Starting with a blank sheet of paper we broke the project into (4) phases:
- In-depth mapping of the company’s operations, opportunities and challenges through workshops and one-to-one interviews with x-functional key stakeholders as well as review of all relevant documentation.
- Creation of more independent data following our own desktop research.
Definition of use cases. Using a rich, yet raw bundle of insights, the crucial part of the whole project was in extracting meaning from it using our data analytics experience to xxx insight and define use case:
- Failure management. - Precise knowledge of estimated operating life of machineries in the grid, allowing upgrades and maintenance with a higher level of accuracy.
- More effective investment management through fraud detection.
- Prediction of changes in the grid caused by the upcoming technological and capacity needs.
For each use case we assessed its implementation demands (tailored to current situation of the company) and its potential return on investment.
Based on these outputs, ZSD was able to prioritize the use cases.
Design of a technical solution for big data analysis tailored to company’s situation, choosing existing and proven services with focus on flexibility and scalability, instead of heavy upfront investment in the creation of new in house architecture.
With this architecture, the company can test data analytics on a particular use case and with low initial costs for near instant return on investment.
More use cases, more data and more predictive models can be added later which makes the implementation reasonable and sustainable in long terms.
Start small, think big.
Creation of a simple live demo, leveraging various types of publicly available data, that demonstrated the principle of our solution in the context of ZSD and they needs.
The demo helped to convince key stakeholders in the company, that with the right tools a functional and user-friendly data analytics solution can be delivered within few weeks.
As an „extra mile“ activity, we helped one team to solve their particular needs with a quick win solution, creating internal ambassadors of using the data for the business purposes.
What we defined as a generic use cases on the company level, then fragmented into smaller parts and found almost immediate use.