Transforming a Data-Driven Company Culture Starts with Management
Data and data analytics have immense potential, at least in theory. An alarming survey revealed that 85 percent of all data projects fail, which is a troubling figure for a cutting-edge technology.
These projects tend to start out ambitiously and optimistically. Then they devolve into busted budgets, missed deadlines, technical labyrinths, and worthless insights. The scale and complexity of data often gets the blame. But that is hardly fair. In reality, it’s mostly because of managers that these projects go off track.
It’s ultimately up to those at the top to call the shots. That requires confidence, along with an equal understanding of one’s blind spots and weaknesses. Everyone, executives included, struggles with self-awareness from time to time. And it’s because of faulty assumptions like these that data projects sometimes produce tiny results:
- Databases Are Ripe for Analysis – Actually, most databases are full of transactional data with redundant entries and little overall value. Success in data starts by collecting better data.
- Databases Are in Perfect Shape – Databases start out with a structure, then quickly deviate once they begin to grow and evolve. Integrating and leveraging databases requires better databases overall.
- Database Contains No Errors – Most of the information in databases was entered manually, making it ripe for errors and omissions. Until these are cleared up, data-driven insights can’t be fully trusted.
- Databases Work in Aggregate – Combining disparate databases is a complex process where it’s easy to lose the connections between information. Rushing to integrate databases without careful planning turns the final product into a dubious asset.
- Databases Only Need IT – Data is a tech-driven process, but software alone can’t solve the inherent challenges. Relying on a data-literate workforce makes any initiative more successful.
- Databases Are Open to All – Databases are designed for a specific type and number of users. Repurposing that database for a larger purpose can stress it to the point where it’s unusable for any purpose.
Managers are typically not tech professionals, so it makes sense that they misunderstand some of the tech behind data. But that’s not an excuse for overseeing ineffective initiatives. It’s also not an obstacle that is impossible to overcome. To the same degree that managers can trip up data they can also facilitate its success.
That begins by understanding that architecture drives performance. Smartly-structured databases collecting high-quality data are the foundation of any data project. Investing in those first improves everything that comes after.
Establishing clear goals and expectations is also important. Management is often opaque about what they really expect data analytics to do. The project moves forward without a clear purpose simply because it’s never been defined. As early as possible managers should establish concrete data analytics goals, metrics to monitor, and benchmarks to define success.
The final and most important piece of the puzzle is to stop thinking of data in isolation. Managers treat data like a precious object that can only be understood by an elite group of scientists and executives. This attitude simply limits how many resources go into a project upfront. And it has a similarly limiting effect of the results those projects produce. In general, the best approach is to treat data as a resource that is important and accessible to all.
If you’re a manager at a company struggling to become more data-driven, don’t let this analysis leave you dejected. One survey of executives showed that only around 1/3 had succeeded at creating a data-driven culture. This is an extremely common problem, and an extremely preventable one. All it takes is for those at the top to acknowledge their mistakes and embrace something different.