This is the first in a series of blogs where Data Scientists Cory Everington and Anna Godwin discuss five Analytics Best Practices that are key to building a data-driven culture and delivering value from analytics. In this installment Cory discusses the benefits of having a shared framework of Analytics Best Practices to allow you to focus on what's most important—the results.
Have you ever collaborated on a complex data science project? How did you manage the workload? Did your code break when multiple people were trying to develop it?
Successfully collaborating on a data science project within time and budget constraints is a challenge! A large, perplexing problem has to be broken down into individual project tasks, and assigned to owners -- whose progress needs to be tracked, verified, versioned, documented, shared, joined, and communicated – constantly. A shared framework of Analytics Best Practices helps smooth the work process, and allows you to focus on what's most important—the analytic insights.
Through more than 20 years of data science consulting experience Elder Research has developed Analytics Best Practices to efficiently manage our project work. This enables us to quickly provide data-driven, actionable solutions to our clients at a reasonable cost. In this multi-part blog series, we will share five keys to building a well-rounded analytics team and successfully deploying analytics projects:
An overview of each topic is presented below, and further details will be provided in our ongoing Analytics Best Practices blog series.
Data scientists benefit from employing agile and lean methodologies. We focus on deployment starting with the earliest stages of the analytic lifecycle. This facilitates collaboration between data scientists, software engineers, and Subject Matter Experts (SMEs). Agile Data Science allows teams to iteratively refine deliverables, generate early buy-in with stakeholders, and deliver a solution the client will use.
“There are no unicorns” is a common refrain in the data science community, reflecting how unlikely it is to find one person who has all the skills required to succeed at data science. The best strategy is to build out a diverse team and select the right people based on the project requirements to ensure project success.
Whether you are a team of one or thirty, use version control for all your projects! Formally checking in your code and research ensures that your work and all of its previous incarnations live on.
Do you find breaking up large analytics tasks into manageable chunks a challenge? We will share our tips for identifying the right task for the right level.
Communication and Collaboration
Few data scientists work alone. If communication is poor, adding people to a project can actually slow progress, but if collaboration is strong, amazing productivity is possible. We will share our experience on how to bring your team together and ensure everyone on the project team is on the same page.
Whether you're a data analyst, business intelligence analyst, data scientist, project manager or other analytics professional, we invite you and your team to follow our blog series and learn some Analytics Best Practices that can make your day-to-day work easier and more productive.
Request a consultation to speak with an experienced data analytics consultant.
Download the eBook The Top 10 Data Mining Mistakes
Read the blog Avoiding Common Data Science Business Mistakes
Check out the book Mining Your Own Business for more best practices on how to harness the power of data science and predictive analytics, and avoid costly mistakes.