the Power to change anything.

Product Distribution Analysis

Challenge

Our client, a data science company serving the life science industry, meticulously scrutinised market data reports against pharmaceutical manufacturers’ ex-factory data. Until the project’s start, this critical comparison was laboriously executed by manual analyst efforts.

The client aspired to transform their ex-factory comparison analysis, seeking automation and heightened visibility. While analysts crafted insightful reports, the process was arduous, hindering agility and impeding code evolution. The project aimed to empower analysts with enhanced data visibility for seamless, accelerated insights—bolstering productivity while ensuring business continuity.

Our goal was to equip analysts with easily interpretable data for client presentations, preserving code quality. Despite their data wealth, the client’s pipelines demanded continual enhancement and customisation, especially during transitions from projects to established data pipelines. Aiku aimed to introduce a solution seamlessly integrating into the client’s technical roadmap, akin to a modular component within a larger system.

We wanted to create something that was truly customised to the needs of the analysts for the purpose of client servicing.

How do we engineer an advanced, automated, and user-friendly analytics pipeline and interface?
For already industry-leading data reports, how do we infuse innovation to further elevate its capabilities?

Solution

We provided a holistic solution from the design of the code stack to data pipelines and analytics tools to perform in-depth analysis of transactional data and track the flow of pharmaceutical products to identify of anomaly events.

Software Package Development: Conducted essential code redesign and refactoring, extracting specific code segments without disrupting the existing heritage code—a crucial step for modular utilisation.
Data Loading and Validation: Imported external ex-factory datasets, meticulously validating the data for accuracy and integrity, as well as matching it against internal reference data.
Automated Data Transformation: Orchestrated merging diverse data sets into a standardised data model, accommodating multiple transactional data sets—facilitating comprehensive sales tracking across wholesale, hospital, and pharmacy domains.
Advanced Data Analysis: Leveraged statistical, geographical mapping, and time series analyses to detect anomaly events within product flows, offering actionable insights vital to the pharmaceutical industry.

Result

Automated Data Transformation: Implementation of enhanced automation and streamlined data transformation, expediting report generation and insights extraction.
Modular Code Stack Redesign: Redesigned Python code into a modular structure, paving the way for reusable solutions across diverse Data Science projects.
Client Recognition: Acknowledgment with the client’s Innovation Award, affirming the project’s exceptional impact and innovation in the field.

Tech Stack delivered

Strategy: Vision and goal setting with the sponsors and subject matter experts.
Product: Tracked data on progress on Jira and all technical documentation on Confluence.
DevOps: Software package development using programming languages Python, SQL. GitLab for version control, continuous integration, and software package hosting.
Data:
- Load data from file extracted from external order management system such as SAP.
- Fuzzy matching algorithms of wholesale partners, hospitals and pharmacies.
- SQL queries to source various internal data sources.
- Data pipelines and transformation with Python (Pandas).
- Storing integrated data model in Exasol datawarehouse.
Analytics:
- Data visualisation via MicroStrategy Dashboard with live connection to datawarehouse.
- Conducted statistical analysis, geographic analysis, time series analysis.
- Drafted PPT presentations for data storytelling.

Do you want to know how we change everything?

Schedule a call and we can show you how to boost your business.