Data Platform | 5 things to consider before the data platform is ready

Many companies are currently implementing a new data or analytics platform. It takes a lot of time and attention, especially from those of you with leadership responsibilities related to data and analytics work. But as you approach the finish line, the relief can quickly be replaced by a different kind of unease. Because how do you actually make this work in practice?

When we change technology, there is a lot else we need to change as well. This should not come as a surprise, but it is nevertheless not uncommon for us to start far too late.

Here are some things you should have considered and acted on before the data platform is ready.

1. Before you begin, you obviously have a data strategy

I say obviously, but it is absolutely not a given. You are not alone if your data strategy feels a bit vague, and this does not stop companies from investing millions in a new platform. But even if the data platform is just around the corner (or is already in place), it is still not too late.

A data strategy is your guiding star for everything you want to achieve. If you think that having a direction might be a good idea, or if you are wondering whether the strategy you already have is good enough, this article on why you need a data strategy can give you an extra push.

2. You need to define new ways of working

Will everything be the same, just better? Most likely not. You need to answer questions like “How should we use our new tools?”, and my personal favourite: “How should we not use them?”.

Methodology for development and analysis, access management, quality assurance, and all the tedious things that stand in stark contrast to the “agile” company you believe you are, need to be in place. If we are to benefit from our investment going forward, it must be treated properly. After a demanding implementation project, it is tempting to go for immediate short-term gains. But when data quality fails, temporary solutions turn into capacity-draining traps, and information ends up in the wrong hands, it is you who are responsible for data who must answer for why things went wrong.

3. You need to set measurable requirements for change among your employees

Data and analytics is something entirely different today compared to 5-6 years ago. This means that you may have some employees who work with data in the “old way” and whose skills do not quite match today’s role descriptions or what a new platform demands.

It is common for us to have a chat about what is needed to fulfil the updated version of an employee’s role, but we often forget to ensure that we can measure the change we want to achieve. General descriptions of the need for change mean that some employees who either cannot or will not change continue doing much in the old way. This also means that achieving results from the change takes time, or never materialises at all.

By setting measurable requirements for change, you and the employee have something concrete to refer to when discussing how the transition is going. An example could be completing a course or a certification. If there is no progress, or the employee does not pass the certification, it is easier to have a new conversation about how to achieve change. Without measurable requirements, this can quickly become an uncomfortable situation where the employee feels they have changed, but you are still not satisfied with the results.

4. You need to know what you cannot do, and bring in someone who can in time

Even with good transition plans and skills development, there may be areas you will not be able to cover, or where you will need deeper expertise. The competency fields within data are many, and the definitions of a data analyst today stretch from one extreme to the other.

It is not enough to count how many employees you have; you need to know what they can do and what they cannot. The market for skilled data professionals is red hot. It typically takes at least 6 months to get new people in place, and even longer before they are performing well in their roles. First, it will take time for the recruitment agency to understand what you are looking for. Then it takes time for them to find the right candidates. Finally, it also takes time to get them out of their current workplace. It all adds up, and in the meantime, the organisation’s patience is running out.

To avoid a situation where the platform is ready but no one knows how to use it, you should start recruiting now.

5. You need to be personal in communicating the need for change

When we are in a change process, it affects those around us. Adapting to new technology and new working methods takes time, and we need to be granted that time by those we work with. The management team may have originally only said yes to a new data platform, and here you come with a whole lot of additional needs?

We know that in change management it is important to communicate why we need to change. Yet it sometimes happens that we stop a bit too early in the chain of consequences when conveying this message. Perhaps because we think that these are conclusions everyone can draw themselves, or because we feel it becomes too exposing of individuals if we give concrete examples of things that can go wrong. But when we keep it general, the arguments often do not fully land with those we need on board for the change journey.

If, for example, you need to communicate the need for new data quality routines, it is common for us to explain this by saying that in the worst case it could lead to errors in our analyses. We tend to think that everyone can imagine the consequences of erroneous analyses, but it is easier to spark the desire for change if each person understands how this will affect them and what they personally care about: “If we do not have good data quality routines, we risk that our analyses may contain errors; this could lead us to make wrong decisions about which initiatives to prioritise, and could cause us to lose our market position.”

If you manage to paint the full picture, you will likely gain more support for your plans. That support is worth its weight in gold on the road to the goal of a new well-functioning data platform.

Now that the data platform is almost ready, it is time to think about which other areas need to keep up on the journey, and what challenges and opportunities you have there. Talk to people about their expectations, hopes, and concerns to get an overview of where the most effort is required.

Start the work now to prepare everyone for the new everyday, and life will be a little easier later on.