Mashed up by machine learning? Dumbfounded by data science? Agnostic about AI? Nick Frank, Managing Consultant, Si2 Partners doesn’t promise to the provide all the answers, but he can offer some crucial insight into the management process on turning your field service data into profits…
Recently I have been working with Data Scientist Eric Topham co-founder of The Data Analysis Bureau, to understand why many company leaders are struggling to turn data into profits. Eric solves data problems. He is the professional who will understand if it is a Data Science or a Data Analytics challenge and then deliver the appropriate math-based algorithms.
Data Science is about discovering new patterns in data in order to make predictions and take real-time action. The mathematical technologies used in this process are dynamic and self-learning, sometimes being grouped under the ‘Artificial Intelligence’ label. In Field Service, the types of data problems addressed by these technologies might include scheduling or predictive maintenance.
Data Analytics deals with historical and more ‘static’ data, where the desire is to test ideas or hypothesis, understand relationships and develop insights into historical patterns.
Data problem solvers such as Eric will tell you that the hardest part of his job is not developing the data solution, it is defining the problem to be solved in terms of reducing costs or increasing revenues or hopefully both.
The companies who can to articulate their business problem in terms of money and performance, make it much easier for his team to create the mathematical models to answer the questions posed.
One of the ways of defining the business problem is to use value mapping tools, such as the Value Iceberg described in February’s issue of Field Service news “Don’t be caught in the Emperor’s new clothes. First focus on the customer”.
These help companies articulate not only the direct benefits to the customer, but more importantly the hidden value of their product or service, such as improved material through-put, lower energy costs or reduced risk.
A good example would be a manufacturer of air conditioning systems who targets facility managers for whom 30% of the building’s running costs is energy. This company targets their products and services to reduce their energy by 10%, enabling a very compelling sales argument.
However, the vast majority are far blander and generally fall into three broad categories:
- Bland USPers: Ask people about their value and they will trot out a predictable unique selling point(USP) such as 24/7 spare parts delivery. The question is do they know what this means to the customer and price accordingly.
- The Easy and Obvious: Many can tell you what their customers tell them, but not much more! Do you hear phrases such as. ‘My customer needs fast and right-first-time resolution!’. What does this really mean to the customer in terms of money and performance?
- Know, but cannot say: Then there is also a significant proportion who intuitively know their customers, but struggle to move themselves beyond the immediate need. They need help to articulate how they make their customers more profitable.
If the key to monetizing the data is to never separate the business problem from the data problem, how should companies approach this challenge. Many lack the confidence to take the journey due to the intimidating jargon and fast pace of change.
This high-level roadmap is our attempt to demystify the process by breaking it down into 5 key common-sense steps:
- Define the business problem: Whether it’s internal service operations or new services, a value mapping exercise such as the Value Iceberg is the essential start point. But do not just look at the customer. Look at the end to end industry supply chain and in particular the data hand-offs between the different actors in the supply chain. We discussed this more in our 2016 Field Service news article ‘ 5 patterns to discovering new data-driven service revenues’.
- Solution and data needs: Identify the solutions you might offer, the critical data you need and how you will collect it. In their rush to create data services solutions, many companies jump to this step first without a clear view of the business problem. The result can be developing IoT platforms with no revenue stream or data they cannot analyse.
- Define data problem: Formulate and scope the problem. Then scope and design the solution. Here matching internal capabilities matched with external expert partners is often the key to success.
- Implement & evaluate: Start with a manageable pilot, revisit the business problem and ensure the solution is able to add the value you desire.
- Scale Up: When successful, you are ready to scale up across your organization
If data is particularly relevant to growing your field service business, then you can reach me @ firstname.lastname@example.org
Be social and share