Implementing Self Service with Tableau pt I

Web Editing in Tableau is like Angry Birds, but a version where instead of hitting a brick wall with their analysis, the bird opens its wings and flies away freely. However not everyone knows it exists and often if they do server admins will lock it down, due to concerns over the impact on the servers. Preventing their analysts from flying through the data to find the insight they are looking for.


This was the theme of my talk at the last Tableau London User Group. “How can we enable self service at enterprise level.” 

Over the next few days I’ll be sharing with you how we implemented Self Service using Tableau.

Garter definition of self service:

Self-Service Analytics is a form of business intelligence (BI) in which professionals are enabled and encouraged to perform queries and generate reports on their own, with nominal IT support.

Self-Service Analytics is often characterised by simple-to-use BI tools with basic analytic capabilities and an underlying data model that has been simplified or scaled down for ease of understanding and straightforward data access.”

It is important to distinguish the above from regular reporting as we are often accustomed to see across organisations. Modern BI tools allow us to filter, drill-down and interact within the scope of the allowable dimensions or measures. But the interaction pool is very shallow and analysts will quickly find themselves armstrung by the limitations of the report or dashboard.

There is also another factor to consider when looking at implementing self service. The myths surrounding it. Unfortunately self service still suffers from two very distinct myths.

  1. Chaos – due to organisations structure there’s often a divide between the analytics teams and the teams managing databases and IT systems. Often Database administrators will fear that security will be compromised as analysts don’t have the knowledge necessary when it comes to governance or their fear the performance of the databases will suffer as analysts push large queries across. These are not unfounded fears and should be heeded when looking at the implementation of a self service analytics culture.

  2. Panacea – This is the analytics version of rose tinted glasses, where organisations get sold the idea that everyone in their organisation will from day 1 adopt the self service culture, and all previously created reports can be binned and never looked at ever again.

Neither of those above are true, and with care and planning there’s no reason why you cannot lead the change in self service in your organisation.

Here are some of the drawbacks I have found with organisations not implementing a self service approach:

  1. Stretched IT teams – struggling to quickly respond to requests by the analysts to modify or add in features to the DB

  2. Analysts don’t see the whole picture – potentially overlooking crucial trends in metrics not reported upon

  3. Frustrated analysts – left feeling they can’t be as productive and valuable to the company as they could be

  4. Reactive organisations – if you can’t see the whole picture how can you take timely decisions?

On the other hand company that embrace self service analytics are seeing quite a few benefits

  1. Analysts can proactively find out nuggets of information on data that was not previously available to them.

  2. Free up your IT team – IT teams can concentrate their efforts elsewhere rather than responding to intermittent request from analysts

  3. Proactive – both of the points above lead to proactive organisations

  4. Saving time and money as result

The above should give you an understanding of self service analytics and in the next “episode” I’ll go into detail about the challenges faced by our team at work and how we addressed those.

Thank you for reading, until next time.