TC18 - Advanced Mark Types

Going beyond Bars and Lines

How many chart types are there in Tableau?

There is no answer as really it’s unlimited depending on the techniques you use

1967 - Jacques Bertans wrote the ‘Semiology of Graphics’

Jock McKinley’s research

2003 - Chris Stolte creates VizQL

Tableau writes all of these elements together into one product

Heatmap Calendar

This isn’t a chart type - it’s taking date parts and placing on to shelves

Weekday on Columns, Week on Rows etc

Pareto chart

Is a line chart and a bar chart in combination

Donut Pie Chart

A dual axis pie chart with the pie in front, whited out. Mrunal used min(1) where as I use avg([Number of Records])

Hex tile map

Using a scaffold of the states and their X Y co-ordinates

Does this mean I should be using all different chart types?

Stop and think - challenge stakeholder requirements - ask why? Need vs Want

Discuss the analytical value

Build vs Analyse mentality - rather than just build the report, if you are asking questions of the data then you will create a better product

If you want the challenge to test your skills then there are lots of blogposts to help you work through the charts

Best way to learn is deconstruct and recreating it

This is what Tableau Public hugely helps with

TC18 - Questioning your marks

Neil Richards - HESA

Zen doesn’t mean technical impossibility

What constitutes a Viz?

Tableau means a picture of a scene - a nice way to look at the overview of a Viz

A framework, one or more visual encodings, annotations

What are the Marks?

The actual representations of your data - ie the Visual Vocabulary from the Financial Times

Removing annotations and axis titles makes most data visualisation useless

What is data Viz consumption?

Does the Viz need to be consumed in 2 seconds?

Really, the fast element is important but don’t ignore the ‘longer term learnings’ that your consumer should be able to be deduced from your work

How important are titles?

Remove them and find out!

Be careful with screenshots as people will often chop titles

Not just titles but also add short descriptions to add more depth or detail for new consumers

What is White Space?

Defined as the bit that isn’t showing data (it doesn’t have to be white)

Data to Ink ratio - what you are using to show the information vs all ink showed on the page. Ie remove any unnecessary elements

Is white space always your friend?

Remove obvious axis titles (years) and just leave the scale

Use colour in your title to remove colour legends

Reduce the darkness of non-data elements to push the detail in to the background

What makes a visualisation memorable?

This is changing over time - style and tools change over time

Intrigue leads to insight (Beaumont) - make people curious

Chart junk can add to memory as it can create a more unique style

How can you make best use of annotations and labels?

Annotations can add to pinpointing insight but also adding more clarity

Empty space can make strong, clear messages with the absence of data

How important is self review and feedback?

Very - especially when you are challenging the normal conventions

TC18 London - Getting out of your Dataviz Comfort Zome

Dataviz Comfort Zone (DCZ) - being ok with what you produce but ultimately something that is containing information and not insight

Strava is used a lot by Eva but their data Viz doesn’t allow Eva to ask the questions she wants answers to. Is it distance per week or distance per day?

Garmin doesn’t do much better- their dashboard doesn’t lead to intrigue, it leads to ‘ok’ but no further chances to ask more questions. Also shows lots of info that is not useful for the specific questions Eva wants to ask.

‘Iterate or Die’ - culture change happens continually so do your dashboards

Analysts

- Be Brave as you are the author and you are driving the improvements. Challenge the status quo, you’re the expert so don’t let those asking the questions tell you how to do it.

- Keep working on improving your skills; speak up - communicate confidently; keep an open mind; take feedback; share your knowledge and skills (teach and support others)

- Quick wins - land some benefits ASAP and keep iterating as you won’t land the perfect dashboard but sharing things early will mean better questions will be asked and you will be challenged to do more (and learn new techniques)

- Take inspiration from other people

- Research a new subject to allow you to explore the subject

Team Leaders

• Let go! Let your analysts do all of the above for themselves. Your job is now to remove blockers to your team being able to do the above

• Build a feedback culture - fostering training and create exciting roles
• Attract great talent...then support their development through training and conferences. Make it part of their reward structure. Mentoring - pay it forward by getting the mentee to double the time they receive and give it to others
• Inspire - share ideas and use related worlds to find differing approaches

Executives and Decision Makers

• Be open to new ideas
• Encourage Growth
• Create a sharing Tableau Culture - The Tableau Community is open to sharing knowledge and learnings so invite them in.
• Encourage people to play in their own environment (Tableau Public) so they can then bring those skills in to the workplace
• Create live, in person events to get your analytical individuals together. A place to play will become a great place to learn. A great way to show off how your organisation and how you work with data

What happens after that?

1. Connect, learn and have fun - share ideas with others
2. Engage, challenge and enrich
3. Empower your people

TC18 Opening Keynote

New Tableau functionality

For the Creators:

Connecting to Data in the browser through Server

Joins and connections in the browser so can form your own data set

Workbook formatting on the web

Quick number formatting to make the number formatting process less hidden

• Able to see instant updates

Dashboard starters for Tableau Online (Salesforce, Marketo etc)

Viewers:

Able to comment on images and snapshots taken from dashboards. Can lasso the elements of interest and write custom comments over the top of free form lasso

Vegas Data17 - Opening Keynote Brief Notes

Adam Selipsky

61,000 customers in 100+ countries

Data Myths - created to replace the unknown and create reasons for what happens

#1 AI will replace the analyst. Actually AI is likely to assist the analyst. Tableau is smart software. Drag and drop of Clusters and Trend. Natural Language Processing will take out some of the technical barriers.

#2 Data is only for the analysts. More data programs in Higher Education. 800 million knowledge workers. Excel used to be taught in university, now school children use it. Tableau aiming for the same.

#3 Data governance means no. Data is valuable so needs to be protected. But that’s the old model - the bottleneck has been removed. Governance should means secure enablement.

Honeywell deals in human safety. It needs to have appropriate data ready at appropriate times. Giving multiple environments with clear transitions between them helps to manage creation at all levels. 20,000+ users within two years as the right mix of options gives governance and flexibility.

#4 There can be one, perfect source of truth. Innovation is so rapid that you can predict the different sources and combinations of these. We live in a world of many sources of truth. We have to embrace that. Tableau invests in all of the flexibility that you need. Go past the hype and try the tools. 

Francois Ajenstat

Myth - BI platforms take power away from the people. Often designed for specialist. Tableau focuses on people. 

100+ features added in Tableau over the last year. 50+ came from the community.

10.5 in beta today

Hyper

Linux

Viz in Tooltips 

Data Engine

Hyper is the new Data Engine. Tableau Data Engine was great. But time to scale up. Hyper instant compared to 25 second load where heavy calculations are used. Extract creation will also be massively improved (3 million rows live SQL) double the speed in Hyper. Hyper doesn’t sort the data like a TDE. 500 million rows isn’t possible with TDE, Hyper is possible. No migration necessary - just use 10.5 and it’s there.

Data prep

Still working with existing data prep partners but not everyone has this - introducing Maestro. Data profiles gives you a sense of your data. Filter outliers with normal exclude functionality that you find in Tableau. Grouping through Fuzzy Clustering to sort poorly entered data. Clear view of the changes made through the transformation. Drag and drop joins and unions. Very simple join and cleaning of joins. Maestro in beta this quarter

Extensions API

Makes additions in to Tableau rather than Tableau in to other applications. Dashboards becoming their own applications. Dashboard Extensions create two-way communication with the data source.

What’s Next for Tableau - Andrew Beers

“This next hour isn’t about me, it’s all about you”

Three principles and how Tableau is answering them:

1. People who know the data should ask the questions

Profile: bring domain knowledge rather than the SQL-coding skills to get to the data to answer the questions

2. Software should be designed for deeper thinking

Profile: Not simplistic products, simple ways to do complex things

After the first five minutes of simplicity there has to be more [complexity]...to let you have the impact you want to have

3. Analytics at Scale can drive change

Profile: Analytics at scale can solve the big problems in the world

Last 12 months in the Tableau world has gone from 10.0 to 10.3

Half of the 130 improvements came from ideas on the Ideas forum

Loving that Tableau is on a ‘quarterly’ cadence

So how does Tableau help us achieve this?

Connect

Data Prep: “If you can’t connect to the data then you can’t analyse it”

So need to connect to all data (pdf, cloud files, Eloqua, Service Now etc)

Hybrid engine: Fast Live & In-memory, Cross DB Federation

Data Prep: Self Service, export to csv

PDF - shows all the tables in the pdf when loaded into Tableau. Can pivot

JSON - Tableau detects the structure. When connection to JSON, if you have multiple rows then Tableau will create the LoD measures removing the duplication for you

Answer More Questions

More ways to change lines (regular, stepped or jump lines)

Tooltip selections: Using dimensions in the tooltip to enable highlighting other other marks that allow the common elements to be highlighted.

Dashboard element sizing - ‘Distribute Evenly’

Python / R integration - use SCRIPT_REAL function

Collaboration at Scale

Sharing / Governance

Discussion

Altering

Anywhere Web & Mobile

Tableau is now watching how the whole community is using the common joins done to ‘recommend’ what tables to join on

Alerting - can schedule regularity of alerts

Big changes coming soon:

Linux

Hyper

Maestro

Linux: cost saving / ease of management / technical skills of IT / more security control

Hyper: support for very large data sizes (billions of rows), faster queries, much faster extract creation, scalable (mid summer public beta)

Maestro: Complete visibility for immediate understanding (coordinated views aid understanding)

Iron Viz - We can't breathe

It’s that time again with IronViz feeder competitions starting up again. Every year, I take this opportunity to explore a technique I haven’t used before, or alternatively, a subject that I care about.

As a keen cyclist in London, Air Quality is a really important subject to me. With increased cycling infrastructure, the move to electronic cars and banning older vehicles from the city centre, I was expecting to download a set of data that showed dramatic falls in the levels of pollutants that are all around us Londoners. Nothing could be further from the truth.

Using data from londonair.org.uk, I looked at the Borough level of detail as there are only three or four monitoring points per Borough. The data proved a challenge with sporadic measurement across an inconstant set of pollutants. This meant that finding trends and understanding geographical patterns using the new Spatial file connector in Tableau was a good challenge.

The biggest break-through I had was whilst researching the subject, was coming across the European legal limits that have come in to force across different pollutants, at various points since 2005. This is shown as the reference line on the trend data to show how often the limits are broken across London.

The answer in many cases is the daily average often doesn’t fall below some of the limits. Worrying stuff!

The spatial mapping allows the reader to see the impact on the city centre. Although Lambeth, outside of the true centre of the City, seems to be hit particularly hard by pollution levels.

As with any IronViz entry, it’s a chance to use a little imagination and have some fun with charting. When I played with some charts, it looked like exhaust fumes so, I kept it and features it at the top of my visualisation.

Overall, the visualisation has led me to want to dig in further in to this subject and what I can do to start making a difference as currently there isn’t any improvement of any note.