The Road to Better Business Insights: Deciphering Dashboards and Decoding with Data Product Managers

Over the last decade, companies have poured billions into data infrastructure. Petabyte-scale warehouses, real-time pipelines, and machine learning platforms mushroom every corner of the business field. However, the torrent of data has only clouded the waters. One question flung at the operations lead and you’re greeted with a whirl of conflicting dashboards. A query to finance about performance reconciliation and the response is a vague ‘it depends on who you ask’. Amidst all the chaos, one truth stands tall - the problem isn’t in the data, but in how we understand and manage it. For years, data teams played the part of internal consultancies - reactive, self-driven, and ticket-based. This ‘data-as-a-service’ model worked well when data requests were small and the stakes, low. However, as companies increasingly leaned into being ‘data-driven’, the old model stuttered and crumbled under its weight. A prime example here is Airbnb. Before the launch of its metrics platform, the product, finance, and ops teams each had their own versions of metrics. The simple Key Performance Indicators varied by filters, sources, and depended on who posed the question. Consequently, different teams presented different numbers during leadership reviews, sparking arguments over the accuracy of metrics rather than actions to be taken. These aren’t technology shortcomings. They denote product failures. The dawn of such discrepancies seeds distrust in data. Analysts find themselves being second-guessed, dashboards are deserted, while data scientists spend more time clarifying discrepancies than generating insights. The system thus falls into a pool of inefficient redundancies and delay in decision-making. At closer inspection, it’s not a data quality issue at play, but a data trust problem: the system works, the SQL checks out, but the GPS of trust in the output is lost. This is where a new saviour in the form of role, strides across the data-driven battlefield – the Data Product Manager (DPM). Unlike their generalist counterpoints, DPMs navigate the brittle, unforeseeable, and cross-functional terrain of data. Their task is not just to manage data but to interpret it accurately, build trust in insights, and ensure seamless transition of these insights into actionable strategies.