Welcome to another episode of data digest. In this episode, you will learn how best to build a data analytics portfolio.

💡 Data digest is a series of webinars that Resagratia hosts fortnightly where data professionals in visualization, analytics, management, etc., share their knowledge and experience with the audience.

Instructor: Fatai Sanni (LinkedIn)

What is a data analytics portfolio?

A data analytics portfolio is an essential tool in a hiring process that showcases the skills you hone and the experience that you have gathered over time. Usually, hiring managers tend to request candidates’ portfolios along with their curriculum vitae to have an idea of the kinds of projects, skills, and experience the candidates possess.

A portfolio can be in form of a repository, website, blog, or document. It is an easy way to promote, yourself, your brand, or your business. It helps to increase your visibility to clients, and managers. You can think of a portfolio as a digital business card that makes people connect with you from anywhere.

Importance of a portfolio

When you are aware of the benefits of something, you tend to appreciate it more. Let’s go through the importance of a portfolio:

1. It stands you apart

Amidst all the pools of challenges, a great portfolio makes you stand apart from the rest of the candidates during a job hiring process.

2. It serves as a repository

A portfolio serves as the repository for all the projects that you have worked on; from data visualization, machine learning, and artificial intelligence. It makes it such that all of these projects are accessible from a single place.

3. It provides a great first impression

It is a very good way of creating an amazing first impression of you. It helps you break the ice and show your technical competence to your prospective employers in a short time.

4. It increases your visibility and online presence

Your portfolio will help increase your reach and also inform them of what you are capable of doing.

5. It serves as a digital CV

In addition to your LinkedIn, and other social media platforms, your portfolio serves as a digital CV that showcases all the work that you have done to your recruiters in a single click.

6. It increases your chances of getting hired

There are instances where people get a job by sharing with their employers or network the portfolio they have built without interview or qualification exams.

What makes a cv different from a portfolio?

Most times, people confuse a cv to be the same as a portfolio.

A cv contains your career history including your educational background and achievements, while a portfolio shows your actual technical competence. For instance, as a photographer, your cv will show the companies you have worked with, your educational background, and so on but it is your portfolio that will display the images, and pictures that you have captured over time.

Who really needs a portfolio?

Different professionals make use of portfolios to market themselves. Some of them are:

1. Marketer (print, design)
2. Web developer
3. Graphics designer
4. Data analyst
5. Creatives (photography, video, etc)
6. Technical writer

What should be included in the portfolio

Each profession has its own peculiarity when it comes to portfolio designs. While most professionals like UI/UX designers, web developers, and graphics designers pay careful attention to the outlook of their portfolios, data professionals’ portfolios do not require a fancy interface (it doesn't mean that you can't add effects to your portfolio if you want) which makes it easy for them to leverage on existing platforms.

Note that there is no hard and fast rule as to what should be included in a portfolio. However, there are some basic things that should be found in an average portfolio.

1. Name: This is your official name which includes your surname, first name, and other names you bear.
2. About me: This is an important section in the portfolio. This can be in one or two paragraphs that describe who you are, what you do, and what your interests are.
3. Picture: You should have a professional picture of yourself on the front page of your portfolio.
4. Case study projects: This is the essential part of the portfolio that shows what kinds of projects you have worked on.
5. GitHub link: Your GitHub link basically shows your GitHub account to your recruiters
6. Social media links: You can include your LinkedIn account, medium account, and Twitter account. Ensure not to include your Facebook account as it is not necessary for you as a data analyst
7. Clients’ testimonials: This is an optional section. It is applicable in case you have done some freelancing work or you have volunteered for some projects. This gives you some credibility to work with other clients if they know that you have worked well with similar clients before.

Top platforms to host your portfolio

There are a lot of platforms that you can use in building your portfolio. You don’t necessarily have to code your portfolio from scratch. All you need to do is to have a repository for all your projects and include them in your portfolio.

Some of the platforms that you can use to host the projects you have worked on include:

1. Disha
2. Linktree
3. Microsoft PowerPoint
4. Microsoft Excel
5. Microsoft Power BI
6. Microsoft Word
7. WordPress
8. GitHub pages

There are a lot of these platforms which are mostly free. All you need do is be creative in the way you design the portfolio. You can as well take your portfolio to the next level by procuring a domain name that displays your works once a search is made online using your name.

Where to get datasets

A lot of people ask where they can get datasets to build projects that will be included in their portfolio. There are a lot of sources where you can find datasets that you can work on. Some of the free sources include:

1. Kaggle
2. AWS dataset
3. Google dataset
4. Azure dataset
5. Web scraping
6. Maven analytics
7. Onyx dataset
8. Local businesses
9. Freelance projects

Local businesses and freelance projects are two distinct sources of datasets that are recommended for building an exceptional portfolio. These sources have datasets that are mostly in their raw form filled with lots of errors that need to be corrected. By doing so, your expertise increases as compared to the other sources that have refined datasets.

What kinds of projects should I build?

In building a project, it is important to be able to answer the following questions:

1. What exactly are you trying to achieve from this project?
2. What problem are you trying to solve?
3. What approach will you adopt in solving the problem?
4. Which tools will you use?

There is actually no hard and fast rule on the types and numbers of projects to include in your portfolio. It all depends on the projects that you can handle, find interesting, and have passion for. Here is a list of projects that you can work on.

1. SQL projects
2. Machine learning projects
3. Statistical analysis projects
4. Data visualization projects
5. R projects
6. Python projects
7. Programming projects

The life cycle of a data analysis project

Data analysis goes beyond just building visuals by drag and drop. There is a defined process that you have to go through. the earlier you understand this concept, the faster you thrive in the industry.

Let’s take a look at the processes involved in building an end-to-end data analysis project

Business issue understanding:

This is the first step. It involves that you:

1. define business objectives i.e. the goals of carrying out the data analysis and the business problem that you want to solve,
2. gather the required information necessary to facilitate the project,
3. determine appropriate analysis method and tools,
4. clarify the scope of work, and
5. identify the deliverables.

Data understanding

This is the second step and it entails that you:

1. collect initial data,
2. identify data requirements,
3. determine data availability, and
4. explore data and characteristics

Data preparation:

In this step, you will

1. gather data from multiple sources,
2. cleanse the data,
3. format the data, and
4. blend the data

Exploratory analysis and modeling:

This step requires that you:

1. develop a methodology,
2. determine the important variables,
3. build a model, and
4. assess the model

Validation:

This is the fifth step that involves you:

1. evaluate the results,
2. review the adopted process,
3. determine the next steps (if the results are valid you proceed to visualization and presentation of the results and if not you repeat the steps a-d)

Visualization and presentation:

This is the last step. It requires that:

1. you communicate the results to the audience in an appealing form,
2. determine the best method to present insights based on analysis and audience,
3. craft a compelling story, and
4. make recommendations.

You can check these links for inspiration on building your portfolio.

• Fatai Sanni | GitHub
• Sanni Fatai | Linktree

We do hope that you found this episode of the Data digest series exciting and insightful, for more access to such quality content, kindly sign up to the Resa platform by clicking here.

Thanks for learning with Us!