5 Passive Income-Generating Products You Can Develop in Your Spare Time as a Data Analyst | by Madison Hunter | Sep, 2022

Finding your niche in these products will help you generate extra income on the side.

Since the data science craze began, everyone and their cat has been developing products to sell to generate passive income.

And when those people begin advising on how other data analysts can do the same, it’s always the same: make a course, write a book, start a Youtube channel.

Except now, there are thousands of data science courses on Udemy, there are over 40,000 results when you search “data science” on Amazon looking for an ebook, and there are more data science videos on Youtube than I care to count.

The advice to make a course, write a book, or start a Youtube channel to generate passive income as a data analyst, is so 10 years ago. Nowadays, you need to be proactive, creative, and unique in figuring out how you plan on standing out from the crowd and developing products that people actually want to buy.

The problem with the data science product market is that it’s difficult to sift through the gravel to get to the gold. The fact is that not everyone can teach, not everyone can write, and not everyone can create interesting videos. And yet people still do it. This means that, even if you’re extremely talented in any of these categories, your ideal customers will have to wade through a sea of other products before they get to yours — if they even get to yours.

Therefore, you need to think five steps in advance of your target market. What are they really looking for? What do people need? What can I provide this market that no one else can?

The trick here is to niche down from the beginning. Don’t develop a course on how to learn Python, don’t write an ebook on how to become a data analyst, and don’t create Youtube videos teaching people how to use Excel and Tableau to analyze data. These ideas have been beaten to death and won’t gain you any extra income.

But how do you figure out what you’re good at? For example, the articles I publish on Towards Data Science are geared toward people looking to make a transition into data science, or those who have just entered the field. I focus on writing articles about how to get into data science, the tricks you need to stand out against other competing candidates, how to write better code, how to study data science, and more. These articles are focused on the beginner steps towards becoming a data scientist, though they may prove useful for more experienced individuals in the profession. I know that I wouldn’t make a dime if I was writing more complex articles on more technical aspects of data science, but I do know I can help people just starting out.

So, with this in mind, it’s time to look at five simple projects you can develop in your spare time as a data analyst. These projects can be sold to generate passive income without you having to develop unique products for new customers each time. These projects focus on developing plug-and-play tools that people can buy and then modify to their heart’s content. Not only does this make the job easy for you, but it also opens the door if a client is looking for someone who could modify the product to suit their needs — which could be you.

Tableau templates generally fall into one of two categories: standalone visuals or dashboards.

As most data analysts will tell you, visuals are often the icing on the cake when it comes to presenting an analysis. Much of the verbiage or summary will often be discarded or ignored if the visuals provide a clear glimpse into the heart of the problem.

A quick search on Fiverr found that tableau dashboards are sold from anywhere between $6 to $800, and data visualizations are sold within a similar range. The way to stand out in such a broad market is to develop templates that both appeals to a particular set of customers and that are easily customizable by a customer once they’ve received the file.

For example, your templates could be geared toward the oil and gas sector which you know has a particular preference for a certain type of dashboard. Or you could focus your visualizations to function better for small businesses that need simple graphics to describe their data.

Time should be spent carefully constructing and commenting these files to ensure that colors can easily be changed, labels can be modified, and the amount of data represented can be increased or decreased accordingly.

An extension of the previous product is to offer data-driven dashboards as a service that customers access via a monthly or yearly subscription.

This is a slightly more involved service that will require you to work closely with a client to develop and maintain a dashboard (perhaps using an existing template) that provides to-the-minute information for a company in an easy-to-understand format.

Essentially, you would work as a contractor to provide dashboard services for the term of the subscription and would entail a hands-on approach. Bug fixes, data updates, analysis code, and more may be involved depending on the level of service you would like to provide (pro tip: this could be determined using a tiered subscription that provides different levels of service at different pricing).

When you’re deeply involved in the data science world where everyone is a statistician, master programmer, data engineer, and graphic designer, it can come as a surprise when you surface and realize that not everyone knows how to code. What’s even more surprising is that people are often more willing to shell out money to purchase code instead of writing it themselves.

This is where you can provide a wealth of expertise by simply selling useful Python code snippets. The trick is to write code that you can’t find elsewhere.

For example, command-line Python tools can be found for free and probably won’t generate a lot of passive income.

However, highly specialized snippets, such as web crawling code that targets a specific industry, could provide a lot of bang for its buck. Alternatively, you could sell an API (a piece of go-between code that allows two applications to talk to each other). Data-related APIs could help a company report data, sync dashboard updates, and more.

If you’ve ever seen an Excel file developed by a person with no data sense, you would know that it’s a hot mess (I know that while reading this you had one Excel sheet in mind that you’ve seen in the past).

Believe it or not, people are selling Excel spreadsheets on Fiverr for anywhere between $6 to $240. Sometimes, it’s just easier to have someone else develop an Excel sheet than try to do it yourself.

To stand out amongst the 2,000+ Excel spreadsheet services available, it’s vital that your spreadsheets service a particular niche market. As mentioned before, this could be small businesses, freelancing individuals, or unique sectors within an industry. Furthermore, these spreadsheets should have an additional feature that takes them one step further. For example, they could have a certain level of automation or perhaps a simple dashboard that presents the data in a fancy yet easy-to-read way.

In short, if people are willing to buy Excel sheets, you want to make it a special experience every time they use it.

The most complex of our product options is a machine learning model that you sell through a subscription service or as a file.

Stream.ML, a company based out of Alberta, Canada, has developed a platform that allows anyone to build and sell their own machine learning model. At its most basic level, the user uploads and tags their data and then trains their model — simple as that! Model hosting and testing are handled by Stream so you can focus on building a model that yields results, as well as some passive income.

Alternatively, machine learning models can be built by hand and sold on freelancing websites such as Fiverr. Machine learning services are currently being sold for anywhere from $13 to $400.

As mentioned above, to stand out against 7,000+ machine learning services on Fiverr, your model will need to target a niche market. Perhaps you want to develop a model that helps small businesses estimate how many website visitors will actually buy products, or you want to build a model that gives a scientist a template that can be modified to begin recognizing particular plant species based on the image data supplied.

The key is to find a problem that your niche market has and solve it using a machine learning model.

Finding your niche in these products will help you generate extra income on the side.

Since the data science craze began, everyone and their cat has been developing products to sell to generate passive income.

And when those people begin advising on how other data analysts can do the same, it’s always the same: make a course, write a book, start a Youtube channel.

Except now, there are thousands of data science courses on Udemy, there are over 40,000 results when you search “data science” on Amazon looking for an ebook, and there are more data science videos on Youtube than I care to count.

The advice to make a course, write a book, or start a Youtube channel to generate passive income as a data analyst, is so 10 years ago. Nowadays, you need to be proactive, creative, and unique in figuring out how you plan on standing out from the crowd and developing products that people actually want to buy.

The problem with the data science product market is that it’s difficult to sift through the gravel to get to the gold. The fact is that not everyone can teach, not everyone can write, and not everyone can create interesting videos. And yet people still do it. This means that, even if you’re extremely talented in any of these categories, your ideal customers will have to wade through a sea of other products before they get to yours — if they even get to yours.

Therefore, you need to think five steps in advance of your target market. What are they really looking for? What do people need? What can I provide this market that no one else can?

The trick here is to niche down from the beginning. Don’t develop a course on how to learn Python, don’t write an ebook on how to become a data analyst, and don’t create Youtube videos teaching people how to use Excel and Tableau to analyze data. These ideas have been beaten to death and won’t gain you any extra income.

But how do you figure out what you’re good at? For example, the articles I publish on Towards Data Science are geared toward people looking to make a transition into data science, or those who have just entered the field. I focus on writing articles about how to get into data science, the tricks you need to stand out against other competing candidates, how to write better code, how to study data science, and more. These articles are focused on the beginner steps towards becoming a data scientist, though they may prove useful for more experienced individuals in the profession. I know that I wouldn’t make a dime if I was writing more complex articles on more technical aspects of data science, but I do know I can help people just starting out.

So, with this in mind, it’s time to look at five simple projects you can develop in your spare time as a data analyst. These projects can be sold to generate passive income without you having to develop unique products for new customers each time. These projects focus on developing plug-and-play tools that people can buy and then modify to their heart’s content. Not only does this make the job easy for you, but it also opens the door if a client is looking for someone who could modify the product to suit their needs — which could be you.

Tableau templates generally fall into one of two categories: standalone visuals or dashboards.

As most data analysts will tell you, visuals are often the icing on the cake when it comes to presenting an analysis. Much of the verbiage or summary will often be discarded or ignored if the visuals provide a clear glimpse into the heart of the problem.

A quick search on Fiverr found that tableau dashboards are sold from anywhere between $6 to $800, and data visualizations are sold within a similar range. The way to stand out in such a broad market is to develop templates that both appeals to a particular set of customers and that are easily customizable by a customer once they’ve received the file.

For example, your templates could be geared toward the oil and gas sector which you know has a particular preference for a certain type of dashboard. Or you could focus your visualizations to function better for small businesses that need simple graphics to describe their data.

Time should be spent carefully constructing and commenting these files to ensure that colors can easily be changed, labels can be modified, and the amount of data represented can be increased or decreased accordingly.

An extension of the previous product is to offer data-driven dashboards as a service that customers access via a monthly or yearly subscription.

This is a slightly more involved service that will require you to work closely with a client to develop and maintain a dashboard (perhaps using an existing template) that provides to-the-minute information for a company in an easy-to-understand format.

Essentially, you would work as a contractor to provide dashboard services for the term of the subscription and would entail a hands-on approach. Bug fixes, data updates, analysis code, and more may be involved depending on the level of service you would like to provide (pro tip: this could be determined using a tiered subscription that provides different levels of service at different pricing).

When you’re deeply involved in the data science world where everyone is a statistician, master programmer, data engineer, and graphic designer, it can come as a surprise when you surface and realize that not everyone knows how to code. What’s even more surprising is that people are often more willing to shell out money to purchase code instead of writing it themselves.

This is where you can provide a wealth of expertise by simply selling useful Python code snippets. The trick is to write code that you can’t find elsewhere.

For example, command-line Python tools can be found for free and probably won’t generate a lot of passive income.

However, highly specialized snippets, such as web crawling code that targets a specific industry, could provide a lot of bang for its buck. Alternatively, you could sell an API (a piece of go-between code that allows two applications to talk to each other). Data-related APIs could help a company report data, sync dashboard updates, and more.

If you’ve ever seen an Excel file developed by a person with no data sense, you would know that it’s a hot mess (I know that while reading this you had one Excel sheet in mind that you’ve seen in the past).

Believe it or not, people are selling Excel spreadsheets on Fiverr for anywhere between $6 to $240. Sometimes, it’s just easier to have someone else develop an Excel sheet than try to do it yourself.

To stand out amongst the 2,000+ Excel spreadsheet services available, it’s vital that your spreadsheets service a particular niche market. As mentioned before, this could be small businesses, freelancing individuals, or unique sectors within an industry. Furthermore, these spreadsheets should have an additional feature that takes them one step further. For example, they could have a certain level of automation or perhaps a simple dashboard that presents the data in a fancy yet easy-to-read way.

In short, if people are willing to buy Excel sheets, you want to make it a special experience every time they use it.

The most complex of our product options is a machine learning model that you sell through a subscription service or as a file.

Stream.ML, a company based out of Alberta, Canada, has developed a platform that allows anyone to build and sell their own machine learning model. At its most basic level, the user uploads and tags their data and then trains their model — simple as that! Model hosting and testing are handled by Stream so you can focus on building a model that yields results, as well as some passive income.

Alternatively, machine learning models can be built by hand and sold on freelancing websites such as Fiverr. Machine learning services are currently being sold for anywhere from $13 to $400.

As mentioned above, to stand out against 7,000+ machine learning services on Fiverr, your model will need to target a niche market. Perhaps you want to develop a model that helps small businesses estimate how many website visitors will actually buy products, or you want to build a model that gives a scientist a template that can be modified to begin recognizing particular plant species based on the image data supplied.

The key is to find a problem that your niche market has and solve it using a machine learning model.