In a constantly evolving technological society, markets are becoming more complex and customers more demanding. A structured, strategy-driven S&OP process is one of the key elements to ensure fast, data-driven decision making. Data-driven. A well-known adjective amongst supply chain professionals. Despite of its abundant use, companies rarely think about the hidden meaning: in order to have data-driven decisions, we must ensure that the data is available, structured and reliable. And in any case, prevent it from being a showstopper.
Data quality is a concept that companies are well-aware of. On the other hand, they are also hesitant to invest time and money in it because the benefit does not lie in the process itself, but in the subsequent processes of planning and decision making using the data. Furthermore, data cleaning is often perceived as an internal task that can be done without external partners. However, there are several pitfalls to companies who identify with this approach.
1. Where to start?
Companies have numerous amounts of data and are - as a result - often unaware of what they can do with it. In the context of a planning process or project, people become aware of which data is important and which changes are needed. So, ask the advice of your planning colleagues. Their insights will help you to create focus on improving data where it matters most. Expanding this effort into various departments of the organization will result in having more relevant data available.
The quality of data is often a “feeling” not endorsed with the right metrics and quantitative information. An external view not only provides the knowledge and expertise on which information to use but also accommodates a critical, objective look. For example, in a typical S&OP process, demand planners will clean their data for the COVID-19 period in order to obtain an unbiased statistical forecast for future sales. By combining your internal knowledge with external indicators, you’ll provide insights in the drivers of your demand. It will allow planners to more reliably clean and assess the impact of the current crisis period or any upcoming events.
Few companies perceive data quality as a continuing process. It is perceived as a one-off exercise at the beginning of a project or as a separate project. However, a one-off snapshot will only get you so far. Data is an always changing and evolving matter. As a predecessor to other processes, it is important to establish data quality as part of the entire process rather than a single exercise. The above-mentioned example on COVID-19 affirms this statement. Is data-quality embedded in your next supply chain improvement track?
Not uncommonly, the detection of bad quality of data is used as a seemingly valid excuse to postpone the start of planning software implementation projects. However, bad quality should never stop those implementations, as the gradual implementation itself can contribute tremendously in finding discrepancies and errors in the data – if proper data quality reports are available. Also, planning systems will help in making decisions, but should not – or at least not in a first stage – create their own (wrong) decisions based on that (wrong) data. So, when managed in the right way you can leverage the project to increase your data quality on the go.
Being perceived as time-consuming and costly, data quality processes are seldomly established as part of the planning process. Yet, the success of every advanced planning process heavily depends on the availability and accuracy of data. “Garbage in, garbage out” is a concept that has been around since the 60s, known but often forgotten.
Data quality is a foundation for data-driven decision making. And, without a solid foundation, the house will collapse. So, start by gradually tackling that big pile of data you’re sitting on, use external indicators to ameliorate the output and embed this procedure in your overall supply chain process.