Recipe to become a Data Analyst-Analyze your data at every step
Welcome back to Episode 5
Recap: In my earlier blogs, we jumped right into the world of data analytics. We learned why structured thinking was key to solving problems, explored the best ways to collect and store our data and gained an understanding of clean data along with data integrity. We’ve also identified how to ask the right questions and learned to clean data. Now we’ll take our skills to the next level. In this blog, we’ll learn how to come up with clear and objective answers to any data question we encounter.
As a reminder 6 phases of data analysis are:
What is analysis?
Basically, analysis is the process used to make sense of the data collected. The goal is to identify trends and relationships within the data so that we can accurately answer the question we’re asking.
There are 4 phases of analysis:
- Organize data
- Format and adjust data
- Get input from others
- Transform data by observing relationships between data points and making calculations.
It’s critical to keep our data organized throughout our analysis. How our data is classified and structured will greatly impact our findings. And once we know how our data is organized, we’ll be able to capture/collect the information we need and make informed decisions. Once we have the data organized and formatted, we’ll be ready to sort and filter it to find the insights we need.
Sorting and filtering are two ways we can keep things organized and adjust data to work with it. For example, a filter can help you find errors or outliers so you can fix or flag them before your analysis.
The bottom line is that it’s important to have our data in the correct format. So always be prepared to adjust, no matter how far we are into our analysis.
A big piece of being an analyst is troubleshooting and problem-solving. You’re as good of an analyst as your ability to ask the right questions, which is why we’ll spend some time learning about problem-solving strategies you can use during analysis.
When you’re dealing with data, you can encounter multiple inconsistencies and hurdles before you reach your end goal.
1. Data Conversion:
Let’s say you wanted to sort the movies in the spreadsheet by the most recent date. If the spreadsheet cast date as strings instead of DateTime, it might sort them alphabetically. And you would end up giving a wrong list of the 20 most recent movies, It’s also possible that your datasets contain inconsistent units of measurement that you’ll need to convert. Like, say, a table that includes both US dollars and English pounds. Incorrectly formatted data can lead to time-consuming mistakes in your analysis and might end up affecting your stakeholders’ decision-making. But taking the time early on to convert and format your data can help you avoid that.
2. Missing Data:
Imagine you have multiple datasets of an ice cream shop. One is the customer metadata and their favourite ice cream codes and another is the master table which consists of ice cream metadata according to their codes. And you have been asked to find out the most popular flavour.
Worry not, my most fav tool at your rescue
To find out your answers, you have to join the two tables. Being able to combine multiple pieces of data can give you new ways to organize and analyze data. There are a lot of different tools to help you do that.
- Excel functions
- SQL Joins
- Python Dataframe joins etc.
Data analysts spend a lot of time problem-solving, and that means there are going to be times when you get stuck, but the trick is knowing what to do when that happens.
Asking other people about a problem you’re having can help you find new solutions that can move a project forward. It’s always a good idea to reach out to your peers and mentors. Your team members have valuable knowledge and insight that can help you find the solution you need to get unstuck. Sometimes we spend a lot of time spinning our wheels saying, “I can do this myself,” but we can be way more productive if we engage with other people, find new resources to lean on and try to get as many voices as we can be involved.
I would like to give a big shoutout to all my team members with whom I brainstorm and discuss and come out of the most difficult situations with the best ideas. And if in some cases all are struggling there’s definitely someone else with the same problem asking the same questions online. Knowing how to find solutions online is an incredibly valuable problem-solving tool for data analysis. Using the thinking skills we’ve learned throughout, the right terms, and your understanding of different analysis tools, we’ll get you ready for actual searching for answers online.
Finally comes the Data Transformation: Calculations are one of the more common tasks that data analysts perform during an analysis. In this part, you explore formulas, functions, pivot tables and SQL queries. All of these are used in data calculations.
Look at the scoreboard from IPL with top teams and their score. How is their performance measured?
Well, this is where the heavy lifting takes place and comes into existence of North Star metrics & KPIs telling the whole story accurately gauging the success, efficiency, context and correlation.
Calculated metrics help companies understand their journey, customers and success. They stand on the most burning and important grounds changing the directions of decision makers with changes in metrics. And this is where the aha moments come in where you’re able to judge and make data-driven decisions.
As data professionals, we should play a vital role in understanding these metrics and the grounds they are built on. We are like the Head chef cooking fav meal for customers and understanding their preferences and taste(key metrics) will result in a happy customer.
Before performing Data calculations we need to understand
- Why our business is seeking a particular metric?
- What will it yield them?
- How will it help the company overall?
- Do we have the right type of data to get the required metrics or not
- What tools will be getting involved and how shall it impact the final goal?
Once you’re done cooking fav meals for your top customers, it’s time for satisfaction and happiness.
This brings us to the end of this read and below is a glimpse of your learnings for Analyze phase.
- Organizing data to begin analysis
- Formatting and adjusting our data.
- Aggregating data for analysis
- Performing data calculations.
