Today’s A:360 discusses a few suggested ways to measure the return on investment (ROI) for your analytics initiatives. A common question I receive is “how do we determine the effectiveness of our analytics efforts?”. This podcast’s intent is to present a few possible ways to answer that very question about measuring analytics ROI.
Welcome to today’s A:360. My name is Brewster Knowlton, and today we’re going to be talking about some ways that you can measure return on investment for your data and analytics initiatives.
As analytics initiatives become much more commonplace in even the smallest organizations, there will always be the question: “How are we evaluating whether or not we are getting a strong analytics ROI – a strong return on our investment, for both time and resources, in our analytics initiatives?”
We’re going to cover a few different ways to measure this. These are good starting points, by no means is this going to be a compressive or exhaustive conversation about analytics ROI. However, it’s going to be something that gets the conversation started, while giving you a few ideas to get going with your own analytics initiatives and ROI calculations.
My starting point for anything related to analytics ROI is going to be, “How can I reduce the time it takes my staff to be able to produce reports or some data-driven analysis?”
Reduce Manual Reporting Efforts
I find that (we will look at this in the context of a credit union or a bank) for a financial institution with a billion dollars in assets, there are between 4,000-6,000 hours (minimum) that can be automated through the use of improved reporting and analytics. This would most likely be through the integration of data and/or the automated extraction of data from applications that don’t allow for easy data extraction.
This is what our previous podcasts and articles talking about data inventories or report inventories are getting at. That is, how much time is it taking on a monthly or weekly basis to produce those reports?
Once you have that time calculation, you can back into the opportunity cost of your staff manually producing these reports. You can do this by taking the percentage of their time spent (in a month, let’s say) on producing what could be automated reports, multiplied by their compensation and benefits expense. Then, you are able to get a number that’s telling you how much it costs your staff to do manual continue to gather data and produce reports. For organizations in the first 18-24 months of their analytics initiatives, this is where you’re probably going to get the biggest bang for your buck and the biggest return on investment. Frankly, you’ll also make a lot of friends in the process as you’re going to give people a lot of time back per week or per month.
Measuring Access to Data and Analytics
The next way that I would recommend you start looking at analytics ROI is by measuring the overall access to data and analytics throughout the whole organization. What we tend to see is that there are pockets of information – silos – throughout an organization where data may or may not be shared and spread throughout the rest of the organization. What this does is creates very limited insights into the organization as a whole as it relates to anything operationally or in the context of analytics.
A way to quickly identify the spread of analytics usage is to measure how many individuals not only can access essential BI portals (usually the front-end to your analytics platform) but also how many people are accessing it on a regular basis. Especially those in roles where having access to and consistently using data is going to be a critical component of success in their jobs in a more data-driven organization.
This is a point that isn’t necessarily financially-driven. (How many CFOs listening are saying to themselves that you only like financially-based ROI calculations?!) But, as we start to talk about overall utilization of any product that we acquire or implement, we need to consider how well it’s actually being used throughout the organization. If two people are using it out of an organization of five hundred, our analytics product penetration is very low. While not necessarily financially-driven, it is a way to measure the overall impact of your analytics platform and initiatives.
Measures of Self-Service
You’ve heard me say this before if you’ve listened to previous podcasts or read our articles: your analytics program should be centrally-driven and broadly distributed. What that does NOT mean is that your analytics team becomes a series of report writers where they’re, essentially, order takers from business users that need data.
So, another way to measure analytics ROI is to analyze how many reports or dashboards visualizations are in your BI portal that have been created by the analytics team and how many of those reports have been created by the business users themselves. This becomes a measurement of the self-service capability of your analytics platform. Again, this is not necessarily financially-based from an ROI perspective, but as we look at overall utilization, you really want to have a platform that enables the business users to get data on their own. If your analytics team is required to constantly create all the new reports and analysis, you’re not going to be able to scale as the data needs and analytics requests rise. Therefore, this self-service piece is a very important and integral component of the success of analytics program and initiative. This metric directly measures the success or failure towards that objective.
Benchmarking Analytics Lift
The last point that I’ll make about measuring analytics ROI – again, this is by no means and exhaustive list but just the starting point for the conversation – is that you can measure analytics ROI and determine the impact of analytics through the benchmarking process.
For example, let’s look at a lending example. Suppose, before you were able to dive into your underwriting and origination data, you had a 5.5% average yield for your consumer loan portfolio. Now, after investing in analytics and you’ve gathered and integrated your loan origination data and loan servicing data, you found there were more opportunities to underwrite loans with lower credit quality borrowers but maintain your same delinquency and chargeoff ratios. In this example, you are able to make more money with a reduced risk or equivalent risk portfolio. After all this, instead of having a 5% yield, you might be at a 6% yield. All of a sudden, you’re making an extra 50 basis-points on loans – directly contributing to the bottom line.
That benchmarking comparison of, “What did we do before analytics?” versus “What did we do after analytics?” is just one way that you can start to show the value. There are going to be a million scenarios where benchmarking applies. Look at your credit card portfolio and look at the number of transactions. Perhaps a dive into the data helps you develop a gamification-based marketing campaign that has an emphasis on signature-based debit card transactions as opposed to PIN-based transactions. This would lead to more interchange income on each swipe. Benchmark the before (i.e your control group) to the “after”. Try to explore the different ways that you can use benchmarking as a way to determine the return on investment or the impact of analytics on your operations.
As I’ve said, this is by no means an exhaustive list for measuring analytics ROI, rather it is just a way to get some ideas flowing about how you can measure the impact and return on investment for analytics.
We talked about measuring analytics ROI through:
- Reducing employee time to create reports, dashboards, and other data-related tasks
- Measuring access and utilization of analytics and the BI portal
- Measuring the percentage of reports and analytical efforts created by the business users vs. the analytics team directly
- Benchmarking your time in the “before analytics” period versus the “after analytics” time.
That’s it for today. Thanks again for listening to today’s A:360.