How do Middle Eastern Companies Keep the Faith Around Data?

Be the first to learn about new ideas in tech and management with the MIT Sloan Management Review Middle East newsletter.

Sign up here and stay up to date with the latest in enterprise tech, research, trends, and insights.

Each year, Virgin Mobile UAE handles information from across its business that could fill the equivalent of 2,000 entry-level MacBook Air computers. Those 500 terabytes of enterprise data improve the customer experience while supporting better decision-making, says Dr. Imran Shaikh, PhD , Chief Technology Officer, Virgin Mobile UAE .?

“We collect and store data such as demographics, usage patterns, transaction history, and feedback, which are extremely helpful in constantly enhancing the end-user experience for the services we provide.”

However, Dr Shaikh and his team must manage increasing amounts of data each year, with annual volumes growing at 20%.?

“We are always trying to find solutions to continuously improve the accuracy and consistency of data across various sources,” Shaikh says.?

While the company partners with some of the world’s top data management firms to ensure actionable insights are always available, he says privacy and governance policies are equally important.

As businesses interact with more digital touchpoints in new ways, they must capture more parameters to stay competitive.?

With that growth comes a new problem: ensuring this growing mountain of data remains accurate, consistent, and trustworthy throughout its lifecycle. Poor data quality can lead to costly mistakes and missed opportunities. Organizations say substandard data causes annual losses of $15 million on average, according to Gartner research.

“The quality of the output depends on the quality of input,” says Prof. Dr. Eman AbuKhousa , an enterprise systems expert who is a Professor at the University of Europe for Applied Sciences Dubai .

When businesses rely on data for crucial decisions and improving operations, it’s essential to have accurate, reliable, and clean data, Dr AbuKhousa adds. “Only organizations with superior data quality can gain insights faster and more accurately, achieving a competitive edge. Maintaining high data quality is [also] essential for supporting sustainability goals.”?

The challenge is becoming more pronounced in the region as economies rush to establish leadership positions in digital economy sectors such as AI, financial technology, and cybersecurity to accelerate growth and improve socioeconomic outcomes.?

Gabriele Obino , Regional VP and GM for Southern Europe and the Middle East at Denodo , says the MENA’s ambitious projects and global collaborations necessitate reliable data solutions.

“Business growth and innovation across many industries, and the speed at which this has happened, have been incredible. However, with speed, there is little room for error. Insights and decisions need to be correct and actionable from the beginning, and for this, trustworthy data is critical.”?

Data and insights available to an organization must be relevant and immediately actionable. Although tools to detect and fix bad data have been around for decades, that is only part of what it means for data to be trustworthy by the business, Obino says. Fixing a misspelled name is easy, but only a more holistic view can identify high-value customers. This is where generative AI is being put to work with its ability to parse large datasets.

Automation typically streamlines and speeds up the process of combining data from many disparate sources, and AI can help non-technical users find insights quickly and effortlessly.

“The ability to access data in any location, at any time, and analyze it in seconds rather than hours is not an opportunity to be passed up,” says Karl Crowther , Vice-President – of MEA at data science and analytics company Alteryx .?

“However, companies that have not already democratized data and analytics at scale to accelerate the data journey will face complex implementation challenges,” Crowther warns.

As organizations look to make the most of data, basic integrity challenges remain.

“In addition to challenges posed by the volume, diverse formats, and rapid production of data, which complicate management, integration, and accuracy, the most significant challenge is securing the data and ensuring compliance with privacy regulations,” Dr AbuKhousa says.?

“Ethical considerations are paramount, as organizations must ensure transparency, fairness, and accountability in data usage.”

From data sanitation to sovereignty: Six routes to quality data

There are several ways to tackle issues around data quality and integrity.

1. Start with a data-driven culture: Dr. AbuKhousa highlights the need for a data-centric culture across the organization: “Involving stakeholders from various departments in data governance and quality initiatives ensures buy-in and collaboration. This foundation is crucial before implementing strategies such as developing data quality management frameworks, adopting data management practices, and establishing ongoing monitoring and improvement processes.” Schedule regular training to educate employees about data quality.?
2. Prioritize governance: Robust data privacy, governance policies, and quality control mechanisms are vital for maintaining high-quality data, including in AI applications. “This involves standardizing data collection processes, establishing clear data ownership, and implementing data cleansing and quality checks tools,” Crowther says. Consider data stewardship and assigning data wardens to oversee specific domains’ accuracy, privacy, and security.?
3. Democratize access to data: More employees want to use data to make decisions, and expanding access can reduce IT bottlenecks and support better performance. “Leaving data management to IT departments means that domain knowledge can be lost. By keeping domain experts engaged in data management activities, you can capitalize on their specific expertise to guarantee better data quality,” says Obino. ?
4. Standardize data management: Without documentation around data management, much more is left to users interpretation, Obino says. One solution is to unify all data at the point of consumption to ensure it meets the users’ needs, minimizing trust issues in the process. “It’s best to have data quality enforced as close to the point of consumption as possible.”? ?
5. Encourage industry best practices: Regular audits, data validation, and employing redundancy systems can ensure high standards, consistency, and quality outcomes, says Dr Shaikh. “We also recently improved our data accuracy by implementing machine learning algorithms to detect and correct anomalies in real-time, reducing error rates by 25%.” Also, consider adopting a center of excellence for data management to guarantee consistency around creating and sharing data.?
6. Secure to assure: Embrace techniques such as security policies and attribute-based access control. Data quality practice will only be rendered useful by breaches, Obino says. Stringent cybersecurity protects against unauthorized access and breaches, maintains regulatory compliance, and bolsters overall data integrity, for instance, preventing data poisoning by malicious attackers.?

Follow us on LinkedIn , Twitter , Instagram and Facebook .