Banking One
Predictive analytics: simple, efficient, profitable and successful

Banking One

As a pioneer, Banking One brings the spirit of innovation and the state of the art of modern and forward-looking technology, which can otherwise only be found in Silicon Valley, directly into the banks of Europe. Following the example of global players such as Alphabet, Meta, Amazon or Tesla, we believe that the right mindset in combination with an outstanding team can change the market and the customers‘ awareness of technology. As a fintech start-up, we have been revolutionised the world of AI-assisted sales in banks for several years with our Predictiv Analytics software. The success of our clients confirms Banking One’s unique approach.

Recognise the behavioural patterns

Real Mashine learning

Banking One AI is designed to learn and adapt autonomously without following explicit instructions. Using algorithms and statistical models, patterns in data are analysed and conclusions are drawn. By permanently adding new data sets, the results become continuously more accurate, the forecasts more targeted and the probability of closing in the appointment higher. The AI recognises the most promising customers:

  • Right customer
  • Right product
  • Right time
  • Right address

Current bank selection

Selection according to a few classic criteria

Customers are often selected for sales campaigns according to well-known schemes. Classic selection criteria are, for example, age, household size, gender, income, disposable income or current product use. From these selection criteria, banks usually face the hurdle of making inaccurate predictions about future connection probabilities. The approach to customers is less precise and the best possible conclusion probability cannot be achieved. Ultimately, the appointment and closing rate suffers as a result of the waterfall approach. What is needed is a customer-specific approach that is also demand-oriented. 

The advantages with Banking One

Selection of the most promising clients

Using AI-based model analyses, millions of data records are used for pattern recognition and accurate lead scoring is created. The scoring makes it possible to address customers with a very high product affinity directly and personally. This maximally increases the appointment rate and the probability of closing a deal. By directly addressing specifically selected customers, it is possible to save considerable effort in the customer service centre and advisor activities. If used correctly, it is possible to reduce the workload in the KSC by up to one day.

Use millions of data sets

Data Science with b1

Account transactions, portfolio characteristics, life events and much more generate millions of data records. Due to Banking One`s KI, it is possible to understand and use this pool of valuable information and to convert it into bank-specific (sales) success according to the respective requirements. For targeted and accurate pattern recognition, diversity is just as relevant to success as the quantity of data records. Only with many different data sets can the models be trained more precisely and in a more targeted manner. This significantly improves the quality of the analysis results and forecasts of Banking One. For this reason, a cross-institutional and cross-bank, anonymised data pool forms the basis for the success of Banking One. Numerous A/B testings in comparison to data analytics solutions from other providers confirm the approach and superiority of Banking One.

Data Lifecycle

From the data set to the lead score.

1. Data collection

Data collection

Data is retrieved from the banking system via an interface. In a further step, these data records are transmitted to Banking One as pseudonymous data records. The privacy of your clients will be respected at all times in accordance with the applicable regulations.

2. Data preparation

Data preparation

All data supplied is brought into a uniform form and all possible references to real persons are additionally anonymised.

3. Application of the models

Application of the models

Models are selected on the basis of the available data. The subsequent processing of the data is carried out according to individual bank application scenarios.

4. Training


The training of the models (machine learning) on the basis of the data sets is carried out.

5. Evaluation


In the evaluation phase, the existing models are reviewed with regard to their performance. The continuous optimisation and adaptation of existing models guarantees satisfactory and best possible results.



The data provided is checked and evaluated by the AI, which assigns an individual rating for each data set.

7. Delivery of the data

Delivery of the data

The processed data is made available in our web application in a mandate configured specifically for your bank. Likewise, the results transferred into a lead scoring can be directly imported into your core banking system via an interface.

Simply understand data

b1 Scores

Banking One analyses the behaviour of your customers and forecasts their needs. The evaluations are stored as a score between 0 and 99. The higher the score, the greater the probability of closing a deal. For better visualisation, the score is displayed in the form of coloured dots in the colours red, yellow or green.

Use data simply

b1 Affinities and filters

In order to be able to control campaigns, certain categories are evaluated by Banking One and displayed as a score. The software offers a wide range of affinities in various special fields. In addition, it is possible to have your own affinities created in the software by Banking One.

  • General
  • Hedging
  • Real estate
  • Assets
  • Precaution
  • Liquidity
  • More:
  • Validation
  • Assets
  • Precaution
  • Real estate
  • Liquidity
  • More

The accurate identification of life events and significant changes offer the ideal starting point for (holistic) advice.

Increase sales performance

Efficiently manage customer relationships

Banking One AI is designed to learn and adapt autonomously without following explicit instructions. By using algorithms and statistical models, it can analyse behavioural patterns in data and draw conclusions. By constantly adding new data sets, the results become continuously more accurate, the forecasts more targeted and the probability of closing a deal in the appointment higher. The AI recognises which customer might be interested in which products and services at which point in time.

Example from practice

Consumer credit

During the three-month campaign period, the entire commission result in the field of consumer loans of the year 2021 could be generated for the bank.

Appointment rate

Graduation rate

2/3 of all degrees
from b1 forecasted

4 x
commission income

Banking One suits you

Fully integrated or standalone

Banking One is used by banks in different ways. Whether adapted to your system or standalone in the Banking One webapp. Thanks to our state-of-the-art technology and modular functionality, you can use Banking One exactly the way you want.

Fully integrated

into your system

Banking One integrates into your system. Regardless of the software you use, it is possible to optimise Banking One for the software you use. For example, you can use bank-specific data or other interfaces of the core banking system to import data records from Banking One into your system.

Banking One Webapp

fast implementation without integration effort

With our intuitive webapp, you use Banking One as a stand-alone application. Besides the integration into your system, it is another possibility to use Banking One in full extent. The webapp allows you to rate records in categories with scores. Adjust your campaign parameters via the graphical interface and export the results in various formats for further use.