Q: Can you tell us about Optimize Ready and why it was developed?
Jessica: “In working with different clients, we saw that Copperleaf Portfolio users need to check the quality of their investment data before proceeding to optimization, and for some users, this can be a time-consuming and manual process. Optimize Ready allows Portfolio users to quickly review the data quality of an investment or an entire portfolio of investments. It leverages our machine learning (ML) engine to provide recommendations that guide users through fixing any data quality issues.”
With Optimize Ready, users can see all the information about the quality of their investment data in one place, save a lot of time reviewing that data, and then move on to running optimizations and multiple what-if scenarios to produce higher-value plans.
Russ: “Another main driver in developing Optimize Ready has been improving data consistency. Most of our clients have multiple people creating investments which can lead to inconsistencies around the quality of those investments. Differences in estimating costs, for example, can make comparing investments at the portfolio level challenging.
Inconsistency can also have significant impacts on how investments are evaluated. Some of our clients have centralized that function to only a couple of people, but other clients have each investment owner perform project valuation. For those clients, there can be a huge variation in how similar types of projects are valued. As a result, when a portfolio owner looks at these altogether, they see valuations that have orders of magnitude difference. The goal of optimization is to identify the optimal combination of investments and timing that meet all constraints and targets and deliver the greatest value to the organization. Development of the best possible plan requires the use of consistent and rigorous assessment of the benefits of all investments.”
Q: How do organizations address data quality and consistency today?
Russ: “Organizations without Copperleaf Portfolio currently juggle investment data across multiple spreadsheets and utilize simplistic valuation techniques to prioritize their investments rather than optimizing.”
Jessica: “These organizations aren’t sure if their data is of sufficient quality so they have to manually check each investment, which is an impossible task with thousands of investments from different lines of business.
With Optimize Ready, users can see all the information about the quality of their investment data in one place, save a lot of time reviewing that data, and then move on to running optimizations to get a higher-value plan.”
Q: How does Optimize Ready help individuals and key stakeholders build more confidence in their investments and resulting plans?
Russ: “We’ve talked about how Optimize Ready improves data quality and consistency. It also enables a consistent process by ensuring everyone in the organization assesses the value of investments in a similar manner and that all business cases are reviewed in a rigorous and consistent way. For example, no one can “game” the system by proposing investments that are artificially valued too high to justify dollars they’ve received in the past.
Optimize Ready also increases transparency across an organization about the quality of the portfolio because investment and portfolio owners can all see which investments have been flagged and why. This improves confidence in the data and the process, and results in a higher-value plan.”
Jessica: “The ML engine initially uses historical data; however, as the system is used, it analyzes patterns of behavior. For example, if a certain investment or certain type of asset is always marked as a “must-do” investment, then the ML engine will learn that type of investment or type of asset is required in the plan. If a certain recommendation is taken (or ignored) each time, then the ML engine will continue (or stop) recommending that action. The power is in the learning and so as time goes on, recommendations are based on stronger evidence and therefore become increasingly relevant to your organization.
Optimize Ready offers several key benefits:
- Build confidence in your data quality so you can optimize quickly: Automate the process of ensuring data quality and consistency prior to portfolio optimization
- Accelerate time to value: With straightforward usability, facilitate onboarding and reduce the time for new users to create a strong plan
- Leverage machine learning models and algorithms: Benefit from ongoing improvements in recommendations as the machine learning engine builds upon historical and actual data”
Q: What does an organization need in terms of data collection to realize the benefits of Optimize Ready?
Jessica: “An organization doesn’t need a specific amount of historical data to gain the benefits of Optimize Ready. Even without the results of the ML engine, Optimize Ready is able to flag investments with data quality issues (e.g. “these 1,000 investments have an issue.”). The system will look at what you are already doing, and the ML engine will start learning from there. If you have investment or portfolio data and are interested in conducting optimization, you can use Optimize Ready. You don’t have to be a long-standing client to get the benefits here.”
Russ: “Organizations are already doing capital planning whether they are using Copperleaf’s solutions or not. That experience can be used as the baseline to start to build up the machine learning models and algorithms.”
Q: Can you share some lessons learned throughout the development process?
Jessica: “To develop Optimize Ready, we worked closely with our clients to understand the most common and important data quality issues that needed to be addressed. From there, we came up with four common data validation rules for our first version of Optimize Ready. We learned that although there are some common data quality issues, many clients have very specific needs, and we will keep working to make this solution more flexible as we move forward.”
Q: Can you tell us about the Optimize Ready team?
Jessica: “We are a small cross-functional team of about 10 people that spans development, quality assurance, design, and product management. Four of our clients, Alectra, Landsvirkjun, Duke Energy, and Essential Energy, were involved from the very beginning—helping us understand their challenges and validate our design prototypes.
Organizations are already doing capital planning whether they are using Copperleaf’s solutions or not. That experience can be used as the baseline to start to build up the machine learning models and algorithms.
Q: What else can we expect to see in the future?
Jessica: “I’m most excited about the continued evolution of machine learning technology within our applications and that this will help our clients produce high-quality plans. There is so much opportunity in the future to use ML in other areas of our products.
We are also exploring other potential opportunities with Optimize Ready, including outlier detection and validations, which will allow clients to add their own rules. We see a huge opportunity there for organizations to self-serve their own needs and check their own business rules.”