I am writing my blog post this week from San Jose, California USA live from the 2016 PASS Business Analytics Conference. For the benefit of those of you that are not familiar, PASS is the Professional Association for SQL Server. Each year PASS holds a Business Analytics Conference – it is the premier conference for Data Analyst Professionals (as opposed to IT Professionals who are catered for at other events). So if you are an Excel or Power BI professional, then this is the PASS conference that was created just for you.
Jer Thorp Keynote Speech
I really enjoyed the Keynote from Jer Thorp today. Jer is a data artist – he specialises in turning boring data into meaningful visualisations.
There were a lot of gold nuggets that I took away from this session that really resonated with me and that I want to share. Here is a list from my notes with some comments from me:
- A great definition of Data is the “measurement of something“.
I love the idea of “Measurement of Something” as a way to describe data. I get asked all the time what I do, and this description of data and analytics is really useful to simplify a complex topic.
- Don’t think about the printed page when thinking of visualisations.
This is a lot harder to execute than it is to say. Jer showed some fabulous examples of engaging graphics using custom developed visual tools. For most of us we are practically tied to generally available tools like Power BI and Excel to create our visuals. Thankfully these tools are developing lots of new visualisations (particularly Power BI) which opens up new opportunities for communicating. But our job as Analysts is to seek out these tools and use them.
- Visualisations should try to create an Ooh/Aah response
- Catch attention => Ooh
- Tell them something they didn’t know => Aah
This is really helpful to me as I sometimes wonder about the merits of “flashy” demos. With the lens of “trying to get attention and then tell them something they didn’t know”, it makes the purpose and value of the whole exercise a lot clearer to me.
- Data Visualisation should strive for Reduction or Revelation.
- Reduction example: Blood pressure is a reduction of a person’s health into a simple 2 number measurement.
- Revelation example: X-rays expand the available data to provide new insights that are not otherwise visible.
Reduction or Revelation is a simple concept that is easy to remember and will help any analyst focus on the task at hand.
- Data analysis can be Confirmatory or Exploratory
- Confirmatory maps to Satisfaction.
- Exploratory maps to Joy.
Another simple concept to help focus you on the task at hand.
- Good visualisations help you to ask new questions – i.e. it is question farming. You create new questions you didn’t think to ask before.
Like I said above, I have often questioned the value of “flashy” visualisations with moving bubbles etc. But if such visualisations help people ask new questions, then they have achieved something great – I never thought about it that way before.
- Upstream data. The closer you get to the source of your data, the better you will understand it.
I think I have always intuitively known this one. I spent my career working in the retail industry and everyone there knows that the answers can’t be found in the office, but in “the trade” (in the retail outlets). You need to get out and look if you want to understand your data.
Why not start planning for next year’s PASS BACON now.
For some reason business analysts don’t seem to hold attending conferences at the same level of importance as IT professionals. I think this is a shame as there is a lot to learn and a lot of networking opportunities to be had. If you are a data analytics professional, I hope you will consider attending a professional development conference at some time in the future. There is no time like now to start planning for this – make sure you get some money put aside in next year’s budget.
May 04, 2016 7:28 am