WIXLIB

10 design principles to boostdata governance adoptionand success

10 design principles to boost data governance adoption and successIn recent years, the volume of data used within organisationshas increased dramatically. However, ungoverned data is messy,lacks rules and inhibits productivity. Hence, being able to properlymanage these growing volumes of data is rapidly becoming vitalto harness data for business value. This increased recognitionis driving governments and companies of all sizes to invest indata and its governance. Companies that are not will be or aresignificantly putting their value creation at risk. Think of it aspurchasing a fancy new car: would you buy one without knowinghow to drive it?The essence of data governance revolves around specifying across-functional framework for managing data as a strategicenterprise asset. In doing so, data governance specifies decisionrights, accountabilities and processes related to data assets withthe objective of ensuring the quality, consistency, usability, security,privacy, and availability of the data. Good data governance helps toensure that those who need to use data can find it, understand itand trust it, which fuels data-driven decision-making acrossyour organisation.Today, there is a widespread recognition that in order tosuccessfully run a data-driven organisation it is vital to: Establish standards, policies, and procedures for the usage,development, and management of data Design the right organizational structure Implement the technology infrastructure to supportdata governanceYet, the successful adoption of data governance remains aproblem for most companies. Many organisations have conducteda proof of concept or successfully implemented the plethora offunctionalities provided by data governance platforms. Proofof concepts manage to eke out small financial gains of theorganisation and excitement often grows as functionalities arebeing materialised. But then months or years pass without bringingthe big wins organisations expected.2Companies substantially struggle to scale from proving thefunctionalities of data governance tools to companywide leverage.These challenges are prominent across industries and irrespectiveof the tools, systems or models used. Why are many datagovernance programmes running out of steam, losing funding, andultimately falling by the wayside?Adoption as the number one priorityThe right technology can ease the path, but any tool is only asgood as what is put into it and the extent to which it is effectivelyused throughout the organisation. In order for data governanceinvestments to produce business value, people have to actuallyenter, update and use the data definitions, business rules, and KPIs.The governance process needs to be a complete feedback loop inwhich data is defined, monitored, acted upon, and changedwhen appropriate.Having worked with multiple companies across the globe, ourexperience shows that slow progress—with few exceptions—reflects a failure to rewire deeply-rooted and well-establishedhabits throughout the organisation. Indeed, we’ve seen that datagovernance programmes, across clients and industries, faceformidable cultural barriers. Ignoring this reality or simply putting itlow on your checklist wouldn’t be a wise course of action.As data governance won’t become less relevant any time soon, it’stime we start thinking about how to approach it differently, withadoption as the number one priority.How can we get data governance programs out of their‘comatose state’ and bring life back to its adoption?In this article, we discuss 10 design principles to maximise properadoption of your data governance programme. These principleshave been inspired through our joint work with clients worldwide.In all of this, there is one absolute truth: automatic adoption of datagovernance programmes doesn’t exist (yet).

10 design principles to boost data governance adoption and success10 design principles to boost data governance adoption and value generation1Budget as much for culture and adoption as for technology2Define an evolving operating model3Take a gradual approach and demonstrate value early on4Make progress measurable5Go where the data citizens are6Be clear on roles and responsibilitiesWhen companies plan their investment budgets for data governance, they typically consider license, implementation and maintenancecosts. The cost of not properly nurturing and adopting data governance is huge. However, when it comes to budgets and priorities, dataculture and adoption are often pushed down the checklist. Why? Evidence shows that companies that have been successful in scaling datagovernance spent more than half of their analytics budgets on activities that drive adoption.There is no single way to do data governance. Therefore every data governance programme starts from defining (and evolving) theoperating model that shapes the execution of it. The operating model can vary based on your company’s priorities, culture, organisationalstructure, and even internal politics! Deciding on the operating model your organisation will adopt is part of the initial steps in setting upand establishing a successful data governance program.One of the challenges is often that the value creation potential of data governance is indirect and can take time to show, while people tendto expect immediate reward for their investments. Therefore, instead of trying to build everything at once, focus on quickly demonstratingthe value of data governance through a few carefully selected use cases that are tied to a significant business challenge. Define andprioritise a list of use cases and specify the benefits or business value they will bring to your company. Start working on the use cases withthe highest feasibility and highest business impact. From there on you expand data governance and its impact across your company. Forexample, if there is a self-service analytics initiative going on, streamline business reporting by setting up a metric or report certificationprocess. Use that value and momentum to expand into other use cases. Instead of communicating the value, show it to people and makethe link between data governance and visible value. From there, leverage your successes and gradually remove more pain points.In line with the previous point, make sure to develop success metrics for your data governance programme. These metrics can providean opportunity to establish a baseline to know what bad data means, and potentially, what good data could mean. Hence, they create astarting point to get the broader organisation to understand the need for better data. Second, the right metrics can help your companyto get aligned on a set of shared goals, which is key to success. Make sure to monitor adoption rates, which give an indication of the extentyour data governance programme is consumable by people, whether they understand it and care enough about it to adopt it. This iswhere you may consider working with Key Behavioral Indicators (KBI’s) or a set of concrete and measurable behaviors which promoteclear ownership, minimize friction between stakeholders involved, increase awareness of individual behavior, and enhance the value beingdelivered by each stakeholder group. Clearly, you don’t just want to measure progress and adoption, you want to make sure that you’remeasuring the value that you’re providing to the organisation. In addition to progress metrics, ensure you have well-established impactmetrics too (e.g. improvement in report quality).Data governance is a new concept for most people today, and whether it is change that scares them or the fact that there are additionalresponsibilities they fear, you have to get to people where they are. Don’t ask them to come to you but go to them and offer solutions intheir work context. For example, consider working with an on-the-go-application, which provides access to data anywhere, anytime. Weknow from neuroscience that high effort significantly reduces the likelihood of new behaviours to occur and be maintained. Hence, ashighlighted in one of our previous articles on building a data culture, think about designing an easy (as effortless as possible), engaging anduser-centric approach.Imagine that your organisation put up an internal data marketplace using a catalogue. A data consumer finds an interesting data asset butmight have questions about it. Who do they turn to? Or in case they want access to the data, who should validate the legitimacy of thatrequest and approve or reject it? Typically, that would be the data owner, and just like that role and its responsibilities need to be clear,communicated, assigned and managed, there are other roles your operating model for data governance will require.3