What’s more important: data quality or data management?

If you ask any researcher, ‘Does data quality matter?’ The answer is almost always going to be ‘yes’. No surprises so far. If you ask a researcher, ‘Does data management matter?’, you are likely to get a mix of responses, or even some quizzical looks.

Which is more important? Well, in my view, they go together, and they are equally important. So, let me explain why I believe this.

Remember the good old days when it was simpler

It wasn’t too many years ago that the process for the vast majority of quantitative research was Fieldwork > Data entry > Crosstabs > Report > PowerPoint presentation (or, possibly, a part of this path). Having an internal process to achieve this efficiently was easier, provided you had the right skills and tools in place. That’s changed, not for every project, but for enough projects.

So, what’s different?

The paths have become more diverse over recent years. What is possible has changed. And, what’s more, everyone knows that there are greater possibilities. Let’s take deliverables. It might be crosstabs, charts, PowerPoint presentations, dashboards, data integration within a customer platform, customised data delivery, API-based connections, Power BI integration, and more. And where does the data come from? Different products, different data structures, different integrations, different, different, different.

What does this mean?

The result of this growth of different data and analysis paths is the need for greater flexibility in data management – admittedly, not the most interesting phrase in any business. The old one-size-fits-all route from fieldwork to deliverables is no longer enough. That is why data management (yes, that word again) is just as important as data quality. Sure, if the quality of your data is poor, even the best data management won’t help, and if you can’t deliver data in its most useful form, the value of that data is diminished.

But I can do everything in one platform, can’t I?

I often hear the claim from many platform providers that you only need their platform to meet all your research needs. That’s absolutely true if your clients ONLY (and I don’t apologise for capitalising that word) have requirements which are achievable within the confines of that platform. And mostly, it will be the confines of the platform that dictate what you can deliver,  some things almost certainly well, other things not so well, or not at all.

How have we responded to these growing demands?

When we designed MRDCL Hub three years ago, we built it around MRDCL’s proven engine. Three years ago, it could number-crunch the most complex data, the biggest surveys, and the most awkward crosstabs. But that wasn’t enough. We needed a platform that could handle whatever data needed to be analysed and whatever deliverables were most appropriate for the project or client.

So does MRDCL Hub do everything?

If that sounds like yet another platform provider claiming that you will never need another platform to satisfy every client, you would be wrong. MRDCL Hub is designed to handle most of what you need in one environment, while giving you the flexibility to easily connect to other systems or outputs whenever a project demands it. Yes, it can do the heavy data processing. It can also provide a dashboard, but this would still miss what makes MRDCL Hub different. It provides the tools to connect MRDCL to the systems and outputs a project actually needs, allowing the workflow to be shaped around the project rather than the other way round.

What does this mean in practice?

When you subscribe to the MRDCL Hub, you are not restricted. You can deliver whatever clients want and build easy-to-manage systems that automate or execute the processes required for a specific project.

Is this just another AI platform?

AI will increasingly help simplify and automate parts of the process, but that does not remove the need for good data management. If anything, it makes flexibility more important. The value lies in being able to take the same data wherever the project needs it to go – efficiently, effectively, within budget and without forcing every project down the same route.

Data quality or data management?

Of course data quality matters. But if you cannot move, process and deliver that data in the form a project needs, some of its value is lost. That is why I think data management deserves to be taken every bit as seriously as data quality. Do you agree?