{"id":7286,"date":"2023-12-08T12:13:54","date_gmt":"2023-12-08T12:13:54","guid":{"rendered":"https:\/\/members.dataiq.global\/?post_type=article&p=7286"},"modified":"2024-03-20T15:39:15","modified_gmt":"2024-03-20T15:39:15","slug":"how-to-be-data-stewards","status":"publish","type":"article","link":"https:\/\/www.dataiq.global\/devstage\/articles\/how-to-be-data-stewards\/","title":{"rendered":"How to persuade data owners to be data stewards"},"content":{"rendered":"

What do data stewards mean?<\/span>\u00a0<\/span><\/strong><\/h4>\n

The first thing contributors to the discussion noted was that there is no\u00a0set definition across the industry as to what constitutes the remits of a data owner versus a data steward.\u00a0Data professionals around the table agreed that the terms held different meanings, but the scope of work expected by owners and stewards varied from business to business.\u00a0<\/span><\/p>\n

Danilo provided each participant with the chance to explain how their team and organisations differentiate between the two so everyone could understand the variables seen across the industry.\u00a0It was agreed that\u00a0although\u00a0the responsibilities of data stewards and owners can vary\u00a0between\u00a0organisations,\u00a0this is not necessarily a problem. The attendees confirmed that the owner must be the most accountable person\u00a0for that data,\u00a0whereas\u00a0the steward needs to\u00a0oversee\u00a0handling the day-to-day concerns of those\u00a0particular data\u00a0sets.\u00a0<\/span>\u00a0<\/span><\/p>\n

\"Danilo
Danilo Sato, European Head of Data and AI, Thoughtworks.<\/figcaption><\/figure>\n

However, this raised a concern\u00a0regarding\u00a0a lack of awareness surrounding the parameters of the roles.\u00a0These need to be addressed by each individual business as the exact scope of work can vary depending on team sizes and niche.\u00a0Once the parameters have been agreed, it makes it easier\u00a0to encourage people to embrace the role of data steward as there are clear boundaries of what the role entails.\u00a0<\/span><\/p>\n

A simple\u00a0definition list\u00a0was agreed by the roundtable:<\/span><\/p>\n

    \n
  • Data owners:<\/span><\/strong>\u00a0<\/strong>Highest person accountable for data, usually a business stakeholder.<\/span><\/li>\n
  • Data stewards:<\/strong>\u00a0Someone more connected to day-to-day\u00a0use\u00a0of data\u00a0who\u00a0can\u00a0understand the technical aspects of data and systems where it originates.<\/span><\/li>\n
  • Data governor:<\/strong>\u00a0Ensure data owners and stewards are in place and keep them accountable.<\/span>\u00a0<\/span><\/li>\n<\/ul>\n

     <\/p>\n

    Who is in charge?<\/span>\u00a0<\/span><\/strong><\/h4>\n

    As with many things\u00a0regarding\u00a0data offices, there is no one-rule-fits-all solution for a chain of command and responsibility. Contributors to the roundtable noted that they themselves have experienced differences for the data teams across\u00a0different positions\u00a0in their careers: some reported into an IT function, others had direct access to the CEO, some found themselves siloed and left to their own devices,\u00a0whereas\u00a0others were a standalone team with equal weighting to other senior departments. <\/span><\/p>\n

    The same problem can be found with assigning data owners and data stewards \u2013 there is no correct way of doing it that must be uniform across all businesses. Data leaders need to assess their\u00a0capacities, the\u00a0objectives\u00a0of the data team, the\u00a0objectives\u00a0of different departments and\u00a0the trajectory of business operations to accurately\u00a0determine\u00a0who should hold responsibility, a chain of command and the\u00a0parameters\u00a0of the positions.\u00a0<\/span>\u00a0<\/span><\/p>\n

    The next step is ensuring that the person in charge \u2013 whether that is a steward or an owner \u2013 is incentivised to perform their job to the best of their ability.\u00a0This can be achieved through a combination of top-down from executives and leaders championing the reasons behind why excellence in data stewardship is needed, and a bottom-up approach from everyone that utilises data for their own\u00a0objectives\u00a0demonstrating\u00a0data quality needs and suggesting improvements.<\/span>\u00a0<\/span><\/p>\n

    Education is a huge part of the challenge, and it was agreed by all participants that, ultimately, everyone\u00a0is responsible for\u00a0data, but this means there must be a minimum level of data literacy across the business. One member of the roundtable used\u00a0the analogy of football to help win over decision makers<\/span>\u00a0<\/span><\/p>\n

    By explaining how FIFA, the governing body is like the data governor, the referees are data stewards\u00a0and owners dictating the rules and enforcing policies and finally the players on each team are the individuals that use data day-to-day and curate the data.<\/span>\u00a0<\/span><\/p>\n

    A couple of participants explained that their data owners and data\u00a0stewards\u00a0attend regular meetings as a group to learn from each other. This gives them the chance to question any problems that may be arising, learn new stewarding techniques and to\u00a0identify\u00a0areas in the business that need more support.\u00a0This is also\u00a0a good time\u00a0for people to request\u00a0assistance\u00a0or guidance if they feel they are misunderstanding the remits of their role, or to check in with a data steward or leader that\u00a0is responsible for\u00a0them.\u00a0<\/span>\u00a0<\/span><\/p>\n

     <\/p>\n

    Should there be quality minimums?<\/span>\u00a0<\/span><\/strong><\/h4>\n

    Further\u00a0exacerbating\u00a0the nuanced description of a data owner and a data steward is the fact that quality standards for data vary throughout an organisation and can vary depending on the situation. It is imperative that whoever is tasked with being a data owner or a data steward has a solid grasp of data literacy to be able to understand these distinctions and\u00a0provide\u00a0the most suitable data solutions.<\/span>\u00a0<\/span><\/p>\n

    It was mentioned during the roundtable that when it comes to data quality, people are only interested in certain aspects that pertain to the need of them and their team. Data leaders find themselves as the only ones working to ensure widespread consistency with quality over an organisation,\u00a0whereas\u00a0the majority of\u00a0people only care about quality when it comes to their specific function. Whether this is due to\u00a0a lack of data literacy, a still-evolving data culture or something else was left for debate. It was agreed, however, that organisations with\u00a0stronger levels of data literacy and more maturity find themselves in a stronger position to address the data quality gulf, but the issue did\u00a0still remain\u00a0in one form or another.\u00a0<\/span>\u00a0<\/span><\/p>\n

    \u00a0<\/span>One way to assess and address the\u00a0different levels\u00a0of quality is to instigate data quality checks, although this can be time-consuming to implement and difficult to\u00a0monitor. As some contributors noted, there are\u00a0frequently\u00a0times where people will simply ignore\u00a0the data quality tests as they consider it a burden and pointless. A solution was put forward by one participant:\u00a0<\/span>use the\u00a0data quality tests\u00a0as blocking tests to ensure quality problems are fixed quickly and by the team closest to the source. This would of course still require a level of monitoring and enforcement, but it would help instil a sense of ownership and responsibility for decentralised teams.\u00a0<\/span>\u00a0<\/span><\/p>\n

    The roundtables noted that there should be some form of incentive for people to\u00a0maintain\u00a0data quality minimums and aspire for higher quality. Whether this comes in the form of monetary compensation, company culture improvements or another\u00a0method would depend on the individual organisation and team, but it was agreed that people need to be rewarded for their time and effort taking on another role.\u00a0<\/span>\u00a0<\/span><\/p>\n

    The use of dashboards is\u00a0a simple way\u00a0to address low-level data quality concerns and to\u00a0maintain\u00a0a touchpoint with teams across the globe. A couple of roundtable members explained how they had implemented explicit KPIs for data governance and quality that can be\u00a0monitored\u00a0through their dashboards, which in turn helped improve the data culture of the business. It was the questioned if there could\u00a0perhaps be\u00a0a level of gamification or competition developed through these dashboards where teams can compare themselves against each other and achieve team benefits. One member also\u00a0stated\u00a0that, in their experience, a fear of missing out \u2013 FOMO \u2013 can be a good incentive and gets people actively wanting to become a data owner or steward.\u00a0<\/span>\u00a0<\/span><\/p>\n

    Ultimately, there\u00a0are a few steps that data leaders need to take to be able to influence and encourage team members into becoming data stewards, and this will take time. It will also depend on the size of the\u00a0business,\u00a0the types of data being utilised and the\u00a0way in which the data is used by the team. The main thing is that data leaders\u00a0are able to\u00a0explain how and why data quality is such an essential part of business operations and that having data owners and stewards is a prime way to achieve excellence. Once trust has been built by the data leader through improving data quality being used by different departments, the evolution of data culture and data literacy will gain momentum and the role of data stewards will become engrained in the wider business operations.\u00a0<\/span>\u00a0<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

    It is one thing to own data sets, but another entirely to be responsible for its use and upkeep \u2013 so how can data leaders encourage team members to become data stewards?<\/p>\n","protected":false},"author":19,"featured_media":7287,"menu_order":0,"comment_status":"open","ping_status":"closed","template":"","format":"standard","meta":{"_acf_changed":false,"content-type":"","_searchwp_excluded":"","footnotes":""},"categories":[129,281],"tags":[343,171,342,341,344,93,233,340],"pillar":[198,194],"class_list":["post-7286","article","type-article","status-publish","format-standard","has-post-thumbnail","hentry","category-editorial","category-roundtables","tag-architecture","tag-data-governance","tag-data-owner","tag-data-steward","tag-ownership","tag-quality","tag-roundtable","tag-thoughtworks","pillar-governance","pillar-leadership"],"acf":[],"publishpress_future_action":{"enabled":false,"date":"2026-05-20 22:43:56","action":"change-status","newStatus":"draft","terms":[],"taxonomy":"category","extraData":[]},"publishpress_future_workflow_manual_trigger":{"enabledWorkflows":[]},"_links":{"self":[{"href":"https:\/\/www.dataiq.global\/devstage\/wp-json\/wp\/v2\/article\/7286","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.dataiq.global\/devstage\/wp-json\/wp\/v2\/article"}],"about":[{"href":"https:\/\/www.dataiq.global\/devstage\/wp-json\/wp\/v2\/types\/article"}],"author":[{"embeddable":true,"href":"https:\/\/www.dataiq.global\/devstage\/wp-json\/wp\/v2\/users\/19"}],"replies":[{"embeddable":true,"href":"https:\/\/www.dataiq.global\/devstage\/wp-json\/wp\/v2\/comments?post=7286"}],"version-history":[{"count":0,"href":"https:\/\/www.dataiq.global\/devstage\/wp-json\/wp\/v2\/article\/7286\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.dataiq.global\/devstage\/wp-json\/wp\/v2\/media\/7287"}],"wp:attachment":[{"href":"https:\/\/www.dataiq.global\/devstage\/wp-json\/wp\/v2\/media?parent=7286"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.dataiq.global\/devstage\/wp-json\/wp\/v2\/categories?post=7286"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.dataiq.global\/devstage\/wp-json\/wp\/v2\/tags?post=7286"},{"taxonomy":"pillar","embeddable":true,"href":"https:\/\/www.dataiq.global\/devstage\/wp-json\/wp\/v2\/pillar?post=7286"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}