Archive

Posts Tagged ‘reporting’

Making Dashboards Relevant

March 11th, 2014 Comments off

Recently I’ve seen a lot of sexy dashboards created in Tableau, doing some very cool things. It got me thinking, these look great but actually they’re useless, the report users would look at them once, think “wow”, and then never use them again as they’re not providing any useful information.

If you’re building reports remember why you’re doing it – to provide meaningful and actionable information, and supplying the information in as few clicks as possible to the user. Don’t make your reports like an Aston Martin without an engine – i.e. looks great but not useful for the main requirement it needs to meet. If you can make dashboard sexy that’s great but remember the main purpose of the report is information supply.

Tableau Dimensions as Columns in a Table

December 18th, 2012 7 comments

If you have created tables in Tableau you might have noticed once you get to a six dimension columns  in the table the left hand columns begin to merge into one column. This only happens when using dimensions as column headers, not measures as when displaying multiple measures in a table this can be achieved by putting only 1 pill, the Measure Names, on the shelf. If you want to know how to add measures read my article describing how to display multiple measures in a table.

Tableau is not really designed for showing tabular data but sometimes it’s useful to show line item data as part of a drill down in a table, or perhaps you need to create a report specifically with the purpose of exporting Tableau to Excel. Read more…

Date Hierarchies in Tableau

May 31st, 2012 10 comments

I recently had a request from someone who wanted to be able to drill into their data both from Year – Month- Day AND Year – Week – Day. The first of these is very simple as it’s more or less a Tableau default hierarchy from any date field, although the default also includes quarter (Year – Quarter – Month – Day is the default).

Adding Week into this adds a significant degree of complexity for the simple reason that weeks generally don’t span months in a neat way – i.e. it’s only on very rare occasions that the first of the month is also the first day of the week and the last day of the month is the last day of the week. Almost always the end of a month and the beginning of the next will be in the same week number – in other words some weeks will span 2 months.

The first challenge is to create the Year – Week – Day drilldown. To do this a hierarchy actually has to be manually built. In a previous post I described how to build a hierarchy. Making use of calculated fields it’s possible to build on the hierarchy concept to build a Year – Week – Day date hierarchy. Read more…

Tableau Dynamic Sorting

May 28th, 2012 19 comments

Recently I created a tabular report where it was a requirement for the user to be able to sort the list by all of the different measure columns. I generally publish my reports on to Tableau Server where sorting is easy by using the Tableau toolbar but I wanted to see how to do this without using the toolbar.

I began to think about ways to do this using parameters and calculated fields as a Dimension has the option of Sort By Field. Luckily I didn’t have to think to hard about how to do this as I stumbled across a blog post which describes how to set up dynamic sorting with Tableau. I slightly modified the more complicated of the 2 techniques described in that article, which I describe here.

For this example I’m going to use the Superstore Sales sample data Tableau provide and create a table showing the Customer State on the rows and the measure values Profit, Profit Ratio and Sales in the columns. If unsure how to do this check my other post on displaying data in tableau as a table. Read more…

Tableau Filter the data on display and not the underlying data

May 16th, 2012 No comments

I was recently set a challenge to filter a chart based on something that happened in the past, but still to show all time periods. The specifics of the task was to look at the number of employees a company had in 2010, and depending on that amount either show or hide all years for those companies that had the right amount of employees. In other words create a ‘Number of employees in 2010’ filter.

The dataset contained only 3 fields, Company, Year and Employees.

Initially it seems simple. Create a calculated field to calculate the number of employees the company had in 2010 and use that as a filter. This is what I did, the field was called [2010 Employee Count] and the formula I used was SUM(IF [Year] = 2010 THEN [Employees] END). Next I had to set up the visualisation to test what I created worked.

I dragged the Year to the columns shelf, Employees to the rows shelf and summed it, the Company to the colour shelf, giving me a line graph with annual employee counts over time split by company. This gives me the following line chart:

Next I tested the filter by dragging the calculated field [2010 Employee Count] to the filter shelf and using the At Least filter to show all companies with over X employees in 2010. This is where it became evident that this wasn’t going to work – the filter affected the entire dataset and not only the selected companies as can be seen below: Read more…

Info About Extracts on Tableau Server 6

May 3rd, 2012 No comments

My current employer is still using Tableau Server 6.1 – or 6.1.6 to be precise. I’ve been trying to work out how the extracts work when pulling from a SQL Server so the same extract can be shared by multiple dashboards. The reason for this is that some of my extracts take a long time to update which is not ideal when the same extract is used in multiple dashboards and instead of refreshing it multiple times for each dashboard it would be more efficient to update just once and all dashboards using it will be updated together.

What I was wondering was if when you connect to a data extract in Tableau is whether the published extract is updated when any dashboard based on this extract is updated. It seems this isn’t the case. Once the extract is published it remains as it was a publication time for all time I believe. It appears that when you connect to that extract and publish the workbook Tableau actually duplicates the extract and updates this duplicate whenever the dashboard is updated. In other words the extract published individually to the server doesn’t alter but the duplicate Tableau made embedded in the dashboard does update. Read more…

Build a hierarchy in Tableau

March 13th, 2012 No comments

One of the most useful things for me in Tableau is the ability to create a hierarchy in a few seconds. I have used this for building product hierarchies – i.e. product category, subcategory, group, etc – and also for reporting on regional data – i.e. Country, State, County, City, etc, which is great for using on maps (assuming you have a longitude and latitude at the bottom of the hierarchy). Read more…

Waterfall Charts in Tableau

March 8th, 2012 No comments

I’ve noticed waterfall charts are becoming quite fashionable now. If you’re not sure why this is used, it’s a great way, for example,  of showing how total revenue is compiled from it’s components. Using the Tableau Gantt Chart it’s quite simple to build. In this article I’ll give a very quick guide to building your own such as the chart below.


Read more…

Tableau Server Passing Parameters in the URL

February 6th, 2012 3 comments

If using Tableau Server it’s possible to pass parameters to filter your dashboard via the URL. It’s very straightforward and detailed in this post on the Tableau Forums. Basically just get the URL to access the dashboard on the server, add a ? at the end, which is how parameters are entered in a URL. After the ? put either the name of the parameter or the name of the field being filtered, =, and then the value.

For example if you want to filter/pass an email address parameter – field/parameter name being EmailAddress – your URL would be as follows: http://ServerName/views/WorkbookName/ViewName?EmailAddress=andrew@reports4u.co.uk

To add multiple of the same filter use a comma (,). For example to filter for 2 EmailAddress’s use the URL: http://ServerName/views/WorkbookName/ViewName?EmailAddress=andrew@reports4u.co.uk,admin@reports4u.co.uk

Note this is case sensitive so make sure this is taken into account. To add additional parameters add an ampersand (&) to the end of the URL for each parameter/filter and the same Name=Value syntax. For example to filter the view for EmailAddress and ReportYear you could try the following:
http://ServerName/views/WorkbookName/ViewName?EmailAddress=andrew@reports4u.co.uk&ReportYear=2012

Bullet Charts in Tableau

February 5th, 2012 No comments

I’ve only just discovered the bullet chart, I’ve had no need to create one until very recently. Both I and my customers find them really useful. They replace the less meaningful ‘traditional’ dashboard tools of guages and dials into something which conveys more information.

The purpose of using them is to compare a value against another value – for example sales this year vs sales last year or order count this year vs budget order count. It compares against the Total of 1 value and the percentiles of that total. In other words it can be seen at a glance if this year is ahead of last year, or to what percentile this year has reached against last year. Read more…