Welcome back, folks! Today I will talk about the process, the story, of choosing, and evolve our choice on Data Visualization tools.
Why Data Visualization?
Maybe you have a question. Why do you need to do this? Well, I am amazed if you can read something like this:
While most of us is definitely able to read something like this:
By looking at the two images, we can safely conclude that visualization is easier to understand for human brain, even though the second image doesn’t have any information on it while the first one has.
So, this is our story, on how we choose our tools. And the evolution of choice itself. Maybe after I told my story, you can also told yours later. It is really nice if we can share our experiences to each other.
Long long story. Lots of problems. We have changed our choice several times. And I think it will change again in the future.
Initial step, a giant step. With D3.js
When we were starting our effort on utilizing data. We are totally unaware about the best practices, as we start everything by ourselves. Including our choice in Data Visualization. D3.js is our first step in diving the depth of our data ocean.
- Open Source. It’s free to use!
- Has a complete set of visualization charts and graphics.
- Has a huge community
- Used by big companies. Even Google Analytics’ charts are made by this amazing piece of technology.
- Self developed. Means another thing to manage
- Hard to make a page that can be customized easily by the users
- The ease of access is limited by the developer’s skills
In conclusion, using D3.js is recommended if you want to manage your visualization by your small team, and has a small number of users. It is really recommended for getting started.
Data driven approach, not getting overwhelmed by moving to Periscopedata
As the company grows, the number of data users also increases. It is really impossible to cater every users’ needs if we stick with our old school tools. We need to move fast into something more easily customized.
Comes Periscopedata, A managed data visualization tools which has a good, slick, and clean user interface.
- Easy setup for any database you have. It supports many popular SQL databases.
- Easily customizable with many filters choices
- You only need to know SQL, and you are good to go
- Unlimited number of users. Means you don’t need to think about it.
- We can leverage our free time to get into a new data process. ETL. Which you can read it here.
- A steep price. You need to spend $1000 to get started. They billed you according to the number of rows you have. With 1 Billion rows as the first upper limit.
- ETL = a new thing to manage, since you have to manage the crontab for each ETL job. The solution for this problem is to use a managed ETL service like AWS Glue.
Too pricey. Any other Data Visualization tools cheaper than that? Comes Tableau
This is the newest thing we use. Tableau, another paid managed data visualization service. We choose this tool because it can solve the two cons of Periscopedata we found for our company. Which is the price, and ETL.
- Easy to use. Like periscopedata
- Have a built in ETL system in Tableau Prep.
- A lower starting up price. $70 for a user monthly. But have no limit on the number of rows processed. This is good for us since several of our data users are resigned. And we need to press the bill.
- Limited number of user. We need to pay more for more users. While this is not a problem for our company, this could be a problem for others.
That was the evolution of our choice on data visualization tools. Our choices are maybe not the best choices for all of us, but at least it can help you decide. There are many other data visualization tools available and you can just google them for good.
An image is definitely better than a thousand word, only if the image is informative.
Please share it if you like it!