Good governance is critical to business success. However, organisations are often let down by their business practices, and a key culprit is poor (or no) data and analytics (D&A) governance.

Organisations that are successful in D&A governance have a data-driven culture. They actively engage and influence their stakeholders, promote data sharing, break down silos and evangelise data literacy. Gartner’s Seventh Annual CDO Survey found that chief data officers (CDOs) who link D&A to prioritised and quantified business outcomes and metrics are more successful than their peers. There are five key steps to achieve this.

1. Identify unachievable business outcomes due to poor D&A governance

First, present business leaders with a problem statement that they can recognise and own. The organisational business strategy should be used as a primary source to understand the goals and direction the organisation expects to take, the market drivers and the regulatory landscape. Following this, engage with key business leaders to find out where D&A is working well, and where it is not, and focus on the business implications of poor D&A governance.

Next, obtain internal audit reports that show the audit observations, audit points and risks that have been raised in the enterprise and assess these in relation to the governance of D&A – within and across business areas – and their cost.

Before proceeding, explore high-level findings to key business leaders and ensure the problem statement is expressed in the right business language and business priorities, and valid assumptions are correctly understood.

2. Connect business performance with information value through metrics

The causal relationship between poor data and analytics and poor business performance must be highlighted if a compelling business case for governance is to be made.

Initially, look to identify the business processes and process owners that are critical in addressing the problem statement. These will often span multiple business areas, so look to focus on key processes rather than on lines of business. This will help break down the silos that have led to the insular and disconnected governance of data and analytics.

Determine the most impactful key performance indicators (KPIs) and key risk indicators (KRIs) for business success, and then identify the specific data and analytics assets that are used in the KPIs and KRIs. These assets are the ones that must fall within the scope of the data and analytics governance proposal.

3. Outline the scope and footprint for governance through people, process and technology

A key characteristic of highly successful D&A governance initiatives is their ability to effectively define and manage scope. Be clear on what is in scope and what is out of scope for governance while identifying the key stakeholders needed in the D&A governance steering group. Also highlight the need for a governance footprint within the business areas, so that decision-making and decision-execution are properly aligned through roles such as the information steward.

Look to then create and evaluate all potential models for delivering key business outcomes for D&A governance, analysing the benefits, costs, risks and assumptions for each of these options. In addition, use knowledge and experience of how the organisation works to select the best option.

Finally, assess the likelihood of success of the proposed governance solution by testing it with the key business stakeholders who will require to share their initial insights and final approval on the proposal.

>See also: Three pillars for scaling intelligent automation: process, technology, people

4. Define the approach, deliverables, time scales and outcomes

To define the approach, use the statement of governance scope to segment the delivery of business outcomes into manageable phases. For each of these planned deployments, identify the business value that will be delivered, to whom it will be delivered and the specific business impact on the organisation.

Select the adaptive D&A governance approach as the governance deployment model, recognising that the application landscape is complex and contains multiple platforms. Identify the deliverables produced through the phases of governance and their specific contribution to business value improvement. Connect these deliverables to data-driven business decisions, improvement in organisational behaviour and the ability to drive new business value.

5. Complete the financials for the proposal

Ultimately, a clear set of financials that numerically demonstrate the value of the governance proposal and the financial implications of not proceeding are crucial – and this will take time and effort. Look to assess the governance proposal in terms of the total cost of ownership (TCO) and develop the return on investment (ROI) model based on a spread of possible outcomes.

Before presenting the final business case to senior business stakeholders, review the storyline, logic and evidence for the governance business case. Include a “do nothing” as one scenario and project the financial implications of this option. Engage supportive business stakeholders to help find the holes in the proposal before it is presented to the formal decision-making board and be prepared to update the business case based on actuals, new data and new issues raised by the review team.

Saul Judah is vice-president analyst at Gartner.

Related:

How to drive business value from edge analytics — With data exponentially outgrowing the capabilities of centralised storage and management, here’s how edge analytics can help organisations overcome this challenge.

IoT governance: how to deal with the compliance and security challenges — As the Internet of Things becomes more prevalent across an organisation’s network, the question of effective IoT governance becomes increasingly relevant.