Do you build a car in your garage? A few rare folks still do but not so common. This was not the case a century ago when cars were just showing up on the horizon. 

We are seeing the same trend in digital tools. It was easy to build simple software a decade back over a weekend and move it to the production stage. The software was also self-contained and straightforward. This is no longer the case now. Also, it is increasingly becoming difficult for a person to be proficient in all aspects from the front end to the back end to distributed architectures, being cloud-native, aware of all DevOps, and many other things that go into building a digital solutions today. 

As part of Entytle’s journey, we see a lot of Industrial OEMs still trying to do a lot of things in-house which soon become self-prophecies. I am not saying that nothing should be built in-house. There are still a number of scenarios that ideally should be done by internal teams. I have dealt with this subject in detail in a white paper.

To make aftermarket growth data-driven, it needs to pull multiple data sets from various parts of the organization. The data sits across multiple data sources from multiple ERPs to CRMs and even mainframes. Do I hear spreadsheets and handwritten notes? Aftermarket data challenges are unique compared to new equipment sales. New equipment sales deal with just the latest generation of equipment whereas the aftermarket has to deal with all the equipment which are operational in the field. They could be a year old, a decade old, or even multiple decades old. So how to deal with such a humongous data challenge?

Data flows in an Industrial OEM

The ability to handle data & AI/Analytics at scale is the KEY in the new world. But how does one do it? What is a data journey before it becomes useful for the organization? What tool sets are needed?

After working through many industrial OEMs, a typical data pipeline goes as follows

Data flow pipeline

Needless to say, each of these stages needs different kinds of tools and different skill sets. If a custom app needs developers well-versed in application programming and full-stack development, AI needs folks with data science skills. I am not even touching the needs of folks with product and requirements management skills. And we all know how difficult it is to hire good resources. 

To be successful, what is needed is a comprehensive approach to data problems so that from sourcing the data from their respective systems they are cleaned and unified so that it helps in decision making. 

As part of aftermarket selling what one needs:

• Where is my installed base and what opportunities are available?

• What are other cross-sell and up-sell opportunities?

• Who are my at-risk customers and how can I stop the churn?

• What happens when my best aftermarket sales or services person leaves or retires?

• I have a limited workforce. How can I prioritize the efforts for maximum ROI?

Digital transformation is not a choice in the modern world but at the same time, you should know how to execute that transformation reliably so that it takes you in the direction of autonomous aftermarket i.e. scalable, and fully automated.