Data Modelling and Reporting Team Lead, Ilza Lindeque, shares how her chance encounter with technology led her from finance to fintech, big data and the leading edge in data modelling.
“My career in data started by chance. After varsity, I ended up in a graduate program with a tech company, where I trained to become a consultant for financial performance software. Having studied financial management, the theory and principles of it all made sense, but I had never even heard of a database or ‘extract, transform and load processing’ (ETL as it is commonly known). A cube was to me still a shape and not a multi-dimensional array of data used for analysis!
It was, however, a dream job — every project was something new: new industry, new client, new challenges, but still with the financial flair that I love.
For a brief period, I wanted to return to my roots and try my hand at a proper finance job. I barely lasted a year though. It was not because of the job or the people — they were all amazing. I had just become addicted to the pace and fast-changing world of working in a tech environment. I wanted that constant change.
By far the most exciting change of my career is the journey we are currently undertaking at JUMO.
We are moving from a very traditional data warehouse solution into the world of big data, data lakes and new and creative approaches to data analytics. For most of my career, I’ve sought to understand the end-to-end workings of a structured, typical star-schema data warehouse. I’ve been lucky enough to have done it all — from the design to ETL builds, OLAP cubes, dashboards and reports. It was fairly easy to learn with nicely packaged software, enabling even someone like me who is mostly self-taught to carve out a solid career as a business intelligence specialist.
Enter JUMO — with an appetite for data, unlike anything I’ve experienced in my career. Consuming data for everything from customer behavioural risk prediction to customer value proposition analytics and performance, financial and monitoring reporting — you name it, we do it all at JUMO. Our little workhorse of a data warehouse is becoming tired. A costly and difficult-to-scale beast that requires effort to conform to the various new data sources we consume as we expand our technology platform to scale for the future.
We are moving away from the traditional conformed dimensional models into a more unstructured environment, consisting of a loosely coupled data lake, event-based data feeds and operational data stores.
This new architecture allows us to reduce the lengthy build and processing time required to conform data into the rigid data schemas designed in the warehouse. We can also explore event-driven data ingestion where there isn’t a dependency on source database tables. With the data lake, data can easily be stored in a repository in its natural or raw format as either object blobs or files.
There is also a massive cost saving in that storing big data in this way is far more cost-efficient, as these types of solutions are geared for volume.
This is all brought together in flexible reporting and analytical data layers. Trying to reduce processing as much as possible so there is little time between when the data lands to when it’s available for consumption. Ultimately this allows us to serve our customers quicker and create greater efficiencies in our ways of work.
Although our journey is still a work in progress, we have a clear vision of where we would like to end up. It has at times required an uncomfortable shift in mindset, but it’s an exciting opportunity to learn and expand my horizons. By far the greatest challenge is getting to grips with all the new technologies and the trial and error in getting something to work. Nothing beats that satisfaction when a process runs successfully though, or your dashboard renders beautifully.
It does come with several exciting challenges, but with the amazing team we have at JUMO and incredibly creative ideas combined with an amazing technology stack at our disposal, only great things will come of it.
I’ve already learned so much from all the talented people I’ve worked with in my career and I can’t wait to see what changes I’ll be lucky enough to experience and be part of in the next 10 years of my career.”
Twitter: @lady_lindeque
