grouping each label into a broader domain. These domains can be defined chronologically (accessing data, analysing data, communicating data) or in other ways. You could also review good-practice data literacy capability frameworks to identify features that might apply.
3. What does good capability look like across the organisation? Different roles require strengths in different areas. Not everyone needs to be excellent at everything. Instead, the organisation as a whole needs access to data skills from among its people. It is important to create avenues for staff to assess their capability against what is expected in their role. Once you understand the capability domains you want to build and the capability levels suitable for different roles, this can be mapped onto a framework so the organisation has a practical tool to aid understanding and build skills. All staff should be able to use the framework, no matter their data skills starting point.
4. Develop resources and processes to integrate capability-building across the organisation. Once staff can assess their capability against what is expected, they need the tools, resources and support to build and develop their individual capabilities, as well as support others to develop. Staff can be supported to assess their own capability through a survey that assesses an individual’s perceived level of capability. The results could map an individual’s perceived capability against capability levels expected for their role and seniority. The results can be used as input for coaching or performance conversations, to understand team capability and to assess ongoing development. The survey could be rerun periodically to assess progress.
5. Develop an implementation plan. Without a plan for applying the capability framework, it will be ineffective. As well as detailing key responsibilities and timing, the plan should consider how the framework can be communicated across the organisations, how it can be embedded into existing structures and processes, the key resources required to keep it meaningful to staff learning and how its successes will be measured.

Broad involvement can help to achieve buy-in

It is understandable that some staff may take a sceptical approach to an effort to achieve a step-change in data capability. For people who feel use of data may challenge their judgements made on instinct, or who find numbers intimidating and would prefer to leave them to specialists, it can be difficult to accept the greater prominence of information-based analysis.

To achieve buy-in it can be effective to involve staff at all levels and in all roles, so the effort is rightly seen as organisation-wide rather than targeting particular individuals. Broadening the pool of people benefiting from the data upskilling can help to achieve shared responsibility for learning and development. Ultimately, it can also deliver the shift in culture and expectations that are required to embed these new skills in the organisation. ‘Change champions’ and training tools can help to promote and facilitate organisational changes.

So, what would an organisation with a high level of data skills among staff look like? Clearly it will vary a lot according to context, but we can expect that decisions are made based on thorough analysis of data, that new products and services are tailored to the individual needs of users and that emerging trends are spotted early, allowing the organisation to get ahead of the curve.

Put together, these can give an organisation an edge that can set them apart from their competitors.

Get in touch to discuss how we can work with your organisation to build data skills.

Written by Peter Ellis during his time as a Principal at Nous.

[1] “Exciting facts and findings about Big Data you should know”, Big Data Made Simple

[2] “The Internet of Things”, Cisco

[3] “Data matching”, Australian Taxation Office

[4] “Health System Performance (HSP) app”, NSW Health