• Mark Monfort

Analytics Evangelism - Week 2


It’s nearly the end of week 2 of the my new/old data analytics role and I’ve enjoyed seeing old colleagues and discussing analytics with various organisations not just in the heart of the city but also in other areas around Sydney. I call it new/old because I've gone back to the company where I started on my data analytics path and in the city where I grew up. Week 1 was all about getting to meet the new team and see what's been going on in Sydney (basically work acclimatising). Additionally, it’s been just as good to see the latest developments in the space of business intelligence products.

In terms of what’s new this week I got to attend a Power BI update thanks to the great community focused site Meetup and the Sydney Business Intelligence (BI) User Group - see the link to the group here: https://www.meetup.com/Sydney-Business-Intelligence-User-Group/. Big thanks to Victor Isakov who hosted and despite us being cut-short of our original 1.5 our session we were treated to some very interesting content on the topic of ‘[BI] Power BI Performance Tips and Techniques’. Whilst this was mainly a Power BI group and focused on that tool, the lessons are agnostic. The key takeaway was that we (all of us) tend to focus a lot on the sexier side of building dashboards and analytics tools (aka the front-end) and not much attention is paid to how things are built. I’ve seen this first hand in apps I’ve built and also at various clients I’ve had across the years (regardless of whether the apps were built or inherited by me). It’s an all too familiar scenario however, as the business/end users need a solution so it’s easier to just slap some script and tables together and then come up with the answer in a nice looking chart. The problem with this is that we are building in bad habits into the foundations of our apps and with bad habits comes the strong potential for turmoil later on. This may happen when we have a large number of apps reloading off of an inefficient data model that uses up more memory than it should and ends up costing the business money and time in terms of wastage. Depending on the tools you use there are a number of techniques for building more efficiency and better structures into your model and as developers we need to focus on these and fight the urge to just build shiny cool things. There’s more to BI than meets the eye!

Yesterday saw the release of the latest Qlik software community update video on YouTube. Mike Tarallo of Qlik has a lot of these ‘show and tell’ and even ‘show you how to do it’ videos so there is a lot of great content for advanced and non-advanced Qlik users. Yesterday’s update showcased a new feature that may get glossed over but is of utmost importance in the Qlik Community. Let me take a step back for a bit to describe the world of a Qlik user before their latest tools came out. In the beginning we had Qlikview. This was a great tool and, in many ways, it still is. Like many other software companies, they adhered to update schedules and would have multiple releases a year (typically quarterly for major/minor releases as well as service packs for fixing bugs). Many others follow a path like this as well whether its Tableau or Microstrategy or others. What this can mean is that if you’re looking for a specific feature that’s not out yet you end up waiting on a release date for that feature to come in. The power was out of your hands. Behold, the new king of the deck with Qlik Sense being released a few years ago. It was now a tool in which Qlik focused on their powerful Associative Engine and allowed developers to build extra charts and objects into the tool. They call these ‘Extensions’ and whilst available in the older Qlikview, was something made much more streamlined in Qlik Sense. These were made available in the Community pages and a dedicated Extensions page called Qlik Branch. In the early days of Branch we saw new charts being added all the time either by companies or individuals. These were shared and downloaded and much like GitHub, people could share the ability to develop these ‘Extensions’ to the community. The problem here was that, despite the vast array of choices, not many were well supported. If you had an extension working in an early version of Qlik Sense, it might not work in a new release and this could mean your fancy looking report wouldn’t work anymore. There was no ‘official’ support. Well, that’s changed with the implementation of a Qlik Trusted Extension Developer Offering. This is a certification program where app developers who focus on this space can become accredited and users can know that by using these extensions they are not only trusting the developer, they are also being supported by Qlik as well. One of the 2 highlighted companies is one I know well, Vizlib. The guys are located mainly in London and I know them after meeting their founder Martin Mahler at a London Qlik User Group. Great to see them continue to grow and the video attached here gives them more recognition that they definitely deserve.


Well, that’s a wrap on week 2. Let’s see what develops next week in the wonderful world of analytics. If you have anything you’d like me to talk about or share feel free to drop me a line here or on my personal email on the front page and here (mmonfort2908@gmail.com).

Over and out.

#data #MicrosoftPowerBI #Qlikview #QlikSense #dataliteracy #datamanipulation #datamining #dataanalysis #dataanalytics

11 views0 comments

Recent Posts

See All

Break time

Been a few months since publishing but I've not stopped publishing. My latest updates have actually all been up on LinkedIn articles where I've found its been much better to engage with people who lov

Follow

  • Facebook
  • Twitter
  • LinkedIn

©2018 by Mark Monfort. Proudly created with Wix.com