Are you a fan of Tableau? How about Adobe Analytics? Both? Chances are you have interacted with both tools, which means you have also likely tried to figure out how to get Adobe data into Tableau. If you have faced challenges trying to get Adobe Analytics data into Tableau, this article is for you!
If you are not familiar with why this process is challenging, let me give you a quick rundown:
Tableau is a software that connects to data sources in many ways, one of them being through API’s (Application programming interface). Google Analytics, for instance, has an API that allows Tableau to get data from GA on a consistent basis. However, for Tableau to use this API, there must be a Tableau connector! And guess what, the only people who build Tableau connectors, are people in Tableau.
Now, this isn’t completely true, because you can attempt to build your own connector by using the Web Data Connector in Tableau. However, building this requires a complex system of databases and other connectors.
One does not casually build a database connector that connects to the Web Data Connector in Tableau.
Therefore, this article will outline the main ways in which you can get Adobe data into Tableau, as well as the pros and cons for each one. If I missed one, please leave me a comment below so I can add it to the list.
The different solutions for this process can be split into two main categories: local and hosted. Local solutions are processes that primarily use your local computer to process and store data, while hosted, on the other hand, uses external servers for the storage and/or processing of data.
I will walk through the local options first, then the hosted ones.
1. Adobe Report Builder – Local
The first and most simple option is Report Builder, which is a tool that integrates with Excel, allowing you to refresh data straight from Excel. The general idea is that you set up everything the way you want in Report Builder, then hook up that file to Tableau. Then, whenever you want new data, you can go into the Excel file and simply refresh the data.
*Note – you can also schedule automatic refreshes in Report Builder so, theoretically, you don’t have to touch the file after initial setup.
Dylan Renzulli and Thanh-Trang Hoang-Le at Mass Media have put together a great resource for accomplishing this task. They also give you a few different options, dependent on the tools you’re using. Check it out here.
- More automated than some other options
- You have control over every step
- Still requires a decent amount of manual work
- Requires you to know Report Builder (it’s not too difficult to learn).
2. Adobe Data Warehouse – Local
Another basic way to get Adobe data into Tableau is though the Adobe Data Warehouse. With the Data Warehouse, you have a multitude of options.
First, you can choose to get an excel file or a TDE (Tableau Data Extract). Either option is fine, and to be honest, the TDE doesn’t really save you much time. You still have to open the file in Tableau, which is pretty much what you do with an excel file. Here is a link to learn more about TDE’s.
Your second decision is about delivery options: email or FTP (File Transfer Protocol). Email is the simplest option, as you just type in your email and receive the file. However, if your dataset is too big, you will need to go with the FTP (if it’s too big, you’ll receive an email from DW after you submit the request, saying the file is too large to be transferred over email).
If you don’t have an FTP server, you have two options: set up one for yourself, or get one through Adobe. Adobe does offer FTP accounts for Adobe Analytics users but is a bit of work. You likely have to go through customer care to get your account set up. Here’s a link outlining that process. Setting up your own FTP server shouldn’t be too difficult though, here’s an article outlining the process for mac’s, and here’s one for pc’s.
- This process is simple.
- You have clear control over the data, allowing you to narrow data requests, which therefore decreases the time it takes for you to receive your data.
- It’s not live. You must do this every time you want new data.
- It can’t be refreshed.
- It’s not live.
- It’s a very manual process
- It’s literally the furthest thing from being live.
3. Data Extract – Local
The third option in the local category is another Adobe service: Data Extract. Data Extract is similar to Data Warehouse, where you select data to extract from Adobe Analytics. However, Data Extract is done within the reports section of Adobe. When in a report, you can click on “more” in the top nav, then select Data Extract.
It’s important to note that Data Extract is limited to 50,000 rows of data, whereas Data Warehouse is not limited. Therefore, Data Extract is best used when working with small amounts of data. Data Extract can also work with email or FTP, which is the same as Data Warehouse.
- Very fast
- Easy to use and understand
- Limited on the amount of data you can pull
- Still a manual process
4. Data Warehouse FTP Script – Hosted
This solution is a basically Data Warehouse on steroids, so hold onto your seats! The normal way to use Data Warehouse is to go into the tool, set up your report, and receive it through email or FTP. But the first nice thing about DW is that you can schedule Data Warehouse reports. So now we don’t have to go into DW every week!
But how do we automatically get these reports into Tableau? A custom script!
When you schedule reports from DW, it’s important you schedule these to be sent through FTP. By doing this, you now have a data file sitting on your own server. From here, you write a script to fetch this data and stick it in your own database. Afterward, you simply connect the database to Tableau and voila, your data is in Tableau!
*Note – make sure the database technology you use can connect with Tableau, otherwise you have the same problem as Adobe data.
- This process is mostly automated, therefore requiring little work after it’s set up
- This is probably the cheapest of the hosted solutions as you will most likely store less data with this method than other methods.
- Data Warehouse data is pre-calculated, therefore limiting your ability to manipulate your data.
- You’ll probably have to hire a developer to write the script to pull data from your FTP server and put it into a database.
- It’s quite a hassle to make changes to current data pulls
If you’re using Adobe Analytics, you are likely collecting data from a website or a mobile app. The cool thing about this fact is you have the ability to add anything you want to your website/mobile app.
In this situation, what you want to add is a snippet of code which “piggybacks” off the Adobe implementation and sends data to a second source. The Adobe Analytics implementation will figure out what data to collect, then send it to Adobe servers. Once it decides what data to collect, you add some code to also send this data to your own servers.
This essentially gives you a replica of Adobe clickstream data on your own servers, which means you can do whatever you want with this data. From here, you simply hook up your database to Tableau and bam!, you’ve got a live connection to Adobe data. However, it’s important to note that aggregating and calculating raw data is not easy. You will need someone who intimately knows how Adobe processes their data in order to process your own hit-level data.
- This process is about as automated as you can get (at least for now)
- Your data is unprocessed, meaning you have complete control over your data. You decide how you calculate metrics.
- You will most absolutely need to hire technical people to oversee this process. You need a developer to write the code, and a database expert to make sure you are collecting and processing data correctly.
- Your data is unprocessed. This is also a con because it means you will be storing a lot of data! Get ready to sign up for a major hosting service.
6. Data Feed – Hosted
The last solution is only a slight variation of the second hosted solution, mostly in the way that it sends data. Instead of using code on your website to collect data, you can get unprocessed data straight from Adobe using the Data Feed.
The data feed lets you automatically pull hit-level data straight from Adobe, based upon either hourly or daily cycles. This data is fed into a database, and then from there fed into Tableau through a connector (I told you it’s basically the same).
- The data unprocessed, meaning you have complete control over how you calculate and aggregate your data
- This is probably the most straightforward of the 3 hosted options.
- You are dealing with large amounts of data, meaning you will need to invest in a fairly large database structure. This will cost money
- You will also need to hire people to build and manage this process. This also costs money.
As you have probably figured out, the two categories of solutions (local and hosted) have their own set of issues. Most local options are fairly manual and require multiple steps to get Adobe data into Tableau. Hosted options, on the other hand, require a fairly sizable financial investment and can take time to set up. These are also more complicated, so there is more room for error on the hosted side.
Click here if you would like to learn more about our services. We would love to make your organization more efficient. And as previously stated, please comment below if we missed anything. Thanks for reading!