Updated May 2020
3 Ways to Add Value to Your Business With Data You’re Already Collecting
Yes, your data sets are valuable. However, like oil in the ground, its value isn’t fully realized until it is cleaned up and processed.
And just as crude oil can be valuable for transportation, plastic manufacturing, and heating, company data too can be processed to extract multiple layers of value.
In this post, we will discuss the ways in which your data can provide value for your business and customers. For information on the technology used to execute these processes such as Apache Kafka and Elasticsearch, check out the resources on our blog and website.
1) Improve internal operations
Teams work better when everyone has access to the information they need to do their jobs, and managers do a better job leading teams when they can identify bottlenecks in workflows. By pulling data from multiple pieces of software, for instance your project management program, spreadsheets, and CRM, there is greater understanding of the big picture and how multiple moving pieces work together.
We already know that pulling information from separate sources together is valuable. That’s why business units take the time to build quarterly and yearly reports. By modernizing the yearly report with daily checks and real-time reporting, the flow of information throughout the company is improved. This approach allows teams to identify what is working and not working in real-time, empowering them to pivot to another approach that will get them the desired outcome.
This need for improved operations through the better use of data is industry agnostic. We work with companies in a range of industries from IoT to government contractor to finance to SaaS, and more. Our clients are almost always looking for a way to implement real-time reporting to improve the operation of their teams and companies as at least a component of our engagement.
So what does this look like in practice? It looks like dashboards that are customized to include the key performance indicators for your team, with searching capability to look up individual projects, customers, employees, and more. It’s automation to pull together data as it is updated in individual pieces of software, creating a single source of truth that the whole team and company can refer to with confidence. And it means using threshold based alerting to notify responsible parties when specific events occur or don’t occur, giving teams critical information in real-time.
2) Add value for customers
Customers want an experience that feels customized for them, easy-to-use, and engaging. By visualizing the data your company is already collecting from customers, such as time on website or app, clicks, phone calls to your customer service department, purchases, demographics, and more you can synthesize a complete understanding of the components of your product that are working well for your customers, and identify what isn’t providing them value.
This information then naturally provides direction for your team and company without having to guess or go off of gut instinct. You have the numbers right in front of you to evaluate the best approaches for improving customer experience and retention.
This evaluation naturally leads to understanding the ROI for your customer. By analyzing how your customers use your products, your sales team can confidently relay monetary value to new customers and reinforce value for current customers at renewal time.
An important part of customer experience is the way customers interact with company representatives and the reliability of a product or technology. AT&T knows this well and is leveraging its data to improve customer service communications and identify sources of potential system failures to prevent outages. (https://www.bizjournals.com/dallas/blog/techflash/2015/10/at-t-how-we-leverage-big-data-to-improve-internal.html)
3) Monetize data
When it comes to monetizing data, there are three items to consider.
Is my data valuable to customers? If so, simple analysis and visualization could lead to an additional revenue stream for your company to sell data back to your customers. For example, a clothing manufacturer had access to the data on which kinds of garments take the most time to make. Their customers were interested in purchasing the data to design clothes that have cheaper manufacturing costs. The manufacturer then made money off of selling the data to their customers, and then had more machine time available to fulfill additional orders, creating two opportunities for increased revenue.
Are third parties interested in my data? Sometimes it’s not your customers, but other industries that are interested in the data you’re collecting. Toyota, for instance, sells the GPS data from its navigation devices installed its cars in the Japanese market to corporate delivery companies and municipal planning departments.(https://hbr.org/2016/10/to-get-more-value-from-your-data-sell-it) The fee for access to the data? It starts at $2,000 per month per company.
Could combining my data with other information increase its value? There is a rich sea of open data available at sources such as data.gov that when combined with your data provide even more value. For example, combining weather data with plant attributes and information about invasive weeds in the same area (e.g. zip code) is valuable information for farms and agri-food companies.
Using Machine Learning to Increase Data Value
In some of the examples above, simply combining and displaying the data in visualizations is enough to provide value. In other cases, such as customer ROI, some analysis needs to be completed to answer business questions.
We will have several follow on posts that discuss Machine Learning (ML) in more detail. For now, we will provide a few examples of how ML makes data more valuable.
Find correlations in the data. ML is ideal for distilling out meaningful correlations from a large data set. An example of its power comes from work we did for a movie production company. They wanted to understand what factors affect box office revenue so that they could make data driven decisions when setting budgets. By evaluating film data from the last several decades, we identified the top factors that influence revenue.
Group customers. Most products attract a range of customer personas. For instance, many retailers find that they have a split customer base with some preferring to buy products in stores and some preferring to purchase online. By sorting through the data you can create customer personas and market to those specific groups.
Look for trends. Looking for an increase or decrease in values using linear regression is a classic and helpful tool in the ML toolbox. Whether you are analyzing the performance of your technology, team, sales, or other metric you can compare your current numbers to past values.
Predictions. Trends are helpful for comparing the present to the past, and just as important is using the current data to predict future values. Finance teams have been using projections for years to estimate spending and revenue. Now all teams within a company have the power and resources to make predictions about the indicators that determine their success.
Alerting. In addition to threshold based alerted mentioned above, ML based alerting is a more sophisticated approach. With ML based alerting, the program identifies meaningful events and alerts users on those. This technology can be powerful for identifying problems early, even before traditional indicators.