Return on investment (ROI) is top of mind for everyone. With so many competing priorities, how you spend your time and money, and what you get for it, matters more than ever.
The focus used to completely be on the level of your investment. But the paradigm is shifting because of the data capabilities that now exist.
In this article, we’ll explore how the definition of ROI has changed because of modern technology and approaches, where your data ROI comes from, and how to accelerate it.
Setting goals for your data analytics efforts
Data without analytics is ultimately an investment without return. Most organizations sit on troves of data, but can’t do anything with it. But analytics is a progression. Each of the levels on the data analytics maturity model represent questions you can begin to answer. To go from nothing to cognitive takes a lot — your investment increases substantially.
Descriptive: What happened?
Diagnostic: Why did it happen?
Predictive: What will happen?
Prescriptive: How can it happen?
Cognitive: What can be suggested?
With each step up the model, you add more information and complexity.
For example, the descriptive level of maturity can be answered with a look at history. As you progress, you will need more information and stronger data relationships to better understand the “why.”
Your data quality and integrity are also important. When you get to the cognitive step, you’re expanding outside your universe of data, and the contextual element of what you’re doing gets broader. For this step, consider Microsoft’s Cortana or IBM’s Watson.
But in the modern data world, there doesn’t have to be a huge upfront investment. Shifting your focus from a return on investment to a return on insights can drastically impact how you invest in your data and your results.
Calculating the ROI of data and analytics projects
Ultimately, ROI is realized from leveraging effective data management to enable access to:
- more and better data
- maximizing visualizations
- advanced analytics
- actionable insights for outcomes
For data management, that means:
- Improved quality and completeness
- Confidence and trust
- Accountability through governance
- Improved stewardship
- Advancing culture change to help stakeholders understand the importance and value proposition
By improving your data management, the insights from your data become better and more actionable, including:
- Access to more data and the inclusion of new sources
- Faster and easier access to data
- Greater integration of disparate data
- Easier standup and use of analytics technology
In years past, insights from data analytics might have been limited to data scientists or experts in the field. But now, with analytics tools and technologies, data insights are useful to — and actionable for — people across the entire organization.
The larger the investment in time and money, the more emphasis on ROI, how quickly it can be realized, and the amount of trailing value.
Data leaders have to work with their organizations to understand what the best strategy is – whether that be a smaller investment with a slower return or a big investment that allows you to realize your ROI sooner. It is critical to evaluate your organization’s needs, expectations, and goals before making decisions on strategic data management.
Understanding the classic data ecosystem
In a classic data ecosystem, the setup might look something like this.
In a classic data ecosystem, it requires deep analysis to understand sources and the definition of data, and a considerable amount of time and effort to reach the gold standard you need for your data to be utilized for BI and analytics. Investments are required on all layers. There is no real way to invest in one aspect of your data and analytics and still find value.
There is also significant effort required to ensure that as you introduce new systems, you don’t break legacy systems and processes already in place. Quite often, work must be done upfront to ensure that changes (even upgrades) will not cause disruption.
In addition, significant “time-to-market” factors need to be considered with classic data ecosystems. Often, the slow delivery of data and features forces businesses to make incremental changes without undertaking any kind of larger project. Doing so might be helpful at first, but can cause issues later.
With a slow delivery of data, many organizations using a classic data ecosystem find that they are unable to keep up with the pace of business today. Classic data ecosystems are often built to meet reporting needs, not analytical needs and the analytics piece is a one-off project. The deployment and incorporation of analytical models into production in a classic setup requires a considerable amount of time and customization.
Building a modern data ecosystem
In this more modern data ecosystem, there is a more layered approach.
Now it is much easier to gather, ingest, and integrate the data, and bridge gaps between systems, along with including new concepts like data lakes that can include data layers at the bronze, silver, and gold levels. You don’t have to invest fully in all of the layers, you can invest where you need to. You still have BI & analytics capabilities, but you have more of an application integration framework that serves additional needs. And then there is the trust of the data. This setup allows for more flexibility and customization for all parts of the organization.
Read more: How to build your data analytics capabilities
The value of incremental change
Your investment doesn’t have to be an all-or-nothing proposition. You can incrementally build out components and capabilities and can make data available for exploration without deep upfront analysis that often slows everything down.
Additionally, you can control the degree of your investment in a significant way. Without having to push data through all the layers to make it useful and the use of flexible architecture, you don’t have to make a significant investment and change to make it worth it.
You can also leverage external tools in the interim. Service and subscription-based features allow for fast initiation, and exploratory efforts can be stood up and torn down easily and quickly. New technologies and design/development paradigms enable faster adoption overall. And now, more user groups are able to access data and analytics, create more use cases, and make business decisions on the insights.
Ultimately, it is time to shift your thinking on ROI and leverage modern data technology and tools and focus on the return on insights, intelligence, and innovation.
For more information on data, analytics, and assigning ROI, Fusion’s Vice President of Data, Saj Patel, recently spoke at the CDO Summit. His presentation details how to accelerate ROI and gain buy-in across your organization. Watch the recording here and connect with us if you have any specific questions
