Top considerations for implementing a data and analytics capability in your company

Data has emerged to become a major business asset. Nowadays, companies across almost every industry and sector are leveraging data to understand customers better, create value at every interaction and to gain a sustainable competitive advantage. This is achieved by deploying data and analytics to translate complex data into meaningful insights that enable more purposeful customer engagements and business decision making processes.

To unlock the value that data provides; organisations must ensure that they focus on getting certain elements right to enable a successful and scalable implementation of a data analytics capability. Lasath Punyadeera – Head, Personalisation & Profit science, Consumer and High Net Worth Markets, Standard Bank provides the following considerations for companies.

1. Getting data basics right –Democratised access to data, secure data storage, compliant data environments and stable data platforms are fundamental aspects in setting up a data environment, that can support and grow analytical capabilities. If getting the right data is a struggle or if analytical capabilities are starved of data, there is likely to be less facts and more fiction informing the data-driven decisions.
2. Technology and software setup – The choice of technology and software remains a key consideration when setting up a data and analytics capability. It is imperative to always build data and technology stacks with scale and future requirements in mind. An important point to note is that allowing full flexibility on the choice of technology, software and programming languages can translate into significant integration complexities down the line. Therefore, organisations need to make the right choices up front - consider the use of interoperable technologies, set up coding standards and govern the choice of programming languages.
3. Organisational readiness and buy-in – Organisational readiness, buy-in, and very importantly, sponsorship remains key to setting up a successful data and analytics capability. This is usually the first and most important step to get right, for any data and analytics capability, to attract investment, grow, scale, and survive in an organisation.
4. Data health - Data quality, availability, timeliness, completeness, consistency, accuracy, and reliability remain key inputs in designing the right data environment to complement analytical capabilities. Early focus on getting the health of data right will go a long way to accelerate intended analytical solutions over time. This will further avoid wasting specialist analytical skills on elementary data cleaning exercises.
5. Building a true data-driven culture – The biggest hurdle to building a data culture is less technical and more cultural in nature. It’s not the lack of data specialists, software or technology that derails transitions, but the mindset of people in the system, at times inadvertently working against the system. From the leadership that sponsor, embrace, and encourage, to the managers that can comprehend, appreciate and value the language of data, to the data teams that are passionate, competent, and creative in producing data and analytics. All these gears need to engage smoothly for a data-driven culture to really take shape and flourish. This requires both a top-down and a bottom-up approach converging in unison to make the data culture the norm rather than the exception. Once a true data culture has been built, the organisation needs to avoid the temptation of second-guessing data-driven decisions – if this dominates in an organisation, the data driven culture is likely to have a slow death overtime.
6. Improve data literacy across the organisation – Work needs to be done, continuously, to improve data literacy across the organisation. This does not mean that everyone needs to be a data specialist, however, the basics of data and the value of data in solving business problems needs to be understood. This will in turn help people realise the value of data and will naturally generate demand for the right data from people closest to the problem. The ideal is always to showcase the inherent value of data in the decision-making processes and establish a “pull” adoption from people versus a “push” adoption to the people.
7. Creating purpose and aligning to organisational goals - The purpose of data and analytics needs to be intricately aligned to problems to be solved and organisational goals. In other words, the workbook of data and analytics capabilities need to be informed by the user’s problems, strategic goals, and performance outcomes of the organisation. This will further ensure that the analytical teams remain focused on things that matter, the problems that were asked to be solved versus flashy things that no one asked for. Furthermore, the data and analytics teams need to understand the bigger picture and the way in which their contribution translates into the organisation achieving its strategic imperatives and delivering on commercial performance.
8. Integrated ways of work – Analytical capabilities are often ring-fenced and centralised as a centre of excellence for various reasons. However, this centralisation should not translate into the unintended consequence of creating both physical and psychological barriers for integration. Analytics can neither deliver value nor survive if it operates in isolation of the business or end users. It is important to accept the reality that people closest to the problem have a better understanding of the problems they need help solving. Therefore, deliberate steps need to be taken to create permeable boundaries between business and analytics, complemented with fluidity to allow analytical teams to rotate in business and business teams to understand analytics.
9. Analytical skills and experience – Depending on the type of analytical requirement, the right skills which are fit for purpose need to be mapped-out, upfront. As a first step, this necessitates a skills assessment to understand the current skillset and potential gaps relative to the problems to be solved. Then, work towards getting the required skillset blended with new hires from the market/outside the organisation as well as current teams upskilled for new job requirements. It is important to give equal opportunities to individuals in the organisation - parachuting individuals without giving opportunities to existing teams will undoubtedly destroy the team morale – this can have disastrous outcomes from the onset. Whilst it is important to plan and cater for future analytical expansions, it does not mean that all such skillsets need to be sourced upfront. It is prudent to think big but start small and then scale up based on business demands and proof of value.
10. Managers who speak the data language and drive adoption – Strong analytical teams can fail miserably in the wrong management hands. These management failures are not as obvious, as the blame may be transferred directly to the team and hence organisations need to pay close attention to these dynamics. It is important to note that managers of data teams need to speak and relate to the language of data, comprehend data and analytical methods, and be able to challenge and guide data and analytics teams. This does not mean that every manager of a data and analytics team needs to be a qualified data scientist, however, this does mean that mangers of data and analytics teams need to have a good working knowledge on the subject. The managers need to have a passion for driving adoption of data and analytics to ensure that their teams are working on meaningful use cases, translating business problems into the language of data and the data into actionable insights and tangible benefits. A manager with a strong bias towards data and execution will be able to stretch the teams thinking, surface and elevate the contribution of the team and prove the value of data-driven decisioning.
11. Staying true to the business case – Data and analytical capabilities are established to solve business problems with the expectation of beyond business-as-usual benefits on both customer and commercial objectives. Therefore, analytical capabilities inherently have a high commercial purpose for its very existence. To this extent, it is important to have commercialisation built-in by design and intent and should never be attained by accident or as an afterthought. This is a key step to staying true to the initial business case, gaining credibility from the organisation, sustaining current analytical capabilities, and in securing additional funds for future expansions.
12. Accurate measurement and value attribution - A robust measurement framework, agreed and approved by all relevant stakeholders, that can accurately measure, and attribute value delivered through analytics remains imperative to showcase the true value of data and analytics. This will further enable organisations to truly understand the real return-on-investment of the analytical capabilities, which remains a key input into future investment decisions
13. Experimentation and innovation – Always create space for curiosity and direct this into innovative analytical solutions for solving real business problems. Analytics can surface many ideas; however, these need to be tested through scientific experimentation and proof-of-concepts to identify practical solutions which are scalable.
14. Training, upskilling, and networking – Analytical methods, software and technologies are continuously evolving. Therefore, to maintain analytical capabilities relevant, up-to-date and fit-for-purpose, continuous training and upskilling is required. This remains a critical step in building and maintaining an analytical capability that can survive the test of time. It is further a best practice to upskill on new analytical methods and technology as close as possible to the time of actual application — this will ensure that the fresh learnings and skills can be put to practice on time before the learnings become stale. It is also important to allow space for analytical teams to network with industry experts and to participate in industry forums, competitions, and challenges. This will ensure that the teams are always at the forefront of advancements in the analytical domain.