Data Analysis in a Report

An analysis is not worth much unless it is understandable and relevant to a recipient. If an analysis is only readable as lines of code, or as complicated sequences of numbers, we may not be able to make use of the data we have analysed.
We have previously looked at how we search for relationships and insights in data. Now, let’s look at how the findings can be presented.

Reports

Most analyses are presented in some form of report.
When done correctly, a data-based report is a reliable source of information. In the old days, decisions may have been based on experience, “common sense”, divine revelations and gut feelings. Today, it is data that counts.
Neither a master’s thesis, an investigation for political measures, a survey conducted by a committee, nor a business analysis has much impact today without arguments supported by data.

What can we say with a report?

Let’s say that you have been tasked with designing a report on how Norwegian glaciers have changed since the 60s. The data has been collected, processed, quality-checked and analysed. How do you proceed? Let’s break it down into a few simple steps.
This is not a recipe for how to write a scientific report, but a simple and overarching example of how data and data analysis can be used to formulate a message.

Analysis of business insight

The kind of software referred to as Business Intelligence (BI) helps us to analyse business data. Such software can also be used to present analyses in visual forms: as standalone graphs, tables, diagrams—or as combined dashboards.
There are many different BI tools, such as Tableau and Microsoft Power BI, mentioned earlier in this chapter. Let’s look at an example of how one would work with analysis and visualisation of business data in practice.
Let’s say you work in the finance department of a grocery store with several hundred stores around the country. You use a tool like Power BI to monitor various KPIs and revenue streams in a dashboard.

Fact

KPIs

A KPI (Key Performance Indicator) is a key figure for performance measurement in a business. KPIs are used to measure and evaluate the business’s performance against defined goals. They can be used in different areas, such as sales, productivity, quality, customer satisfaction, and profitability. In a grocery store, examples of KPIs might be:
  • Sales revenue per square metre of shop floor space
  • Sales of frozen food compared to the previous quarter
  • Proportion of turnover from sale items
  • Proportion of turnover from local producers
In Power BI, these KPIs have been predefined, and revenue streams are updated in real-time. This way, you can visually see how you are doing against the KPIs, such as what items bring in the most profit.
The graphs show that sales of frozen food have increased in recent months. The same goes for items on sale, but they do not earn enough to cover the purchase price.
You know that the rental fee for locations close to the city will increase next year, and the income will no longer exceed the rental expenses in 14 of the stores you are responsible for. At the same time, this is where the sales of frozen food are the highest. It’s up to you to make further decisions: Should you close down the stores in the city? Or should you investigate whether there are other measures for increasing profitability in the city?
Making such data-based decisions—with the help of analysis tools and dashboards—is a significant part of the daily routine for many who work with data. But for many of us, it is difficult to relate to just numbers and tables. For this reason, it’s a good idea to also know a bit about visualisation.