Imagine the sheer volume of data that a supplier of rough diamonds to global markets needs to process in order to plan its operations. In the specialist independent jewellery retail sector alone this amounts to millions of transactions, thousands of products, and a huge network of about 1,000 retailers. Adding a further five countries increases the complexity of the data and the analytical work.
This challenge is further intensified by the diverse offerings of the jewellery industry. De Beers needs to understand how diamonds compete within a constantly evolving landscape. Jewellery comes in a range from diamonds in numerous sizes and qualities to pieces with precious colour gemstones and pieces without any gems. This variety makes it challenging to classify and analyse data consistently. It's a data deluge that can be overwhelming, even for the most experienced analysts.
De Beers needed a quick way to turn this firehose of raw data into actionable insights that could drive growth and improve efficiency.
The solution we created together is a custom-automated data analytics system which transforms raw data into a well-structured format, and automatically highlights key insights – so facilitating straightforward human analysis. Every two weeks, point-of-sale data is imported into the model. This leverages algorithms and machine learning to process and analyse large volumes of information. Through this process, the system detects trends, patterns, and picks up on anomalies. This detection of anomalies has significantly improved data quality, as one stakeholder said: “Speed of data anomalies detection has enabled speedy resolution with the data provider, including remapping at their end.”
To tailor the solution to the specific challenges of De Beers, we created a hierarchical structure for the data. This capability allows the De Beers team to drill down from high-level overviews to granular details, providing a comprehensive understanding of their business at every level. As one of the stakeholders at De Beers commented: “The creation of a hierarchical structure for the data has been essential in understanding the full picture.”
Using the system, we generate a clear, concise table in Excel for more experienced analysts to dive into, and a PowerPoint presentation for the senior leadership, giving them a quick snapshot of key metrics. These reports enable decision-makers at De Beers to get an overview of the business quickly and save time for more strategic thinking. This has been well received by stakeholders: “Fast automated monthly summaries provide ‘at a glance’ updates for decision-making. A great win!”
Even with the impact we have made so far, our work isn’t completed. The team are brewing up new solutions to tackle further challenges: inventory data analysis, automated executive summaries, and end-to-end integration.
The results for De Beers? We’ve reduced time-to-insight, mitigated risks, supported more informed decision-making, and uncovered valuable insights from their mountain of data.
