It’s good for quick visualization and being able to quickly consume unstructured data to play around with. This is good way to show a demonstration/prototypes of dashboards and scenarios for design discussions on reporting requirements or to show what the data is telling us when it comes to features of data integration, OLAP services, data mining and extract, transform, and load (ETL) capabilities.
Improvements to My Organization
Good for adhoc visualization of an unstructured dataset which comes from other sources outside of source systems that you can overlay on top of the structured data and you have to get a quick visualization dashboard prototype going. It helps with the Agile design build and can be used in our current operations analytics work to overlay multitude of data sources that we know of. Can always work offline, which is nice. It’s good for organizations with very limited staff to do quick report builds and dashboards that can be put on our SharePoint site for sharing or on reports when responding to data caps. I use it a lot for design discussions so I can communicate the gaps in data sources for data exchanges or to generate a storyboard prototype of how the data is to be used in visualization but where we need to have data exchange/ETLs on.
Room for Improvement
It is good for its use if ad hoc, offline, or needed for quick turnaround on reports/dashboards. It is not so great when it comes to data exchange/integration, data mining, etc. I rely on what’s available in current versions to see what APIs and plugins that I can use and they have Open Source on GitHub is a plus to share things to re-use.
Room for improvement is more on data integration features that are agnostic to any solution platform but can be plug and play to be able to reuse what was built out Tableau in any other platform of work.
Use of Solution
Over five years, and the past three more for integrating the use of a similar family of tools where Tableau is one of a few options in our environment, where these services are considered for quick-hit items as needed, given time, dollars, and analyst skills.
Yes, see other answers. Scalability per user defined elements are okay but not so much for enterprise wide reuse. Per license cost can have some work done to it to make it more affordable on the recurring maintenance end of things. I would like to see more subscription based models.
Customer Service and Technical Support
I don’t have to use it much since I can get much of this through current site materials and social media blogs/videos.
I didn’t switch. It was just a matter of seeing where Tableau makes sense as a service to use in our environment, which is for the simpler, not so complex, but quick turnaround. Worked with other technology stacks that are similar, like Information Builders, SAP, Microsoft, Oracle, SAS, MicroStrategy, IBM, Salesforce, Qlik, etc. I find Tableau the easiest on visualization and its license model straightforward. But when it comes to scaling to other interoperation work, not so great on the wizard template, to do data mining/exchange. It doesn’t have that robust analytics and intelligence self-learning feature that comes with other tools.
It’s straightforward, like any typical software. You have just got to understand what the various versions of API and plugins and what they can do. Though I have noticed that there were situations where it was said it was able to do things, but not until a later version. It needs better communication on that front.
Pricing, Setup Cost and Licensing
License small scale and run with it to get a business case going on its use. Give the licenses only to those analysts you want to do quick turnaround visualizations and those that know the data sources/data (those that don’t will just have access to tool and that compounds a problem with giving you something nice to look at but no meaning behind it, which I wouldn’t recommend). Look to existing platforms of one’s current BI environment and see if you can have a server license which can reduce the per user licenses.
I wish there was more of a subscription model with the pricing when it comes to Tableau, so you can get all the latest version upgrades/features if you pay monthly/annually, rather than buy straight up licenses that you lock to a baseline version and have to pay for upgrades later on. It limits how many users you can get on the thing, and it's not like you will use it all the time.
Other Solutions Considered
Yes, we did an alternatives analysis of all the product line options against our criteria of need in our environment, where recurring cost, time to implement, and other interoperation, security, platform scalability, architecture, etc. factors play a role. The majority were mentioned above.
We always have the latest versions of Tableau (part of the package deal), so we can have the latest in APIs and integration hooks and plugins needed across our platforms of OBIEE, SQL, etc.
Tableau is good for quick visualization once you have the data, but not such a great interoperable tool or getting to multiple sources without a lot of work and know-how. Good for pulling in unstructured data and doing quick reports/prototypes. Does require some stronger business analytical skills rather than your novice user (and technical with regards to use of API and plugins).
If new to the analytics/BI market, use it, as it's good for getting you jumpstarted to understanding your data/data sources and to envision what you can use the data for. It's a good starting tool for that. If more advanced or need it for interoperation, I suggest looking to see how it fits with your current environment and determine where best to use it as it shouldn’t be your only option as the features are not robust enough to scale for everything.
Disclosure: I am a real user, and this review is based on my own experience and opinions.