Thanks to those who read our August newsletter and emailed questions for Mariana, our data scientist. As promised, here are some of the most useful questions, and Mariana’s answers.
Brad: When you say we need to embrace tabular data, do you mean spreadsheets? Is there special software we can use to make this easier?
Mariana: Spreadsheets can be used to organize data in columns and rows, but the concept of tabular data involves more than just inputting information into an Excel sheet. It is a structured method of gathering, arranging, and comprehending data with a specific, well-defined objective.
Before thinking about how to format the data, it’s essential to first clarify what your goal is. For example, do you want to track your energy consumption, or are you also interested in the cost of bills, the source of energy (e.g., solar vs. grid electricity), or even carbon emissions? Does all the data come from the same source, or will you need to connect data from different systems (like electricity usage from meters and billing data from utility providers)?
Once these goals are clear, you can focus on defining how to format your variables, making sure units are consistent (e.g. kWh for energy, $ for costs), and ensuring there’s a culture of data consistency across the organisation. Tabular data, at its core, is the result of thinking critically about why you’re collecting the data, how frequently, what limitations might exist, and what minimum standards the data need to meet to deliver quality insights.
When it comes to tools, the choice really depends on the size of the company and the complexity of the data being collected. For most small to medium-sized companies, Google or Microsoft’s platforms offer a solid starting point for managing energy and emissions data. Sheets or Excel can be used for the initial setup and using it through Drive or SharePoint allows multiple people to collaborate efficiently. Moreover, you can automate some of the data collection and pre-processing steps using tools like Power Automate and analyse your results with Power BI. This solution can be a cost-effective alternative to buying specialised software. For companies that manage millions of data points or hundreds of sites, it might be worth considering more specialised software that can handle larger datasets more efficiently, but for most businesses, the Microsoft suite offers a great place to start.
Melinda: We can’t afford to hire a data specialist. Who would be the best person in an existing organisation to take on this task? How much time per week/month should we expect them to dedicate to data management?
Mariana: If you don’t have the budget for a dedicated data specialist, the best person to manage data would be someone already involved in operations, finance, or IT, who works closely with the information you need to analyse. For instance, if you’re tracking energy consumption or carbon emissions, someone in facilities management or finance might be a natural fit since they’re already handling related data, like electricity bills and operational costs.
In addition, it could be beneficial to form a small data management committee. This group should include representatives from key areas that manage the most important data. Together, they can draft a data management and quality standard that outlines the minimum requirements for data collection, consistency, and quality across the organisation. This way, everyone is aligned on the goals of collecting data and understands the standards needed to ensure it is reliable and actionable.
As for the time commitment, estimating it precisely is challenging because it depends on the size of the task or goal and the state of the data. If we’re talking about a small company looking to track energy usage and carbon emissions for a single building or operation, a rough estimate would be 2 to 3 hours per week. This will vary depending on how frequently the data is collected (e.g. monthly vs daily) and how much effort is needed to clean and process the raw data. If the data is already well-organised, the time spent will be less, but if significant manual work is required to clean and structure the data, the time commitment could increase.
We hope this has helped Brad and Melinda, and anyone else reading this. Let us know if anything else springs to mind, or visit the Newsroom page on our website to read some of our older articles.
