Today’s most competitive companies are data-driven. Consider Uber, which assailed the taxicab industry in San Francisco in just a few years, without owning a single taxicab. How? Among many innovations, Uber brought data to the taxi industry. Using historical data, Uber advises drivers to be in certain hotspots during certain times of day to maximize their revenue because customers tell them with the push of a button where to be.
These companies don’t rely on hunches, siloed spreadsheets, or data on rogue servers to make decisions; instead, they have operationalized data as a part of every process and decision and built cultures where guesswork doesn’t suffice. Operationalizing data, or using data to improve business performance, will be the defining competitive advantage of the future.
There can be roadblocks, but there are ways to outmaneuver them. Here are some of the most common hurdles:
- Data breadlines:
In many businesses today, employees are data-poor. Overtaxed data analysts can’t keep up with demand for queries, and even if they can meet demand, employees are often left with more questions. What if I haven’t asked the right question? How can I look at this old problem from a new angle? All of that takes time, and can lead to employees abandoning their curiosity. Or, worse yet, relying on inaccurate hunches that may lead the business in the wrong direction.
By developing the right metrics and language, educating teams to analyze data without bias, and rewarding curiosity with the best tools to find the right information, companies can maximize the speed of their decisions. Data education, data literacy, and data tooling are three key ingredients to evolving a company’s culture to becoming more data-driven. When a data team trains its colleagues in data literacy, the team becomes a huge lever for the business and a powerful tool for eliminating data breadlines.
- Data obscurity:
Once an employee has been patient enough to reach the front of the data breadline, he has to ask the data analyst team to help him answer his question. That is not always as straightforward as it may seem. Within companies, there could be dozens or even hundreds of locations for data in use by various functional teams—sales, engineering, and so on. Employees often do not know how data is organized, who created it, who manages it, and what it contains. This makes it tougher to tell the data scientist what to query. Only a few select data engineers understand how the labyrinths of data are organized. With such high stakes for becoming data-driven, everyone needs a better way to make data more digestible for the entire company, not just for data scientists.
- Data fragmentation:
Overly delayed by the strapped data team and unable to access the data they need from the data supply chain, enterprising individual teams often create their own rogue databases. These shadow data analysts pull data from all over the company and stuff it into database servers. The issues with this approach rear their heads quickly. Among them, duplicated, isolated rogue datasets that quickly become outdated and fragmented data in silos. The key is to build a single database—a single source of truth—that can be easily queried by everyone in the company, without them having to be data scientists or sophisticated coders.
- Data brawls:
Data fragmentation has another insidious consequence. It incites data brawls, where people labor over figures that just don’t align and that point to diametrically different conclusions. Imagine two well-meaning teams, a sales team and a marketing team, both planning next year’s budget. They share an objective: to exceed the company’s bookings plan. Each team independently develops a plan, using metrics like customer lifetime value, cost of customer acquisition, payback period, sales cycle length, and average contract value. Unfortunately, because they are using two different sources of truth, the plan devolves into a heated argument.
Unifying the silos, quelling the brawls, and satisfying the breadlines are all necessary steps, but inculcating a data-driven culture is far more fundamental. When everyone is capable of using data that is based on a single source of truth, analysts are no longer a bottleneck to getting things done and they can nurture a data-driven culture. They can become available to answer the tougher questions that take a little more time and scale, so they can add more value, train more employees to become data-literate, and help everyone make a bigger difference based on data.
*Winning with Data, by Tomasz Tunguz and Frank Bien, provides a fuller description of these concepts.
Frank Bien is CEO of Looker and co-author of Winning with Data: Transform Your Culture, Empower Your People, and Shape the Future. With over 20 years growing and leading technology companies, Frank Bien built his career on nurturing strong corporate culture and highly efficient teams. Prior to Looker, Frank was SVP of Strategy for storage vendor Virsto (acquired by VMware) and VP of Strategic Alliances at big-data pioneer Greenplum, leading their acquisition by EMC (now Pivotal). He led Product Marketing and Strategy at early scale-out data warehousing company Sensage and was VP of Solution Sales at Vignette/OpenText. Earlier in his career he held executive roles at Dell and the Federal Reserve. Frank recently co-authored the book, Winning with Data: Transform Your Culture, Empower Your People, and Shape the Future (Wiley), which takes a deep dive into big data in business, explores the cultural changes it will bring and discusses how to adapt an organization to leverage data to its maximum effect.
Subscribe to Data Informed for the latest information and news on big data and analytics for the enterprise.