Frequently Asked Questions about CrossEngage

What is the Customer Operating System?

The Customer Operating System enables the development of valuable and long-term profitable customer relationships. At the core of the solution is the Customer Data Platform, a real-time solution for the comprehensive processing of customer data with cross-channel campaign management. This is complemented by a seamlessly integrated Customer Prediction Platform that automatically creates machine learning models to predict future customer behavior. This allows relevant customer groups to be selected and addressed in a personalized manner.

How are 360° customer profiles generated?

CrossEngage uses different identifiers to assign different data points to individual customers in a privacy-compliant manner. This consolidation is also known as “profile stitching” and enables the generation of comprehensive 360° customer profiles.

How does CrossEngage automate campaigns and individual customer journeys?

With our visual journey builder, you create CRM and marketing campaigns and play them out to targeted individual customer segments. With just a few clicks, you can also create complex customer journeys based on customer-specific behaviors and preferences. These campaigns can be triggered at certain predefined times as well as by specific triggers. This allows you to respond to customer behavior in real time. You can customize messages for different channels and subsegments. In addition, you have the option to exclude certain customer segments from campaigns and cap message frequency across all channels.

How quickly can CrossEngage be implemented?

Since your existing infrastructure remains intact, CrossEngage can be implemented in a matter of months. Of course, the effort required depends on the scale of the infrastructure in question. However, in all cases, CrossEngage can be deployed significantly faster than popular marketing cloud solutions, as they replace your existing infrastructure entirely.

What data can CrossEngage collect and consolidate?

CrossEngage can consolidate data from any data source available. This includes basic profile data of your customers, customer history data, and current user behavior across multiple channels in real time. Examples of data sources would include CRM systems, data warehouses, web tracking solutions, or response data from channel-specific marketing tools.

How does audience segmentation work?

With the visual user interface of our intuitive Segment Builder, you create even complex segments with just a few clicks. You simply define any conditions and get the desired segments in real-time. An example of a customer segment would be a concatenation of the following conditions: Male customers over 30 who have visited the website within the last 14 days or opened the newsletter within the last seven days.

Do I need to replace my existing infrastructure?

CrossEngage sits on top of your existing infrastructure as an overarching instance, enabling orchestration of all connected data sources and tools. We integrate your existing infrastructure and ensure future agility by allowing you to easily integrate new tools and channels. To integrate channel-specific solutions, we use APIs or webhooks through which our platform communicates with the appropriate tools.

How to create predictive models with CrossEngage?

Processing and modeling in the CrossEngage Customer Prediction Platform is based on your own customer data. This raw data must first be provided to the software in a defined structure via an upload facility, SFTP server or API.

With our model builder, you can create standardized models for common use cases along the customer lifecycle in just a few steps. All you need to do is answer certain questions about the campaign and the forecast horizon (time period of the forecast). After the check, it can be released for calculation. Based on the calculated model, targeted scorings and forecasts are created for the selected customer group, on the basis of which the selection for the intelligent playout of marketing activities takes place.

For AI experts and data scientists, we offer an advanced AutoML workbench that allows you to create any number of complex predictive models that can be fully manipulated.

How is my data protected?

As a contract data processor, we have no rights to your data. It is stored on secure servers in Europe that comply with the ISO 27001:2005 standard. German data protection laws apply, which are among the strictest in the world. Our platform complies with the new European General Data Protection Regulation (GDPR). CrossEngage has been audited by external data protection officers.

Does CrossEngage comply with the GDPR?

Yes, absolutely. CrossEngage not only complies with the General Data Protection Regulation (GDPR), but by consolidating customer data centrally, it also helps you ensure portability and easily manages and documents opt-in and opt-out processes.

What is machine learning needed for in marketing?

The use of machine learning provides the basis for decision-making through smart analysis and evaluation of customers, offers particularly precise results for direct marketing measures and a considerable gain in efficiency in the analytics process. Machine learning enables valuable forecasts about future customer behavior in order to adjust individual audiences, campaigns, channels, offers and also the entire marketing strategy in a way that optimizes sales or profits.

How does machine learning work?

Predictions of future customer behavior are calculated using statistical machine learning models. What these models have in common is that they identify patterns in customer behavior based on past data, which can be used to predict future behavior. The focus is not on a specific customer, but on the combination of characteristics that make up a customer.

These characteristics are used to create a kind of “fingerprint” that allows conclusions to be drawn about a customer’s future behavior. Customers with a similar fingerprint behave similarly, so that the knowledge gained can be transferred to other customers.

This procedure is called “scoring.” Scoring involves making a forecast for a period of time that the predictive model does not yet know. This enables a real prognosis into the future. Each individual customer is given an individual score value – which usually is a future purchase probability and a future expected sales amount.

What is automated machine learning (AutoML)?

The idea of AutoML is to automate and standardize the individual manual process steps for creating predictive models. In marketing and CRM, for example the CrossEngage Customer Prediction Platform (CPP) provides this standardization. The user of the platform can independently create a variety of predictive models from raw data (in CRM: transaction and customer master data) very quickly and with very little effort.

What is NextGen AutoML?

NextGen AutoML is the moment when machine learning automation (AutoML) advances to a system that makes proactive suggestions. AutoML is the maximum possible scaling of a large number of AI models so that they can be industrially manufactured. NextGen AutoML, in contrast, is achieved when it is not the user who has to ask and define all the forecasting questions themself (and the AI answers them precisely), but instead the AI independently and automatically shows the user opportunities and possible actions.

What data do I need for machine learning?

For pattern recognition in CrossEngage, the platform needs customer transaction and customer master data. By a transaction, we mean the purchase, cancellation, or return of a single item by one of your customers at a specific point in time. Customer master data contains an identifier (e.g. customer ID) and other personal characteristics such as age, gender or place of residence. In order to improve the model and forecast quality, online, inbound, outbound or payment activity data can optionally be added.

We only process anonymized or pseudonymized data in this process. This means that we cannot draw conclusions about individual persons from the information available to us. This fully complies with GDPR regulations.

Which use cases can be solved with machine learning in marketing?

Machine learning can be used to optimize all customer scenarios and use cases along the customer lifecycle. You can recognize the changing future behavior and preferences of your target groups at any time and react in an automated manner:

  • Customer Lifetime Value: What individual customer lifetime value are my customers likely to achieve in the future? Which customers are the most profitable?
  • Customer Acquisition: Which potential customers are likely to buy in the future and which will be the most valuable?
  • First-to-Second Order: Which of my new customers are likely to buy again?
  • Cross- and Upselling: Which customers are likely to buy which product next?
  • Churn: Which active customers are at risk of not buying in the future?
  • Reactivation: Which of my former customers are likely to be reactivated in the future?

What are the challenges in the traditional machine learning process?

To build a predictive model in the traditional way, you need both qualified personnel and the right tools to perform the necessary steps (statistical software such as SAS, SPSS, Python, R, etc.) and a lot of time. If you want to create a model for the first time, you will need not a few days, but probably several weeks or even months. Especially if errors creep in during this process.

After all, to make data accessible for machine learning, appropriate methods of data pre-processing, feature engineering, feature extraction and feature selection must be applied. These steps are followed by algorithm selection and hyperparameter optimization. All these processes lead to (manual) challenges, which are a significant hurdle for the deployment of machine learning models in enterprises.

By using the CrossEngage Customer Prediction Platform, the manual steps in the data science process become automated and scalable.

What is the benefit of AutoML?

In order to successfully compete in the market, to secure competitive advantages or even to gain new market shares, it is essential to manage marketing in a customer-centric way and to personalize the customer approach (down to the “segment-of-one”). AutoML offers the technical prerequisites for this.

With more and precise customer forecasts, customer groups can be segmented and selected in a much better targeted, more accurate and more efficient way. In addition, AutoML enables exploratory testing of the best approach in the direction of customer centricity. Personalized, customer value-oriented targeting leads to increased sales and profits through higher conversion rates, sales figures and shopping carts. If only those customers are selected who are really worth the investment of the marketing budget, advertising spendings can be used more efficiently and costs, possibly, reduced.

How good do my data science skills need to be for CrossEngage?

You don’t need data science skills to use CrossEngage and build predictive models about your customers’ future behavior – for that, we provide the Model Builder in the platform, which guides you through the process with an intuitive user interface.

If you have more in-depth knowledge in this area, the Model Builder’s “expert mode” allows you to manually set and manipulate every single parameter of a model.

Go to top