Skip to main content
Once a customer, or to be more precise, a contract, has been recognized, Enneo initiates a check whether the recognition is sufficiently specific for further processing. This concept is “Authentication” in Enneo. Recognized but unauthenticated customers are suggested to the user, but with an appropriate warning message, so that the user can particularly check whether the AI recognition was correct and all necessary characteristics were mentioned. If a customer is authenticated, this warning message disappears. Dark processing without human interaction is also only possible for customers who have been authenticated. Typical authentication rules consist of 2 independent features that must match each other. This ensures that there is no misidentification. Examples of actual identification methods are listed below. Authentication always takes place at the contract level, not at the customer level, although characteristics from the customer data can also be used for this. Thus, Enneo ensures that contract-related changes are also carried out for the correct contract.

The 4 Stages of Authentication

The customer recognition in Enneo has the following stages of legitimation:
DescriptionMeaning (E-mail)Meaning (Chatbot/Voicebot)ExampleLevelDark Processing
No customer identifiedClerk sees no customerNo recognitionCustomer does not mention contract or customer number, used email is also not registered anywhere0No
Customer recognized, but not authenticatedClerk sees customers on the right strip, but with warning messageLike no recognitionAI has recognized customer 123, whose documents attribute the email [email protected]. However, the email is from [email protected]10-19No
Customer recognized by AI and data matchesClerk sees customer on the right stripRecognition, full execution by AI agentsAI has recognized customer 123, whose documents attribute the email [email protected]. The email is from [email protected]20Yes
Customer manually confirmedClerk sees customer on the right stripNot applicableCustomer was recognized by Enneo only without authentication (level 10), but confirmed by the user30(Not applicable)

Recognition of Authentication Levels in Enneo’s Standard:

  • For Chat and Voice, the setting at which level 20 is reached is determined by the Assistant authentication instructions in the respective chat/voice bot. The default setting is contract number (contractId) and postal code (zip). Only when the combination of the two is mentioned, is the customer authenticated.
  • For Emails and Letters, the setting at which level 20 is reached is determined by a rule editor for authentications. This allows for complex security levels to be defined.

Rule Editor for Authentication in Emails and Letters

[The rule editor is currently under development] This rule editor allows for complex business rules to be mapped, which define when a customer has sufficiently authenticated to process his request. It can also be ensured that dark processing only takes place when the customer has been clearly and sufficiently identified. Examples of rules that can be mapped with this editor include:
  • Contract number OR customer number mentioned AND sender email matches with ERP
  • First AND last name mentioned AND (contract OR customer OR meter number mentioned) AND (billing / shipping address OR email matches data from ERP)
  • Total of 4 characteristics mentioned; thereof at least 1 mandatory characteristic (contract-, customer- or order-ID).
For the concrete implementation of these examples, please see below:

Procedure

Configuring rules in two steps:
  1. Create matching groups
  • A rule consists of one or more matching groups.
  • A customer is only considered authenticated when all matching groups have been successfully fulfilled.
  • Each matching group has:
    • A Name (eg “Identification per ID”)
    • A minimum number of criteria that must be met within this group
    • A selection of criteria (standard criteria from enneo and optionally your own criteria)
  1. Select criteria for each matching group
  • Within a matching group, you choose the criteria to be checked.
  • The group is considered fulfilled if at least as many criteria apply as specified under “Minimum number”.

Available criteria

Enneo has already developed test criteria for the typical criteria, e.g. the mention of a contract or customer number. These can be simply clicked on. The respective data fields, i.e. the specific contract number for a review on mentioning the contract number, must of course be made available as defined in https://docs.enneo.ai/de/guides/customer-recognition/integration-of-customer-data-in-enneo. The corresponding name of the variable is also mentioned in the following table.

Standard criteria

Standard criterionUsed data fields for matching
Sender email matches with data in ERPemail in one of the customer’s contracts
Email mentioned in the message matches with data in ERPemail in one of the customer’s contracts
Phone number matches with data in ERPphone in customer or contract data
Contract ID mentionedContract ID (from ERP/contract data)
Customer ID mentionedCustomer ID (from ERP/customer data)
First name mentionedfirstname in customer or contract data
First name or initial of first name mentionedfirstname in customer or contract dataLast name mentionedlastname in customer or contract data
Exact delivery address mentioneddeliveryAddress
Exact billing address mentionedbillingAddress
Similar delivery address mentioned (≥ 80% match)deliveryAddress
Similar billing address mentioned (≥ 80% match)billingAddress

Custom (user-defined) criteria

If in addition to the standard criteria, you would like to use bespoke characteristics from your ERP (e.g. order number, meter number), you can set up custom criteria. For this, you need to define:
  • The name of the criteria (e.g. “order number specified”)
  • Data path/variable, which Enneo should search for (must correspond to a field in your customer/contract data; dot notation like rawData.orderData.number is supported)
  • Display text in case of failure (e.g. “Order number 0 was not found.”; 0 will be replaced by the sought value)

Individual determination of legitimization levels

Different from the rule editor, customers can also specify rules according to a completely free logic, determining at what point a certain level should be reached, thus implementing customer-specific rules. As part of this, additional legitimization information can be obtained via API. To do this, under ERP integration in the user code for **Feature-based search of contracts **, not only the customer and contract number need to be returned, but also the field
customerLegitimation
Here is an example of a user-code return for a legitimized customer:
{
  "contractId":"1234"
  "customerId":"Cust-1232131"
  "customerLegitimation": 20 // Customer is securely legitimized: Dark processing for e-mail is possible; Full access to AI agents is possible for chat and voice
 }
Here is an example of a user-code return for a non-customer. Here a warning message will appear for the clerk.
{
  "contractId":"1234"
  "customerId":"Cust-1232131"
  "customerLegitimation": 10 // Customer suspected, but not certain
  "customerLegitimationMessage": "Customer has mentioned the correct contract number, but the name does not match"
}
The customerLegitimation field defines what warning signs should be shown to the user.