The world’s leading publication for data science, AI, and ML professionals.

Starting Out With Data Analytics to Boost Your Business

How to get started in the obscure and complicated world of data analytics. Hint: It’s easier than you might think.


Photo by Lukas Blazek on Unsplash
Photo by Lukas Blazek on Unsplash

Data analytics can be a very powerful tool in your arsenal. Whether you own a small Business or if you’re at a larger company. Using data analytics can help you improve your profitability, be more streamlined, and even improve your products. If you are interested in learning more about how analytics can help your workflow, here is another article I wrote about the topic. But for now, let’s imagine that you’ve already taken the decision, what are some of the steps you should take in order to get started?

Do I really need to learn how to code?

The obvious answer to this question is simply no. There are a few things you need to learn in order to be able to analyze data, but Coding is not one of them. Sure, learning some useful languages like R and Python will help you massively, but they are not prerequisites.

Why do data analysts code?

If coding is not required, then why do so many data analyst learn it? After all, if you go to a university, whether that be at the undergraduate or postgraduate level, you will be exposed to a lot of coding classes. In these you mostly learn R and Python. The reason many data scientist prefer to code is that it allows for more flexibility and speed.

Programming languages have a hierarchy, to simplify it, the more readable a language is, the more difficult it is for a computer to run it. This also applies when comparing Python to something like Excel. You can do the same work with both of them, but Excel will be slower when working with large data. If you are comfortable using Python, you will likely prefer it over Excel for the speed differential, and the fact that you are not constrained by Excel functions. Python has a huge selection of packages that can do everything from data analysis (Numpy, Pandas), to creating a website (Flask, Django), to creating 2D games (Pygame). Programming languages are also more scalable. The scalability and future proofing are important when you know that there is more data coming in the future. This allows you to do the work once, and then edit it later to accommodate the influx of data.

How can you get around Learning how to code

Even though I just pointed out some of the benefits of coding, it’s not required. If your end goal is to become a full fledged data analyst, then yes, you are going to need to learn these languages. But, if you are looking to perform some analysis on a smaller scale, there are alternatives.

Excel and Tableau are two good examples of that. There is a learning curve to using both of these programs, but they are great in handling medium sized datasets. Excel is great for running calculations, gathering insights. And Tableau will serve you well if you are looking to visualize your data and show it to other people. Analysts everywhere still use both of these programs, and they are taught at many universities.

Where to start?

Data analysis is storytelling using numbers. Your main goal is to be able to derive useful information from the numbers that your business produces.

Choose your weapon

The first thing you need to do is to pick your tools. You need to choose a program, (Excel, R, Python, Tableau…) and start learning the basics. An important thing to keep in mind here. At the end of the day, these are all tools that do the same basic things. Unless you need to analyze a very specific dataset, you really can’t go with either of them. The two main considerations you should keep in mind are learning curve, and personal preference. How many hours are you willing to put into picking up this skill? And which one do you feel most comfortable using? The best way to never get started is to think that you have to use the perfect tool, when in reality all of them have the same basic functions. Do some quick research, and then commit to a decision.

Figure out your goals

Committing to a project when you have a goal in mind is more likely to help you stick with it in the long run. So before you get started on this journey, try thinking about what you are looking to learn from it. Is it more insights about your product? More information about your suppliers? Your target market? This will also help you formulate a more robust plan and will yield much better results.

Finding inspiration

Finding inspiration for your storytelling is very important. As you get started, you’re going to ask yourself, what information is there to unlock in this data? This is a difficult question to answer, and the truth is, it differs from depending on business, industry, company size, and many other factors. One way of getting started is to look at what others have done online. Doing some research and finding out what analysts at similar companies are doing can point you in the right direction. But at the end of the day, you are in the best position to determine what you need to analyze. No one outside your business can tell you exactly what you need, so trust your instinct.

Gathering data

This can also be a daunting part of the journey. My advice, start from the inside out. Look at all the records within the different departments in your company, find where the numbers are, and start taking a look at them. If that doesn’t work, then work your way out, see if you can find market data that can be relevant to what you are trying to achieve, if that isn’t available, then approximate. You can check is there is informatoin about customers in another city, state, or even country, and start thinking to yourself how that can apply to your market. Again, trust your instincts, if something feels off about the approximation, then account for it. For example if the market you are looking at has a much higher median household income than yours, then keep that in mind when interpreting the data.

Practice

The best way to get started, or improve, as a data analyst is to get some practice. This doesn’t have to be very extensive, a few hours working on a project can go a long way. A good way of getting some practice is to do look at the more famous datasets on a site like Kaggle, download them, and start working on them. The benefit of getting some practice on these datasets is that data scientists of all levels have worked on them before. This provides you with a great reference to be able to go back to and compare your own work once you’re done. You will get an idea of what you did right and what you might have missed.

Conclusion

Data analytics is limitless. You can take is as far as you like, but the important thing is to get started with it. Once you do get started and become more comfortable with the field, you can push it further, but again, you have to take the first steps.


Related Articles