This article was originally published as part of Energy Central’s December 2020 Special Issue, Data & Analytics: Creating Real Value. Please click here to join the conversation and explore other featured posts!
When you think about data, you might imagine a dizzying array of ones and zeroes, or spreadsheets filled with minute details. But as a data scientist and machine learning evangelist, I see data as a rich collection of tools and materials for solving real-world problems, and I’d like to show you why.
Like the bricks, boards, screws and sheets of drywall you’d assemble to build a house, data can (and should) be architectured purposefully to create something of value that’s safe, reliable, familiar and useful for years to come.
I call this process ‘operationalizing’ data. Anyone can do it if they want to make better, data-backed business decisions that solve customer engagement problems.
That’s because operationalized data, especially when it’s infused with artificial intelligence, is fully functional. It’s responsive and predictive. It can be used to fix problems as they arise–even if you’re not a data guru, yourself.
What can you do with operationalized data?
Our utility partners use operationalized data to achieve long-term goals. They’re increasing customer engagement with company programs and services. They’re generating new revenue streams and saving money. And they’re doing so quickly and efficiently, without having to lean too heavily on scarce IT resources.
Internal Data Inventory
Utility providers possess loads of internal data: payment history, energy usage, employee time, call center volume, grid distribution and so on.
But they need to find out whether it’s in tip-top shape to help them reach their goals. They must assess whether to sprinkle in any outside data to enhance what they already know about their customers.
Take an internal data inventory survey
We have our utility partners ask themselves:
- Is my data fragmented, with some bits collected by one department and other bits gathered by another?
- Is it difficult to share one team’s data with another?
- Do employees use multiple methods to collect and record the data?
- Do different departments use different systems to update the data over time?
- Is any important data missing?
If the answer to any of these is yes, it’s time to clean, organize and orchestrate their data.
The goal is to make all internal information accessible to and understandable by everyone across the organization. Regardless of any team’s unique goal, they should be able to put the data to work quickly to get results.
Data Quality Over Data Quantity
Oceans of data are available to us from all over the world. We can find out everything from how many dog owners live in one Census block to which consumers are most likely to respond to a Yellow Pages ad.
But just because we can get that data doesn’t mean you need to have it.
Unfortunately, sometimes people think more data will answer their problems. But more data isn’t going to magically begin generating revenue. And it isn’t going to increase customer satisfaction all on its own.
What matters is what you do with the data that you already have.
The data you just orchestrated and organized may be sufficient for solving your customer engagement problems quickly and efficiently.
With operationalized data, you may be able to answer the following questions:
- Which of my customers have missed two or more bill payments in the last six months?
- How much electricity or gas are my customers using this month as compared to last year at this time?
- How many customer calls have we received about bill payment options in the past 30 days?
Solving business problems
Being able to answer those questions with the right data means you can solve problems like:
- Knowing which customers to engage about enrolling in energy assistance programs.
- Identifying which customers may be good candidates for smart thermostats.
- Understanding whether now is a good time to invest in a new IVR system.
Enhance and Integrate
Once your internal data has answered as many questions as possible, you can begin to see where knowledge gaps exist. You can also determine what kinds of data you’d like to add into your current framework to achieve bigger goals.
Say a utility’s goal is to enter the EV market. But it doesn’t know where real opportunities exist. The utility has organized its internal data, so at this point it knows:
- who its existing commercial customers are,
- which commercial businesses aren’t buying electricity from them, and
- which of its residential customers own electric cars.
What it does not know are which local commercial businesses already offer EV charging on their properties, or how many customers EV charging brings in.
External data will tell them that.
With that knowledge, they can better assess the market and understand their opportunity for entering it.
Choosing the right external data at this juncture is critical. We’ve seen companies overspend on data that was irrelevant and unnecessary. So it’s important to buy only what you need and weave it into your existing framework.
Enriching Data with A.I.
Artificial intelligence-infused data means that a geek like me creates a complex math equation, sometimes known as a machine learning algorithm, and applies it to your data to generate answers.
Following all necessary cybersecurity protocols, machine learning generates predictive models and propensity scores utilities can use to understand things like:
- How likely is Customer A to respond to email advertising?
- Is Customer B more likely to respond to eco-friendly messaging or budget-conscious messaging when being contacted about enrolling in e-billing?
- How likely is Customer C to pay their bill in full upon receiving a 10-day warning letter?
- What type of commercial business is most likely to install EV charging on their property next?
With these insights, utilities can plan effective, targeted outreach campaigns. They know where to start engaging because the data has indicated where to focus resources first. And it’s told them what kinds of messaging will resonate most with different customers.
You’re now launching into action with effective strategies. Deriving results that fulfill your goals. Applying what you know across the organization to generate new solutions. Being able to do all of this means you have achieved operationalized data.
You have architected it purposefully and carefully. It provides safety and privacy for all the functions of your business. Your new data infrastructure is reliable for years to come. It can be added to and refreshed easily, whenever new data becomes available.
Teams across your organization are able to apply your data to a variety of different scenarios and solve problems nimbly. You’re regularly fulfilling long-term goals, like increased customer engagement, revenue generation and cost savings.
We expect to see more utilities working toward operationalizing their data in the near future. It creates real, actionable value in so many ways.
If operationalizing your company’s data to solve customer engagement problems sounds like too a heavy lift to manage in-house, BlastPoint is here to help. Take a walk through our solutions and capabilities, and please reach out to us when you’re ready to get started!