Don’t throw good money after bad

Decision-makers, business analysts and architects must help transform the data culture and thinking within organisations.
Julian Thomas
By Julian Thomas, Principal consultant at PBT Group
Johannesburg, 13 Dec 2023
Julian Thomas, principal consultant at PBT Group.
Julian Thomas, principal consultant at PBT Group.

In my second article of this series focused on making the data industry great, I stressed the importance of having a professional body and framework that has a strong focus on the business component.

In the article, I further stressed the importance for business stakeholders to be heavily involved in committing to and driving this framework in their organisation. I explained how important it was for business to change and adapt to enable the transformation to a truly data-driven organisation.

I believe this makes a great deal of logical sense, and certainly everybody that I speak to agrees with these statements. But my experience is that everybody leaves the conversation, goes back to the office, and proceeds to continue as they always have done before − building the data platform that they believe so strongly in, yet without the critical business involvement that we all believe is so necessary.

It is at this point that I believe we need to focus ourselves. We, as data professionals, believe so strongly in what we are doing, that we push forward, trying to build these platforms without the very first, most critical success factor in place.

Are we really helping by pushing for these initiatives before the business is ready?

Maybe I have been in the industry too long; however, I cannot help but sometimes think: “What’s the point of throwing good money after bad?”

Are we really helping by pushing for these initiatives before the business is ready? Are we really helping our employer by convincing them to invest in a new platform, when we know they are not going to make the business commitments that are required, to make the data platform truly successful?

Are we really helping by implementing data governance initiatives in the data warehouse platforms, when we know that what the business really needs, is data governance that is implemented in the operational systems and processes of the organisation instead?

I have always thought the professionals in our industry were brave and showed high commitment in continuing to implement the strategy, even without true business support. But maybe, a different type of bravery is called for here?

Maybe the type of bravery and commitment that is required here is the kind that does not allow us to embark on these journeys until the organisation is truly ready for it? Sometimes, it is better to do less.

But what does that mean for the technical team? Do we pack up and go home? Change careers? Find alternative employment?

Absolutely not. The decision-makers, business analysts and architects need to focus on the business transformation side and help transform the culture and thinking within the organisation. This will take time. While that is in progress, the technical team should use this time to “clean house”.

It's like the analogy that I always use for my clients when planning for a migration. I tell them, imagine you are moving house. Do you pay the moving company your hard-earned cash to move things that you don’t want into your nice, new house? Or do you first re-organise the contents of your original house? Declutter, throw away all the junk in your garage you never ended up using, and ensure you only take the things you really want to keep to your new house, and potentially save money in the process, since you will be moving fewer items?

This is what we must do as data professionals. Instead of trying to implement a strategy that we know our organisation is not ready for yet, we should focus on optimising what we currently have.

There are generally a variety of legacy challenges in any data platform, that never get resolved, and that will continue to hinder the future strategy implementation if not dealt with.

What are some of these items that we should be focusing on? They can include a variety of aspects, some specific to industries and individual organisation’s needs and challenges. In general, however, I have found there are some improvements required that are largely universal. These include:

  • Update and review the contracts and service level agreements for the provisioning of data to the data warehouse, both internally and externally. These need to be standardised across all parties, to help improve consistency and achieve economies of scale.
  • Update and review data governance principles within the organisation, in particular those that limit the organisation’s ability to deliver on the principles of data democratisation, such as security policies, enabling external access to data (ie, APIs, web services, etc), access open-source technologies, cloud enablement, etc.
  • Review and improve the technical mechanisms for integrating data. It is bewildering how many organisations still rely on file exchanges, when alternatives such as database replication, message queuing, etc, are readily available. Even more bewildering, that we accept costly daily exchanges of data when we know our business needs access to data as quickly as possible. Focus on pushing for as close to real-time integration of data wherever possible.
  • Improve the existing data warehouse solution. You will be surprised how much opportunity there is for even simple improvements to the legacy data model design and associated ETL to yield significant benefit to the platform.

These are just some examples that I regularly encounter. Until business is ready and mature enough to fully commit to becoming a data-driven organisation, we as data professionals must commit to gearing up.

We must focus our energy on preparing the organisation for the implementation and focus on the things that we can do that will have real, technical benefit and impact on the business, both in its current state, and in its future strategic implementation.