What is our primary use case?
We are majorly using it for our procurement business area. We have created almost 50 to 60 dashboards for the entire procurement cycle. We have used it for procure-to-pay, and we have also used it for our retail business. We have a lot of petrol bunks, and we get a lot of data from them regarding sales and other things. So, procure-to-pay and retail business are the main use cases.
We've been using its latest update. It is on-premise. All our solutions are on-prem because we are in the Oil and Gas sector, and data is very critical for us. We have not yet migrated anything to the cloud.
What is most valuable?
It is very easy to build charts and drag and drop the fields that are there. It automatically identifies the dimensions and measures and makes our life easier when we need to build any dashboard. It is pretty user-friendly.
Its visualizations are good, and its features make the development process a little less time-consuming. It has an in-memory extract feature that allows us to extract data and keep it on the server, and then our users can use it quickly.
What needs improvement?
When it comes to visualizations, Tableau has a limitation as compared to Power BI. It has a limited set of visualizations. Power BI has the entire marketplace, so you can connect and import many visualizations and use them, whereas Tableau has only 10 or 15 visualizations. There should be more visualizations, and there should also be data integration with more cloud providers.
Tableau has recently launched a paid version for the documentation. So, documentation has become a little bit challenging when it comes to Tableau development because we do not have any tool to export the data out of it. It is a license-based feature that you need to purchase to prepare documentation. So, on the documentation front, for preparing clear documentation for any dashboard, it would help if we get an embedded option, rather than buying a license for each user for the documentation. To document anything, if I have to connect to each workbook and see what has been written as a formula and then document in the Word document, it is pretty time-consuming.
We have the Microsoft stack, and we are currently evaluating Power BI because Tableau has a limitation of 50 columns for a drill-down report. If we want more than 50 columns, we have found a hack, but there is no ready-made option for doing it. So, we have to use another tool in case we need a drilled report with more than 50 columns. There are many instances where users need 80 or 90 columns for their analysis, and switching between two technologies becomes a challenge. It is not a cost-effective approach for us.
Their support should be improved. We are not happy with their support. Whenever we raised queries, we were pointed to a few blogs, and we didn't get a proper solution from them.
Their licensing should also be improved. They want us to purchase a Tableau Creator license for business users, whereas Power BI Desktop is free for business users. They should come up with a basic license with one or two connectors that our business users can use for preparing their visualizations. Tableau also charges us per user for users who want the data only through email.
For how long have I used the solution?
Tableau has been in our organization for more than four years.
Buyer's Guide
Tableau
March 2023
Learn what your peers think about Tableau. Get advice and tips from experienced pros sharing their opinions. Updated: March 2023.
690,226 professionals have used our research since 2012.
What do I think about the stability of the solution?
Its performance and stability are good. Because we are not on the cloud, and it is on our internal servers, it is performing well.
What do I think about the scalability of the solution?
I have a team of three people who are Tableau developers, and they have been working with me. We have one Tableau senior developer who does the server administration, as well as major Tableau development. Two members are supporting him on small dashboard developments.
When we were given the first project, we had hired a consulting company called PWC. They had implemented Tableau for us for procure-to-pay. They had deployed one project manager, one Tableau developer, and one MSBI because the data is on SQL. After that, for our support, we hired one senior Tableau resource, and then we internally trained two people. They have been using Tableau and supporting us.
How are customer service and support?
We were not happy with their support. We did not get many solutions. Whenever we raised queries, we were diverted to a few blogs here and there.
Even for the production issue, they didn't give proper support to us. There was a lack of clarity about how to resolve the issue. They work in shifts. So, one person hands over the ticket to another person, and we again have to explain. We had also approached our sales representatives, but unfortunately, because of Tableau being acquired by Salesforce, the team had changed. We had lost all points of contact because of which it was a little chaotic to get support when we needed it the most.
Now, the first step we take is to go and search on Google about the issue and also on Tableau blogs and forums. That's because whenever we first approach Tableau support, we anyways get pointed to those blogs. So, we first try to search on our own. If we are still struggling, only then we approach Tableau support.
How was the initial setup?
The initial setup of Tableau was not much complicated for us because we had hired a partner for its implementation.
It was implemented almost four years ago, and it probably took 10 to 15 days because it also involved getting the server, configuring it, and then doing proper configuration of the rights, etc. It is on-prem. So, we had to take care of a lot of security factors, such as opening the ports, etc. The vendor had to develop and establish proper architecture based on our security policies, and that is the reason it took more time. For simpler infrastructure, the deployment would take less time.
In terms of maintenance, Tableau does require maintenance from our end. We need to ensure that the servers are basically up and running. Sometimes, the upgrades come, and they have to be done properly. We had one instance where we rebooted without stopping the Tableau services and our entire server got corrupted. Luckily, with support from Tableau, we found the solution, and we could get back our project. So, it requires monitoring and server administration in terms of closing the services, giving the user access, and ensuring that the database size is proper because they use PostgreSQL as a backend for Tableau. These things are required to be maintained and overseen, and the licenses also need to be monitored. So, it requires a lot of administration activity and server maintenance because we are not on the cloud.
What about the implementation team?
A partner had implemented it for us. We didn't find its installation much of a hassle.
What's my experience with pricing, setup cost, and licensing?
Tableau is a little cheaper as compared to Power BI and other technologies that we have used in the past. However, if the business users in our organization want to make presentations, Tableau has been asking us to purchase a Tableau Creator license, and $35 per month is expensive for business users.
Power BI is giving a free desktop version for business users to connect to any data source and build their own dashboards. That's why we have proposed to use Power BI for most of the business users in our organization. We wanted them to be able to create charts and presentations for the management, and we didn't want to spend $35 per month on a Tableau Creator license. We tried to give them a Tableau Explorer license. We had purchased 15 licenses, but the Tableau Explorer license had a limitation where it did not allow business users to connect to their own data source. So, they had to come to IT to connect to the data source, which didn't work well for us. That's why we told them to use Power BI Desktop. This is where Power BI wins over Tableau.
In 2011, we had purchased perpetual licenses for Tableau, and at that time, they were selling its perpetual license at $1,500. We had purchased these licenses for our business users, and they were making a lot of visualizations for presentations, but about a year or two ago, Tableau stopped issuing that license. Tableau can come up with a basic license for connecting with Excel because most of the business users only connect to Excel for preparing their visualization. They can provide one or two connectors at a cheaper rate than the Creator license. It would help them in promoting it to more business users and increase users across the organization.
Most of our users are on the field, and because we are not on the cloud, one of the challenges with Tableau was about giving access to them. They didn't want to log in to Tableau Mobile, and they wanted the data in email. We have around 400 territory sales managers who want to see the data in email. When we have to publish that, Tableau charges us per user, even if they're not going to the dashboard to view it. For sending an email, they charge you per user. So, if I have to send an email to 400 users who don't even use dashboards, it becomes expensive. That's why we moved to MSBI, which is free for us. We have bought a perpetual core-based license, and we can send an email to unlimited users. We no longer use Tableau in this use case.
Which other solutions did I evaluate?
We are currently using two tools, and we are considering moving to Power BI. We are evaluating whether we want to move to the cloud or not. If we are moving to the cloud, we may completely move to Power BI in the future. If we move to the cloud, we mostly would move to Azure, and integrating Azure and Tableau doesn't make sense. We are also considering the pricing point.
What other advice do I have?
First, I would advise evaluating the data stack or database that you have, and based on that, you should make the decision of going ahead with Tableau or another technology. If you are on Azure, going for Tableau doesn't make sense. If you are on AWS or Google cloud, they have their own visualizations. So, the integration becomes a little challenging. You also need to see whether connectors are readily available for your database stack. For example, Tableau has a connector for SAP HANA, but it doesn't have a connector for SAP ECC, which is the older version. My organization is still on ECC, so we had to buy another connector to pull the data into our SQL, which increased the total cost of ownership for the company. Therefore, you need to first understand your database architecture and the kind of data you have been using and then move to visualizations.
Second, while implementing Tableau, you should not keep developing dashboards after dashboards. That's because they would require maintenance as you grow. The maintenance cost increases as you grow. Therefore, you should first evaluate the scope and then go ahead and build dashboards.
Third, there are many selling partners, and they loop you in with a minimum purchase, such as 5 Creator licenses, 15 Explorer licenses, and 100 user licenses for two to three years, which is binding. You should evaluate licensing options properly because when you are starting a journey, you don't want to spend so much at an initial stage. One Creator license, one or two Explorer licenses, and 20 to 30 user licenses are generally fine. You should negotiate on the number. Otherwise, your licenses are underutilized. This is where we could have saved money while purchasing Tableau. We are now juggling between two technologies for drill-down reports.
Fourth, if you want to send an email to your team on the field, you need to understand and ask about how many users would subscribe to emails. If most of the users are going to just subscribe to emails and not use Tableau as a dashboarding tool, it is a waste of money. In most organizations, senior management doesn't like to go to the mobile app. They want to get information through email, and if you are buying a license just to send an email, it is a waste of money for the organization.
I would rate it an eight out of ten. Two-point reduction is mainly because of the support quality, visualization limit, and lack of documentation capability.
Which deployment model are you using for this solution?
On-premises
Disclosure: I am a real user, and this review is based on my own experience and opinions.