It’s a common misconception that organizations need a robust engineering layer, an available data warehouse, and advanced analytics capabilities in order to derive insights from their information. However, this is not always the case. Companies can still embark on data science initiatives even with limited data and a relatively green information management environment.
If your company is in this situation, read on for some quick tips to guide your efforts and get you started:
- Your business problems should determine the kind of analytics you need
According to Gartner, approximately 80 percent of data science projects fail to deliver business outcomes. One reason for this high failure rate is that many IT leaders choose projects based on available data, skills, or tool sets rather than assessing what business problems need solving.
Embarking on a data science journey under the latter framework enables companies to identify the most critical issues that target users want to see addressed—and determine whether solving them will deliver the desired impact on the business. Answering these questions will then dictate the right analytics approach and, subsequently, the data required for the project.
- The analytics approach should dictate the data you source
It’s not unheard of for companies to spend significant time and resources building a data warehouse only to realize that the data they have collected isn’t robust enough to perform the analytics they need. A better approach is to obtain the necessary visibility into organizational strategy and business problems to determine whether you need descriptive, diagnostic, or predictive analytics and how the insights will be used.
- Data collection begins with easily available small data
Big data certainly has its place, but many business challenges can be solved via descriptive analytics on small spreadsheets of data. By reducing the data requirements to a few hundred rows, companies can manually collate data from systems, digitize paper records, or implement simple systems to capture the required information.
- Adopt an incremental approach to deliver transformational value from data
The final consideration is to develop an iterative process to source the data you need on an incremental basis. Don’t waste time trying to anticipate all the potential sources of data you will need as your journey progresses. In fact, getting started on data analytics projects can actually help build a robust data engineering roadmap for future initiatives.
For more on the above and why having limited amounts of data is no reason to abandon your data science goals, check out this Enterprisers Project article.