Now every business is an eCommerce: I need to analyse data, run campaigns, increase conversion, reduce cost per click, decrease churn rate… Which tool can do all at once? (CMO)

Use Cases & Pain Points Addressed

This tool solves the following pain points, or greatly reduces their impact:

Data analytics projects are expensive, and results are not known.

With this tool...:
  • Quick trials are possible, at no cost.
  • Results can be seen, before the project starts! (you get what you see)
  • There is neither need for IT support to integrate the results, nor for a BI Tool to visualize the results. Output shows immediately to the user, with clear, easily understandable parameters and intuitive charts.

Every business user can perform machine-learning analyses on a prepared sample of tabular data of any kind (e.g. about customer data, products, company, sensors...).

The complexity of algorithms and scores of Python and R codes have been translated into an easy-to-use application with comprehensive dashboards. No need for sophisticated theoretical background.

With a few clicks, and within a few minutes, data can be uploaded and processed to extract future predictions & newly found business parameter values.

Churn, Cross-sell, Fraud and many more data-driven predictions for relevant business questions can be addressed without training or IT-induced latency.

Purchase propensity, specifically for eCommerce businesses: predict your visitors & customers.
  • Improving marketing ROI (+30%): by predicting eCommerce visitors, therefore have a better targetting in campaigns.
  • Expanding the target audience itself (+35%): better identification of the visitors that have higher conversion than “add-to-basket visitors”.

Examples of business scenarios and achievements (non-exhaustive):
  • In a Fortune 150 Payment Services company, revenue loss from churn was reduced by over 70%. It clearly identified churn characteristics and allowed to take preventive action.
  • A retail chain gets real-time (predictive) information about the next best product recommendation, at the cash-it time, through API integration.
  • Government's effort to fight business fraud and smuggling was predicted with over ten-fold accuracy compared to previous methods.
  • For an OTA, target groups for a specific e-mail campaign could be predicted & segmented, leading to click rates 17% higher vs. a control group...

Key Features & Differentiators

The following features and advantages should be noted:

An intuitive & non-technical user interface, helps focusing on business value instead of technical details: “Self-service analytics”, rather than relying on data experts or consultants… It unifies analysis and marketing in one platform.

Simplicity: the platform was designed by predictive modelers, so the modelling flow is smooth.
  • It has a unique responsive app interface: one can log in and build models on a smart phone or tablet.
  • After uploading data, a fully transparent machine-learning model can be built in 3-clicks.
  • Advanced users can then move on to more advanced features and settings in case of need.

None of the competitors are as interactive and transparent:
  • None support playing with parameters in what-if scenarios... literally while sitting in a CXO meeting.
  • None provide trials or else do not open the platform to public for unsupervised use.
  • More than "try & buy": the tool's web-platform is live, ready to sign-up and use. Only one competitor provides such a service (though via a download to local pc).

Predictive analysis-related results are augmented with plug-ins to support marketing towards target audiences, by feeding the results directly into marketing tools.

Support of marketing & targetting:
  • Uses AI-based model.
  • Generates automated customer segments, to be connected with marketing platforms (e.g. GoogleAdd, FacebookAdd, etc), whereas other tools don’t use AI-based models or select targets manually.

No IT footprint: the platform can be used over the cloud and web, unless required by the policies of the enterprise.

Costs/Expenses: competitors are either more expensive and/or have fewer capabilities.

Performance benchmark:
  • In one large customer, prediction of customer churn has an error variance of only 1%.
  • 10 minutes are enough to learn from 1 Mio records (data points).
  • For smaller data sets with less than 100K records, it runs in seconds.