Project examples

We will squeeze valuable information from your data and improve your business. See, how it is possible.


Below we list chosen specific examples of what we can improve in your company. Nevertheless, we do not want to standardize our services because our projects are always tailored to needs of clients and to specific character of the business.

Usually in our projects we are looking for answers to questions. That’s why we ask exemplary business questions to that we are able to answer thanks to data analysis and building suitable statistical models.

Majority of the projects has stated a clear business goal since the very beginning. But there happen projects where no questions are stated at the beginning. In these situations a company holding data wants to know what in these data can be found and how to use the data.

Then we use data exploration methods (data mining methods).

  • Scoring and other predictive models:
    • Predictive models predict occuring an arbitrary event, for example defaulting on a loan, reaction for a bad debt collection action.
    • Scoring for example answers a question: Which clients should we grant a loan, which clients should we refuse?
    • Fraud detection: How to detect whether some transactions are fradulent or not? How to prevent them?
    • Learn more. It is a particularly high strong point of our offer.
  • Optimization of bad debt collection activities:
    • To whose debtor call, whom should we visit to maximize profit?
    • What time should we call or visit?
    • How long should last soft collection actions? Maybe it depends of type of a customer?
    • How to price a portfolio of debts before buying it?
    • Learn more
  • Analysis of clients’ behaviour basing on information from data bases, for example in telecoms (analytical support for CRM):
    • Why some of clients stop using our services?
    • Which clients do leave?
    • How to choose clients who can be interested in other offer (cross-selling) or in bigger shopping (up-selling),
    • How to learn what clients are not glad and will leave us for the competitors?
    • Learn more.
  • Forecasting of future sales::
    • How forecast sales using more robust methods than just Excel trend lines?
    • How to detect and take into acount trends, seasonality and other important factors often occuring in time-dependent data?
    • How is possible to take into consideration in analysis holidays and different length of months?
    • What to do when it is needed to forecats many series, for example for products related to each other?
    • How correctly clean historical data from influence of marketing campaign on sales volume?
    • Is this possible to automatize a forecasting task with help of suitable or even dedicated custom software?
  • Analysis of points of sales:
    • Why effectiveness of particular point of sales: why some give better and some worse results?
    • Why some POSes sell better some products and are far worse in selling others?
    • Is it possible to group points of sales accordingly to their properties to be able to manage them easier and betters what in consequence will increase profit?
    • How optimally locate new points of sales? Is it needed to change location of the present ones?
    • How use additional external information for this purposes: income level, economic factors?
  • Profiles and segments of clients:
    • What are kinds of our clients?
    • How shortly describe these kinds?
    • Do they have other buying or shopping preferences?
    • How find behavior patterns of clients?
    • What offers should be prepared for them?
  • Market basket and buying analysis:
    • Are there any regularities and patterns in buying behavior of clients?
    • Are there products that are bought togehter?
    • Is it possible to offer to a client a personalized offer basing on buying history of this and similar clients?
    • How to recommend effectively products in online shops?
  • Analysis of textual data (text mining):
    • Often textual data are just written as open text by human: without any standardization or dictionaty.
    • Examples of such data are notes from bad debt collection activities, customer support emails, emails with diverse enquiries, any textual data from the Internet.
    • Nevertheless, valuable knowledge can be squeezed out of these unstructured data.
  • Industry:
    • How to improve production process?
    • Which factors influence on problems in a production process?
    • How to forecast product demands and request for resources?
  • Bioinformatics:
    • We specialize in applications of multivariate statistical methods to analysis of medical and biological data.
    • Proteomics in our particular area of expertise.
    • We have several years experience with a Swedish bioinformatics company MedicWave and some of the projects are executed in collaboration with it.

Try again