GoldenSource Blog

It’s time to talk about data onboarding

on December 2, 2024

Over the coming weeks, I want to explore the often-overlooked subject of data onboarding.

In my experience, there are a number of common challenges and roadblocks that come up frequently during a typical data onboarding process. Dealing with them early on is a good way to achieve a more successful implementation in the short-term (and a more successful initiative in the long-term).

For the purposes of this discussion, I want to start from the assumption that the funding and budget have been approved, the staffing resources have been established, and the data platform has already been chosen.

After all that, you have to bring in the data. So, what does that mean? Here’s a list of the questions you’ll want to ask:

1) What data is it? Determine what the data should comprise specifically, how it interrelates with other data, what takes priority – and what isn’t a priority.
2) Which datasets are foundational? which need to be onboarded on top to service the foundation? This dat-centric consideration may differ from the data that the business needs first.
3) What is driving the initiative? Is it meeting an immediately tactical need, or is there a long-term strategic objective?
4) Are my data sources already fully confirmed? If not, who’s making the call and by when? More often than not, decisions are revisited when the data sources are not decisively determined during the implementation.
5) Which data classification am I working with? Is it current data, which is real-time or near-real-time information that supports the operational aspects of the business? Or is it historical data, which supports long-term analysis and reporting?
6) Which applications are needed to meet the goals? Different applications handle different data domains, whether it’s instrument data, counterparty data or market data.

If you can answer these questions definitively, you’ll be off to a great head start in your data onboarding process. But that’s not to say that it’s going to be easy. For me, it’s always been preferable to do the hard work upfront rather than kick the can down the road and hope for the best.

Next week, I’ll dive a more deeply into the first question, ‘What data is it?’ A simple enough question to ask, but depending on how you look at it, the answer can be surprisingly complex.

All Posts