pharmacy image

PHARMACY

Data collection
and analysis system

Our data collection and analysis system gives pharmaceutical companies the latest news about medication distribution and directs towards what measures should be taken for achieving better performance indicators.

computer vision logo

Expertise

  • Data Science
  • Business Intelligence
  • Data Mining
blockchain logo

Technologies

  • Node.js
  • Docker container
  • AWS Cloud
  • R

Whereas there is a website that provides the information on all medications available in the country (namely, in which pharmacies they are in stock, how much they cost, where to find the cheapest offers, etc.), the pharmaceutical company felt it would be time-consuming to accumulate all necessary data manually.

They asked us to gather the information about their production and competitors and visualize the required data insights.
The idea transformed into the automated data collection and analysis system.

Challenge

Tight cooperation with pharmaceutical distributors doesn’t guarantee impressive sales across the selected regions.
A large pharmaceutical manufacturer experienced some gaps after selling their production for further distribution to target pharmacies in Belarus.

The requirements are:

To track the following metrics:

  • Price range across all pharmacies
  • Actual regional sales for any period chosen (a day, week, month, year, etc.)
  • Medicine packages delivery all over the country/in the region/in the city
  • Distribution index, i.e. product availability in different cities and regions (including its dynamic changes)
  • The drug pricing and quantity depending on the pharmacy network, and many more
  • Sales representatives’ engagement

Solution description

Our solution includes a bot developed in accordance with three concessive steps of data handling — collecting, analysis, and visualization — the result of which provides at-a-glance views of key indicators relevant to a particular customer’s objective.

The data collecting step realized through Node.js serves as a starting point of the working pipeline.
Here we took into consideration the fact of miscellaneous security levels to be encountered while collecting data from web.
To overcome the most of similar scenarios, we applied the number of appropriate manipulations with cookies, IP, and request limits.

Collecting data and organizing it in accordance with appropriate structural blocks containing title, location, and product supply and sales (product flows) info, we utilize the raw data for making use of it in the subsequent operations.

Designed to configure the data and calculate the key metrics, the analysis step can be viewed as a core for all the essential calculations.

Given the lists of medications, sales representatives responsible for the particular pharmacies/pharmacy chains, and competitors, the step includes three specific algorithms: Product Distribution Analysis, Sales Representatives Analysis, and Competitors’ Products Analysis.

The visualization step of the solution is represented by informative dashboards provided by the Tableau Online service.

The customer is free to refer to any pharmacy chain and explore a delivery status, sales statistics, index distribution, prices, and so on.
Thanks to the easy-to-use filtering system, all parameters (a period, city, region, and much more) can be updated in one click, after which the dashboard will automatically provide new numbers.

The solution presents an up-to-date picture of any medication available in the country.
Our solution also fulfils similar requests when it comes to our customer’s competitors.
On the way to increasing sales performance it could be not enough to achieve the highest distribution index — to be a step ahead of business rivals, the pharmaceutical company wants to be aware of their activities.

That is what our system supports too.

Development аpproach

Our first step was the investigation of the website security patterns in order to realize which systems our bot will need to beat.
Then we built a bot using Node.js and utilized Docker, which allowed us to easily launch and run the solution on any hardware in a cloud.

As for the bot for the pharmaceutical company, it automatically starts at 9 p.m. every day.
We can configure it to check the data non-stop or get started once an hour, for instance, depending on the customer’s request.

Results

Data is the new oil. Who gets it the first, wins.
Driven by an ambition to boost their sales and outcompete other pharmaceutical enterprises, our customer banked on the processed data and hit the target.

Our data collection and analysis system gives them the latest news about medication distribution and directs towards what measures should be taken for achieving better performance indicators.

If to look at a bigger picture, our solution is a tool for a complex market and turnover research for any required time span.
If planning and predictions on the market сonditions or price fluctuations are needed, the system will help to make them as accurate as possible.

"Lorem Ipsum is simply dummy text of the printing and typesetting industry. Lorem Ipsum has been the industry's standard dummy text ever since the 1500s, when an unknown printer took a galley of type and scrambled it to make a type specimen book"

Joran Lee (Head of Marketing)

ceo victor

Computer vision and data scientist with 15+ years of experience in R&D.
About 45 scientific publications and 15 patents. Worked for Samsung as a head of R&D department. Ph.D. in computer science. MBA degree from Open University. Extensive knowledge of image processing and pattern recognition.

CEO, Chief Data Scientist