Building a firm foundation for your future proof data strategy

In my first post, I talked about the hell of using Excel for all your data needs; in my second post, I covered the building blocks and initial first steps that will lead you to a future proof data strategy. Now let’s zoom in on the three domains that underpin it all.

1. Data management

It all begins with Data management – that’s what will help you consolidate and manage enterprise data from a variety of source systems, applications and technologies. Depending on your needs, tasks like cleaning, migrating, synchronising, replicating and promoting your data will be required. In the selection phase, look for options that are optimized for a variety of analytical tasks that you’ll want to perform in a later stadium.

It’s often said that 90 percent of the tasks performed when working with data are “data management.” That’s often the case, but it doesn’t have to be that way.

With the right data management tools, including data visualization, you can make huge improvements. Doing some quick validations and testing before creating an industrialized ETL job will help you switch faster. Visual data exploration of the results will help you detect overlooked issues and determine the origin of any problems.

2. Business intelligence

Next, let’s consider business intelligence – BI allows your users to create, produce and share reports and analysis. With BI, your goal is to empower different types of uses within your organization to answer business questions. 

You’ll still provide standardized reports for information consumers, but you want to provide self-service capabilities as well (for more on this, read my post on balancing self-service and governance).

The visual aspect of business intelligence and its importance is something I explained in more detail in: See what you never expected to see with data visualization.

3. Analytics

Last, but not least, the third (and perhaps most valuable) data domain: Analytics. Depending on the analytical maturity of your organization, you need basic or more advanced analytical tools to do forecasting, data mining, predictive and prescriptive modelling and text analytics.

For many people analytics sounds like something for the math-magicians and data scientists (and to some extend that’s true), but as an organization you have to build up experience in analytics. Start small to grow big. This is something I explained in a recent post about building analytical expertise. You can use the latest developments and technologies to start benefitting from analytics with a minimum of investment -- and without expensive new hires. 

Things to keep in mind as you move forward

Data management, business intelligence and analytics do build on each other, but only to a certain extent. For example, you need data management before you start BI, but your data management doesn’t have to be completely finished before you start your BI projects. An agile method of working is recommended, especially when you’re building experience and maturity in a domain. Smaller cycles and iterations will help you adjust more quickly, and thus improve faster.

When you start evaluating tools for each of the three domains, you must ensure that your end users are confident about working with the new tools (otherwise you’ll never escape ‘the hell of Excel’). Involve them in the selection process and take their current knowledge and capabilities into account. Do everything you can to lower barriers to adoption.

Also keep in mind the integration and modularity of your data platform. For your first data management project, you’ll likely opt for a basic or intermediate tool. As you advance, you’ll run into limitations with the basic/intermediate tool and will want to replace it, or add a more advanced tool.
If you’ve prepared for this, you’ll be able move forward with a minimum of effort -- and without losing your previous work. Minimizing integration challenges and complexity needs to be your goal from the outset. That will save you a lot of work in the long run, reduce the support needed from IT, and reduce technical waste as your tools fit and work together.

For example, when you’re looking for a data visualization tool, you might want to consider choosing one with analytical capabilities because that’s one of the next steps in the execution of your strategy. Doing so will also help your users start building analytical knowledge and curiosity.

Conclusion

Your data strategy is not just a list of technology and trends that magically help prepare your organization for the future. It’s a firm foundation of initiatives and projects produced by the strategy, all geared toward helping your organization harness and exploit enterprise data.

Your data strategy includes three major domains: Data management, business intelligence and analytics, and provides both self-service (sandbox) and more industrialized (governed) capabilities that match with the levels of expertise within your company. Equipping everyone with the right toolbox will help speed things up and quickly improve the use of data within your company.

If you follow these guidelines for building your data strategy, you’ll have three key advantages:

  1. Consistency of data and business rules.
  2. Fast and easy reporting and analysis.
  3. Analytics available to all users.

If you have any questions or want me to go deeper into a specific area, please let me know in the comments below. In my next (and final) post in this series, I will help you ‘stress-test’ your data strategy.

To see other posts in this series, search the tag data strategy series. And don’t forget: organizations of any size can benefit of their data as I discussed here (see my post Self-service BI & approachable analytics for all).

Blog series Data Strategy by Natan Meekers 

  1. The Hell of Excel, or why you need to future proof your data strategy
  2. How to future proof your data strategy
  3. Building a firm foundation for your future proof data strategy

SHARE

SHARES