The Shift Synchronization feature integrates Microsoft Teams' Shifts with the Formal Agents of Social Queues on the ROGER365.io platform.

When the Shift Sync feature is enabled, and an Agent in a Social Queue is matched to a corresponding Microsoft Teams user with a defined Shifts schedule, Microsoft Shifts becomes the primary source for managing that agent’s availability during their scheduled shift times.This means that when the agent is On Shift, their availability is controlled by Microsoft Shifts and not by manual status changes in ROGER365.io.

Synchronization of the Shifts runs each 10 minutes.

‍

Prerequisites:

1. Create a Team in Microsoft Teams (or use an existing Team where you have Owner permissions).
2. In Microsoft Shifts, create a Scheduling Group and define Shift records for the assigned users - Microsoft reference.
   NOTE: Remember to share the created Shift records with the assigned users!
3. Authorize the Shift Synchronization Application on the General admin page.

How to enable Shift Sync?

After confirming that all prerequisites have been met, in ROGER365.io admin portal navigate to "Queues" page (under Social Messaging section) and open settings of desired Social Queue where Shift Sync feature has to be enabled.

Scrolling down through Queue's configuration items will expose Shift Sync section:

• Switch “Enabled” toggle
• Select created Team
• Select Scheduling Group (each Team can have multiple scheduling groups)
• The optional "Enable Verbose logging" switch enables detailed logging that can help the ROGER365.io support team investigate potential issues with the Shift Synchronization feature.

Save Queue's configuration to apply Shift feature, afterwards open Queue settings again and press "Sync Now" to make it effective.

Once synchronization is complete, you will be able to see which agents are managed by Microsoft Shifts:

Formal Agents who are not part of the Scheduling Group will not be managed by Shifts.

Behavior in Contact Center application

When an agent’s shift starts, they are automatically signed in to the Social Queue that has the Shift Sync feature enabled. Neither Agents nor Supervisors can manually log out from the Queue while their Shift is active (On-Shift flag).

After shift ends, the On-Shift flag is disabled, causing the agent to be automatically logged-out from the Queue. Once logged out, agents regain the ability to manually manage their Formal status:

SocialConversationInfo Export

This API endpoint retrieves main details of social conversations for a given time range.

To extract data from this section you have to create API Key with ‘DataExport‘ scope.

/v1.0/contactcenter/export/SocialConversationInfoExport

‍

‍

Example

{
  "items": [
    {
      "tenantId": 9999,
      "id": 2135,
      "orchestratorConversationId": "b3cfd762-1234-9876-b3f6-3b2d7e9794db",
      "assignedToAgentId": 93,
      "firstAssignedAt": "2026-07-06T06:52:45.9654328",
      "teamsLink": "https://teams.microsoft.com/l/message/19:22229675d5bc1111ab12bf49acedf9a2@thread.tacv2/2137321069874",
      "activeWithAgent": false,
      "inQueue": "2026-07-06T06:52:15.9129548",
      "socialQueueId": 81,
      "waitForAgent": false,
      "lastMessageReceived": "2026-07-06T07:29:02.6543291",
      "lastQueueActivity": "2026-07-06T07:29:02.9644125",
      "queuePrio": 5,
      "lastMessage": "Thanks for your support!",
      "lastMessageRead": true,
      "teamsEndpointConfigId": 4503,
      "picUrl": "https://admin.roger365.io/Content/Teams/images/AdaptiveCard/Email.svg",
      "participantId": "user@company.com",
      "participantUri": "mail:user@company.com",
      "ended": "2026-07-13T07:30:02.5878251",
      "note": null,
      "firstAgentMessageAt": "2026-07-06T06:56:49.9654328",
      "agentMessageCount": 8,
      "userMessageCount": 12,
      "useAutoAssignment": true,
      "isDeleted": false,
      "lastModified": "2026-07-13T07:30:02.5879886+00:00"
    }
  ]
}

‍

Model

‍

‍

SocialQueue Export

This API endpoint lists all currently configured (or past) Social Queues.

To extract data from this section you have to create API Key with ‘DataExport‘ scope.

/v1.0/contactcenter/export/SocialQueueExport

‍

‍

Example

{
  "items": [
    {
      "tenantId": 9999,
      "isDeleted": false,
      "lastModified": "2026-04-09T07:02:27.115628+00:00",
      "id": 81,
      "queueName": "Social Queue",
      "statisticsResetAt": "0002-01-01T02:00:00",
      "powerFrameUrl": "https://PowerFrameURL.com/?PhoneNumber={phonenumber}&Company={company}",
      "powerFrameLabel": "PWF-1",
      "secondPowerFrameUrl": " https://SecondPowerFrameURL.com/?Name={displayname}&Email={mail}",
      "secondPowerFrameLabel": "PWF-2",
      "useAutoAssignment": true,
      "lastAssignedAgentId": 64
    }
  ]
}

‍

Model

‍

To extract data from this section you have to create API Key with ‘DataExport‘ scope.

This API allows to retrieve information regarding calls that has been set to Virtual state either by Make Call Virtual action or Outbound Call Campaign API.

Virtual Call Info Export

GET /v1.0/contactcenter/export/VirtualCallInfoExport HTTP/1.1
Host: api.roger365.io
Auth-Key: <authorization key>
Content-Type: application/json

‍

‍

Example

{
    "items": [
        {
            "tenantId": ,
            "id": 279,
            "virtualDateTime": "2026-05-22T08:29:00.8547471",
            "callId": 30452,
            "requestId": 20813,
            "originalCallId": "b3f3641b-c684-4894-be53-a0327a47509b",
            "callOutDateTime": "2026-05-22T08:29:21.7001162",
            "connectedDateTime": "2026-05-22T08:29:32.7152562",
            "target": null,
            "isDeleted": false,
            "lastModified": "2026-05-22T08:29:32.7152973+00:00"
        },
        {
            "tenantId": ,
            "id": 280,
            "virtualDateTime": "2026-05-22T08:32:29.1374994",
            "callId": 30453,
            "requestId": 20814,
            "originalCallId": "9ba66067-2213-4d6b-a320-03d022f3d2f5",
            "callOutDateTime": "2026-05-22T08:34:16.3830255",
            "connectedDateTime": null,
            "target": null,
            "isDeleted": false,
            "lastModified": "2026-05-22T08:34:16.3830716+00:00"
        },
        {
            "tenantId": ,
            "id": 281,
            "virtualDateTime": "2026-05-22T08:54:38.3463646",
            "callId": 30455,
            "requestId": 20816,
            "originalCallId": "02005180-61a1-4985-99d9-6cf04534a0e1",
            "callOutDateTime": "2026-05-22T08:54:46.419894",
            "connectedDateTime": "2026-05-22T08:54:55.0994419",
            "target": null,
            "isDeleted": false,
            "lastModified": "2026-05-22T08:54:55.0995076+00:00"
        }
    ]
}

‍

Model

To extract data from this section you have to create API Key with ‘DataExport‘ scope.

This API allows to check results of Quality Assurance Action of the AI Flow feature.

GET /v1.0/AI/export/QualityAssurance HTTP/1.1
Host: api.roger365.io
Parameters: startFromDate
Auth-Key: <authorization key>
Content-Type: application/json

‍

Example

{
    "items": [
        {
            "tenantId": ,
            "id": 258,
            "createdAt": "2025-02-26T14:47:18.2736993+00:00",
            "isDeleted": false,
            "lastModified": "2025-03-10T16:14:25.3073564+00:00",
            "result": {
                "total_score": 2,
                "pass": "No",
                "flagged_for_review": "Yes",
                "detailed_feedback": [
                    {
                        "question": "Did the agent greet the customer appropriately at the start of the interaction? (Yes/No)",
                        "answer": "No",
                        "comment": "The greeting was unclear and informal."
                    },
                    {
                        "question": "Did the agent introduce themselves and their role clearly? (Yes/No)",
                        "answer": "No",
                        "comment": "The introduction was vague and lacked clarity."
                    },
                    {
                        "question": "Did the agent correctly understand the customer’s issue or request? (Yes/No/Partially)",
                        "answer": "Partially",
                        "comment": "The agent acknowledged the request but did not confirm understanding."
                    },
                    {
                        "question": "Did the agent provide a clear and complete resolution to the customer’s issue? (Yes/No/Partially)",
                        "answer": "No",
                        "comment": "The interaction did not lead to a clear resolution."
                    },
                    {
                        "question": "Was the agent’s language clear, professional, and free of jargon? (Yes/No)",
                        "answer": "No",
                        "comment": "The language used was informal and unclear."
                    },
                    {
                        "question": "Did the agent maintain a positive and empathetic tone throughout the interaction? (Yes/No/Needs Improvement)",
                        "answer": "Needs Improvement",
                        "comment": "The tone was somewhat positive, but lacked empathy."
                    },
                    {
                        "question": "Did the agent follow the correct process and company guidelines? (Yes/No)",
                        "answer": "No",
                        "comment": "The agent did not follow a clear process."
                    },
                    {
                        "question": "Did the agent properly verify the customer’s identity, if required? (Yes/No/N/A)",
                        "answer": "N/A",
                        "comment": "Identity verification was not applicable."
                    },
                    {
                        "question": "Did the agent summarize the solution before closing the interaction? (Yes/No)",
                        "answer": "No",
                        "comment": "There was no summary provided."
                    },
                    {
                        "question": "Did the agent close the conversation with a proper closing statement (e.g., thanking the customer)? (Yes/No)",
                        "answer": "No",
                        "comment": "The closing was abrupt and lacked a proper farewell."
                    }
                ]
            }
        }
   ]
}

‍

Model

To extract data from this section you have to create API Key with ‘DataExport‘ scope.

This API allows to check results of Sentiment Action of the AI Flow feature.

GET /v1.0/AI/export/Sentiment HTTP/1.1
Host: api.roger365.io
Parameters: startFromDate
Auth-Key: <authorization key>
Content-Type: application/json

‍

Example

{
    "items": [
        {
            "tenantId": ,
            "id": 201,
            "createdAt": "2025-02-05T12:18:44.2156091+00:00",
            "isDeleted": false,
            "lastModified": "2025-03-10T16:14:24.9948296+00:00",
            "result": {
                "id": "0",
                "sentiment": 0,
                "confidenceScores": {
                    "positive": 0.58,
                    "neutral": 0.4,
                    "negative": 0.01
                },
                "sentences": [
                    {
                        "sentiment": 0,
                        "confidenceScores": {
                            "positive": 0.58,
                            "neutral": 0.4,
                            "negative": 0.01
                        },
                        "offset": 0,
                        "length": 7,
                        "text": "Thank you!",
                        "warnings": null,
                        "opinions": []
                    },
                    {
                        "sentiment": 1,
                        "confidenceScores": {
                            "positive": 0.07,
                            "neutral": 0.92,
                            "negative": 0.01
                        },
                        "offset": 7,
                        "length": 3,
                        "text": "you",
                        "warnings": null,
                        "opinions": []
                    }
                ]
            }
        }
   ]
 }

‍

Model

‍

‍

To extract data from this section you have to create API Key with ‘DataExport‘ scope.

This API allows to check results of Summary Action of the AI Flow feature.

GET /v1.0/AI/export/Summary HTTP/1.1
Host: api.roger365.io
Parameters: startFromDate
Auth-Key: <authorization key>
Content-Type: application/json

‍

Example

{
    "items": [
        {
            "tenantId": ,
            "id": 732,
            "createdAt": "2025-05-13T07:52:36.4952292+00:00",
            "isDeleted": false,
            "lastModified": "2025-05-13T07:52:36.5065838+00:00",
            "results": [
                {
                    "aspect": "recap",
                    "text": "The customer reported an issue with their system not updating."
                },
                {
                    "aspect": "issue",
                    "text": "The customer inquired about problems with various statistics used all over the system."
                },
                {
                    "aspect": "resolution",
                    "text": "The agent confirmed that there's a problem which has to be solved as soon as possible"
                },
                {
                    "aspect": "narrative",
                    "text": "Agent and customer were discussing an issue where something was not working, specifically, it was not updating."
                },                
                {
                    "aspect": "Follow-Up Tasks",
                    "text": "The customer will share more details via WhatsApp."
                },
                {
                    "aspect": "Follow-Up Tasks",
                    "text": "Agent will take care of this case via text messages."
                }
            ],
            "editedByUserId": 47,
            "editedResults": [
                {
                    "aspect": "resolution",
                    "text": "The agent identified problem which recently introduced by the solution provider and notified the customer about it."
                },
                {
                    "aspect": "narrative",
                    "text": "Agent and customer were discussing topic of problem with various statistics data, specifically, it was often incorrect."
                }
            ]
        }
   }
}

‍

Model

‍

To extract data from this section you have to create API Key with ‘DataExport‘ scope.

This API allows to check results of Transcript Action of the AI Flow feature.

GET /v1.0/AI/export/Transcript HTTP/1.1
Host: api.roger365.io
Parameters: startFromDate
Auth-Key: <authorization key>
Content-Type: application/json

‍

Example

{
    "items": [
        {
            "tenantId": ,
            "id": 1161,
            "createdAt": "2025-10-15T14:09:03.6995424+00:00",
            "isDeleted": false,
            "lastModified": "2025-10-15T14:09:03.6942848+00:00",
            "results": [
                {
                    "language": "nl",
                    "id": "1",
                    "role": "Agent",
                    "participantId": "1",
                    "text": "Hallo hallo wat voor soort probleem heb je?"
                },
                {
                    "language": "en",
                    "id": "2",
                    "role": "Customer",
                    "participantId": "2",
                    "text": "Hello! Recently I have some strange issues with my printer."
                }
            ]
        }
    ]
}

‍

Model

The MCP integration relies on API Key based authentication, API keys are available from the Business subscription and above.

‍Prerequisites:

• A Copilot Agent already set up in Microsoft Copilot Studio.
• API Key with scope “Conversations” (created earlier in ROGER365.io admin portal)

Disclaimers:

• The ROGER365.io does not supply the Copilot Agent configuration and is not responsible for unexpected results or being flagged as a violator of Terms of Use.
• Please be aware of the cost-aspect for utilizing your Copilot Agent configuration used in this integration – Pricing.

Overview

The Model Context Protocol (MCP) Integration is an open standard for connecting AI agents to external tools and data sources through a consistent, secure interface. Instead of building a separate integration for every action you want an agent to perform, the agent talks to an MCP server, which exposes a ready-made set of tools.

ROGER365.io provides such an MCP server. When you connect it to a Copilot Agent, the server’s available tools, their descriptions, and their input requirements are pulled in automatically – there is no per-action wiring to maintain. If ROGER365.io adds or updates a tool on the server, your agent picks up the change without extra configuration.

Once connected, a Copilot Agent can answer questions and take actions against your Contact Center using the tools the server exposes – for example retrieving call summaries, looking up queue statistics, or checking an agent’s current status. The agent decides which tool (or sequence of tools) best matches each request and calls it through the MCP server.

‍

Authentication of the Copilot Agent

Before connecting the MCP server, configure your Copilot Agent’s authentication.
Navigate to its settings, then open the “Security” section followed by “Authentication”:

Set the authentication option to ‘Authenticate with Microsoft’.

Since the MCP server exposes tools that can retrieve Contact Center data, access must be restricted to identified users rather than left open. Authenticate with Microsoft applies Entra ID sign-in automatically across the Teams, Power Apps, and Microsoft 365 Copilot channels, so no separate authentication setup is required. Each user still completes a one-time sign-in to authorize the MCP server connection itself, covered later in First-time sign-in.

‍

Connecting the MCP server in the Copilot Studio portal

In the Copilot Studio portal, open your Agent configuration and go to the Tools tab. Select “+ Add a tool“, then choose “Add new MCP“:

‍

‍

After entering the required information for the MCP Server, press Create.In the “Connection” drop-down list, choose Create new connection. You will be prompted to provide an API key from the ROGER365.io admin portal to authenticate your user account with the newly created MCP server.

Enter an API key, then select Create, followed by Add and configure.

Once configured, the MCP server becomes available as a tool within Microsoft Copilot Studio. Administrators can use the configured MCP server when creating new agents and can reuse the same connection across multiple agents as needed.

‍

First-time sign-in

The first time a Copilot Agent uses the MCP server, Copilot Studio asks each user to verify their credentials. This is a one-time sign-in that authorizes the connection on that user’s behalf; afterwards the agent reuses it without prompting again. Each new user is asked to sign in once before their first conversation.

To complete it, open the Test pane and send a message to the agent. If a sign-in is required, you’ll see a prompt with an Open connection manager link.

‍

Press on the “Open connection manager” link, locate the created MCP server (by its name) and press “Connect”:

From the drop-down list, press “Create new connection” and authenticate your user account to MCP server using API Key from ROGER365.io admin portal. Afterwards, press “Create” followed by “Submit”.

Once the status becomes Connected, the user account ready for the first conversation with the Copilot Agent (with MCP server as its tool)!

MCP Server Tools

At the Tools tab of the Copilot Agent, you can find list of all existing at the moment tools that the MCP Server provides. Based on the user’s question or request, the Copilot Agent selects the most appropriate tool (or chain of tools) and invokes it through the MCP Server, which executes the action and returns the result.

If ROGER365.io adds or updates a tool on the server, your agent picks up the change without extra configuration.

Depending on the use case, there’s possibility to disable the “Allow all” switch and selectively enable or disable individual Tools that a given Copilot Agent does not need. This allows for maintaining multiple Copilot Agent configurations, each tailored to a specific task or scenario.

‍

Customer data

To extract data from this section you have to create API Key with ‘Conversations‘ scope.

This API returns the found contacts in a conversation, and shows which contact is selected by the user in the Teams adaptive Card

/v1.0/CustomerData/foundContacts/<conv-id>

output

{
    "crmResult": {
        "id": ,
        "conversationId": "",
        "replicatorReferenceId": "",
        "contactId": ,
        "originalContactUri": "",
        "dataSources": ""
    },
    "contacts": [
        {
            "contact": {
                "id": ,
                "firstName": "",
                "lastName": "",
                "displayName": "",
                "email": "",
                "sipAddress": null,
                "street": "",
                "zip": "",
                "city": "",
                "country": "",
                "company": "",
                "phoneNumber0": null,
                "phoneNumber1": null,
                "phoneNumber2": null,
                "phoneNumber3": null,
                "phoneNumber4": null,
                "phoneNumber5": null,
                "custom0": null,
                "custom1": null,
                "custom2": null,
                "mobileData0": null,
                "mobileData1": null,
                "mobileData2": null,
                "type": null,
                "contactUrl": "",
                "contactId": "",
                "companyId": ""
            },
            "replicatorId": ""
        }
    ]
}

‍

CRM Results in bulk

This API allows for returning CRM results in bulk.

POST option to:

/v1.0/Crm/CrmResults

‍

with Auth-Key (Conversations key scope), Content-Type – application/json, and an array of call GUIDs or Conversation GUIDs

[
     "00000000-0000-0000-0000-000000000000"
    ,"10000001-1001-1001-1001-100000000001"
]

‍

output

[
    {
        "id": 12345,
        "conversationId": "00000000-0000-0000-0000-000000000000",
        "replicatorReferenceId": null,
        "contactSourceId": null,
        "originalContactUri": "sip:user@mail.com",
        "dataSources": "00000000-1111-2222-3333-444444444444",
        "created": "2023-12-06T13:42:51.2812911"
    },
    {
        "id": 12346,
        "conversationId": "10000001-1001-1001-1001-100000000001",
        "replicatorReferenceId": "20000002-2002-2002-2002-200000000002",
        "contactSourceId": "30000003-3003-3003-3003-300000000003",
        "originalContactUri": "tel:+31123456789",
        "dataSources": "00000000-1111-2222-3333-444444444444",
        "created": "2023-12-17T09:09:35.5164645"
    }
]

‍

With this information you can get the data from the replicators by doing a POST to

https://admin.roger365.io/v1.0/Crm/00000000-1111-2222-3333-444444444444/Contacts

‍

where 00000000-1111-2222-3333-444444444444 = ID of the replicator (from output of the bulk results)

[
     "tel:+31123456789"
    ,"mail:user@mail.com"
]

‍

Output has the found data from the replicator

[
    {
        "sourceId": "40000004-4004-4004-4004-400000000004",
        "firstName": "Name",
        "lastName": "Surname",
        "displayName": "Surname, Name",
        "email": "name.surname@mail.com",
        "sipAddress": null,
        "street": "Edison Street 10",
        "zip": "00-123",
        "city": "City",
        "country": "Country",
        "company": "ROGER's Family Donuts",
        "phoneNumber0": "1071",
        "phoneNumber1": null,
        "phoneNumber2": null,
        "phoneNumber3": null,
        "phoneNumber4": null,
        "phoneNumber5": null,
        "custom0": null,
        "custom1": null,
        "custom2": null,
        "mobileData0": null,
        "mobileData1": null,
        "mobileData2": null,
        "type": null,
        "contactUrl": "https://website.xyz.net/xxx/yyyyy/zzzzzzz/melon/callercard?action=new&field0=persoonid&value0=40000004400440044004400000000004",
        "contactId": "40000004400440044004400000000004",
        "companyId": "50000005-5005-5005-5005-500000000005"
    },
    {
        "sourceId": "60000006-6006-6006-6006-600000000006",
        "firstName": "Peter",
        "lastName": "Giraffe",
        "displayName": "Giraffe, Peter",
        "email": "pgiraffe@mail.com",
        "sipAddress": null,
        "street": "Tall 10",
        "zip": "NY 10053-2713",
        "city": "New York",
        "country": "United States",
        "company": "Zoo",
        "phoneNumber0": "+11234567789",
        "phoneNumber1": "+19876543210",
        "phoneNumber2": null,
        "phoneNumber3": null,
        "phoneNumber4": null,
        "phoneNumber5": null,
        "custom0": null,
        "custom1": null,
        "custom2": null,
        "mobileData0": null,
        "mobileData1": null,
        "mobileData2": null,
        "type": null,
        "contactUrl": "https://website.xyz.com/xxx/yyyyy/zzzzzzz/leaves/callercard?action=new&field0=persoonid&value0=60000006600660066006600000000006",
        "contactId": "60000006600660066006600000000006",
        "companyId": "70000007-7007-7007-7007-700000000007"
    }
]

Native Microsoft Teams Integrated

The ROGER365.io Contact Center solution is fully integrated within your Teams application. That means that you don’t have to install any additional software except adding the ROGER365.io app from the Teams App store.

Azure AD Sign in

ROGER365.io makes use of you existing Azure Active Directory, that means that you don’t get another usermanagement environment. From our admin portal you can select and assign roles to your existing users.

Interactive Voice Response / Automated Attendant

Interactive Voice Response (or Auto attendant) let people call your organization and navigate a menu system by pressing a digit to speak to the right department, call queue, person, or an operator. You can also use a callers’ input to validate against a database.

Graphical Callflow Editor

Use the graphical callflow to create how callers are routed to the correct queue. You can add actions like audio prompts, provide an IVR option and insert music on hold. By drag-and-drop all actions you’re very flexible in creating your own custom flow.

Queuing / Multiple Routing Methods

The contact center solution provides multiple methods to route incoming calls to your agents:

• Longest Idle: routes each call to the agent who has been idle the longest time. An agent is considered idle if their presence state is Available.
• Serial: rings all call agents one by one in the order specified in the agent list. If an agent dismisses or does not pick up a call, the call will ring the next agent and will try all agents until it is picked up or times out.
• Round Robin: balances the routing of incoming calls so that each call agent gets the same number of calls from the queue.
• Parallel: Rings all agents in the queue at the same time. The first call agent to pick up the call gets the call.
• Skill based routing: routes the call from a queue to an agent, who has the highest skill in a certain subject (eg. language)

Neural Text to Speech

Audio prompts for callers can be uploaded from a file to the platform, but you can also use the text-to-speech option. With this you can enter de text, select the language and your preferred voice and the audio is automatically generated. Because we are using a ‘neural’ text-to-speech engine, the audio result is really amazing. Alternatively you can also upload your own WAV files.

Music on Hold

While your callers are in the queue, you can let them listen to some music. You can easily upload your own provided music files and there are some examples available on the configuration pages.

CRM Integration

With CRM Integration you can utilize your CRM to recognize callers. While they are in the queue and offered to an agent, ROGER365.io displays their name and company name. You can also use CRM information for more advanced functionalities, like routing a caller to a specific queue based on their contract or extract details like open support calls or invoices when answered by an agent. ROGER365.io has default connectors with several well known CRM systems, but we also provide an API for you to create a custom connection. Importing a CSV file is also an option.

Schedule Based Routing

A schedule can be configured to define open and closed hours. Based on that schedule callers can get the correct audio prompts and can be routed to an alternative destination, for example a voicemailbox. Also holiday’s can be configured.

Webhooks

Webhooks can be triggered during several different events like when a new call arrives. These webhooks can trigger your own applications for example a logic/power automate application.

Powerframe for Custom Apps

The powerframe can be used to create your own applications which are opened when a new call is offered to an agent. The powerframe can trigger an url which included variables like the calling phone number, but also variables from the CRM Integration. You can, for example, create an application which shows information from your CRM system, but you could also open a Microsoft Form or a PowerBI report.

* Some features are not available in every subscription. Here you can find an overview

‍

Disclaimer: Before proceeding, please make sure you have good knowledge about your own Microsoft Office 365 and Entra ID situation before continuing.

To enable the Email integration, you must first create an App Registration in the Azure Portal: Download PowerShell Script (preferred way, execute as Global Admin)

General steps (manual method):

1. Create an App Registration in Entra ID environment for Email integration (with API permissions: Mail.ReadWrite – Application, Mail.Send – Application, MailboxSettings.ReadWrite – Application, User.Read – Delegated).
2. Create a Secret for the App Registration.
3. Configure an application access policy and limit the scope of application permissions according to your environment’s situation – Microsoft documentation.

Disregarding point 3 will cause the Email App Registration to have an access to every mailbox within tenant which brings potential security risk.

‍

After successful creation of an App Registration, you can proceed with creating the Email TouchPoint.

Insights & Analytics module settings

In the R365_Settings table, records can be created to change the default settings. The default settings can be overwritten by creating the records in this database:

Here you can check basic examples of how to setup some of the permissions group, with that you can base your future ideas. Before configuring any of these examples make sure that you setup everything correctly according to Setup Access documentation.

Queue Admin/Viewer

Suppose that you want to grant the Admin/Viewer permissions to Queues for selected group of people. Beside of adding Read or ReadWrite rights for Queues Functional Permission you might consider granting Read or ReadWrite rights to Skill Based Routing and CRM Replicator Functional Permissions. It will allow selected group of people to change or view crucial parts related to Queues.

Functional Permissions:

• Queues – Read or ReadWrite
• Skill Based Routing – Read or ReadWrite
• CRM Replicator – Read or ReadWrite

This configuration will restrict an access to only Queues, Skill Based Routing and CRM Replicator sections of ROGER365.io admin portal. Depends on your rights configuration you are able to view the configuration or manage it. (it’s possible to mix rights)

CallFlow Admin/Viewer

Suppose that you want to grant the Admin/Viewer permissions to the CallFlow section. Beside of adding Read or ReadWrite rights, consider adding Read or ReadWrite rights to Audio Prompts, Schedules and CRM Replicator. It will allow selected group of people to change or view crucial parts related to CallFlows.

Functional Permissions:

• CallFlows – Read or ReadWrite
• Audio Prompts – Read or ReadWrite
• Schedules – Read or ReadWrite
• CRM Replicator – Read or ReadWrite

This configuration will restrict an access to only CallFlow, Audio Prompts, Schedules and CRM Replicator sections of ROGER365.io admin portal. Depends on your rights configuration you are able to view the configuration or manage it.

Read access for CallFlow Functional Permissions allows to check/edit configuration of callflow without possibility to Save any changes, users can use Import/Export functionality.

‍

CallFlow Supervisor Admin/Viewer

CallFlow Supervisor Functional Permission works in slightly different way than CallFlow Functional Permission. User with this permission will see the CallFlows only when they’re containing the Queue action – Queue assigned to that action has to have the user assigned as a Supervisor. User is able to see the Subflows of the CallFlows which user have an access to. Consider adding Functional Permissions to Audio Prompts, Schedules and CRM Replicator which will allow the selected group of people to change or view crucial parts related to CallFlows.

Functional Permissions:

• CallFlows Supervisor – Read or ReadWrite
• Audio Prompts – Read or ReadWrite
• Schedules – Read or ReadWrite
• CRM Replicator – Read or ReadWrite

This configuration will restrict an access to only CallFlow, Audio Prompts, Schedules and CRM Replicator sections of ROGER365.io admin portal. Depends on your rights configuration you are able to view the configuration or manage it.

Note: Read access for CallFlow Supervisor Functional Permissions allows to edit configuration of callflow which users have an access to, and use Import/Export functionality. There is no possibility to Save any changes (case for Read right). Administrators can combine the CallFlow Read, and CallFlow Supervisor ReadWrite Functional Permissions together. It will allow selected users for seeing every callflow, but editing only those which they’re part of Queue’s Supervisor.

You can create new CallFlows and save them only with CallFlow ReadWrite Functional Permission.

‍

ROGER365.io only uses Microsoft industry standard technology. The ROGER365.io is an Enterprise Application within the Microsoft Entra ID. Via this Enterprise Application you can allow users to sign-in to ROGER365.io admin portal or not, and into which sections users can have an access to. It is strongly recommended to configure the ROGER365.io Enterprise Application within Microsoft Entra ID according to your own company policy.

Role reference can be found at this documentation page.

Disclaimer: The information in this document must be used as a guidance, please make sure you have good knowledge about your own Microsoft Office 365 and Microsoft Entra ID situation before continuing.

Azure / Microsoft Entra ID:

1. Create a security group at Microsoft Entra ID tenant. Save the Object Id of a freshly created security group, it will be needed later.
2. Assign selected users.
3. Navigate to Enterprise Applications, and choose “ROGER365.io” application.
4. Add created group to “ROGER365.io” application via “Users and groups” section from a Resource Menu. Group have to have “TeamsUser” role assigned.

Note: Assigning “Admin” role to a security group will grant Admin permissions to ROGER365.io admin portal for every member of a group!

ROGER365.IO admin portal:

1. Navigate to “Permissions” tab.
2. Press “+ Add Functional Permissions” button.
3. Paste Object Id of a security group into “Group ID” field, from Azure’s step 1.
4. Give a descriptive name to your group of Functional Permissions.
5. Set proper rights for chosen Functional permissions (sections of a portal).

Examples of usage is described at this documentation page.

Important Notes:

• General, Resource Accounts and Call Park pages are restricted only to Admin role.
• Any permission changes are effective usually after ~10 seconds (requires re-login to admin portal).
• You cannot combine user “Admin” permission with RBAC group permissions (restrictions) where a user is part of “Admin” permission will always have priority over RBAC group restrictions. Not respecting this will result with User having full access to admin portal.

‍

Add TOASTAR app in Teams client

Follow these steps to enable the TOASTAR App in your Teams client:

1. Go to Microsoft Teams and select the [Apps] icon in the left bottom of the navigation bar
2. Search for [Toastar]
3. Click the found application and select [Add]

Configure simultaneous ring

Final step is to configure simultaneous ring, so when a call arrives the call is also redirected to the TOASTAR app. Follow these steps:

1. In de Microsoft Teams client, click your profile picture on the right top of your screen
2. Select [Settings]
3. Select [Calls]
4. Select [New number or contact] in the list next to [Also ring]
5. Enter the phone number which you configured in the step Create Teams Virtual User
6. Close the window

Now when you receive an external phonecall, TOASTAR will search the selected CRM Replicator (Datasource) and display the found contact details.

‍

Via the ROGER365 users can now be enabled to use the function. You can create multiple profiles with different settings. For example you can create a profile for users at a sales department and a different profile for the financial department. In this way you can connect seperate CRM sources to different users, or send them different adaptive cards.

Add a profile

To create a profile, you can follow the next steps:

1. Go to the ROGER365 admin portal
2. Select Toastar for Teams in the left navigation bar
3. Click [+ Add Profile]
4. Fill in a [Name] and [Description]
5. At Data Sources you can select in which sources can be searched
6. When a new call arrives, a webhook can be triggered for example to Microsoft Power Automate or a Logic App
7. To search your Active Directory you can select Users Groups, you have to login with administrator credentials te retrieve these
8. At [Available Users] select the users to add to the profile and click the arrow to move them to the [Assigned Users]
9. Click [Save]

‍

Webhook

When an incoming phonecall arrives, you can trigger a webhook to an external system (like Power Automate/Logic App) to perform certain actions. This can for example be a retrieval of extra CRM data, or create a phone note on the CRM contact card. To configure a webhook, follow the next steps:

1. Go to the ROGER365 admin portal
2. Click [Toastar for Teams] in the left naviagtion bar
3. Select the Profile
4. Enter the URL to your HTTP listener in the [Webhook] field
5. Click [Save]

The JSON displayed underneath will be send. If multiple contacts in a dataSource (CRM Replicator) are found, a button appears to select the correct contact. After selecting another contact, the webhook will be send again with the new data in the “selectedContact” field.

{
  "conversationId": "8ecc770b-1685-4770-cffa-08d84434a9b2",
  "originalContactUri": "+31882757600",
  "lookupType": "phone",
  "dataSources": "ef4f1179-e890-44cb-9748-24f590c66b88",
  "selectedContact": {
    "id": 25001,
    "firstName": "John",
    "lastName": "Doe",
    "displayName": "Doe, John",
    "email": "john@doe.com",
    "sipAddress": null,
    "street": "Nieuwe Plantage 58",
    "zip": "2611 XK",
    "city": "Delft",
    "country": "The Netherlands",
    "company": "ROGER365.io",
    "phoneNumber0": null,
    "phoneNumber1": "+31882757600",
    "phoneNumber2": null,
    "phoneNumber3": null,
    "phoneNumber4": null,
    "phoneNumber5": null,
    "custom0": "Customer",
    "custom1": "Doe, John",
    "custom2": null,
    "mobileData0": null,
    "mobileData1": null,
    "mobileData2": null,
    "type": null,
    "contactUrl": "null",
    "contactId": "null",
    "companyId": "null"
  },
  "selectedContactDataSource": "ef4f1179-e890-44cb-9748-24f590c66b88",
  "userAadId": "c1f1a379-3075-461e-8456-193db13ade17",
  "type": "newIncomingCall"
}

Postback URL

After receiving the JSON via the Webhook you could, for example, retrieve extra CRM info. This new info can be posted back to the already displayed info in Microsoft Teams. You have to send this info in an adaptive card format. More information how to create an adaptive card, click here.

To make sure the info is posted back to the right user, you must include the “conversationId” in your postback url (https://api.roger365.io/v1.0/tft/postcard/<convid>). In this example the postback URL will be:

https://api.roger365.io/v1.0/tft/postcard/8ecc770b-1685-4770-cffa-08d84434a9b2

‍

Add TOASTAR App to Microsoft Teams

There are two options to add the TOASTAR app to the Microsoft Teams client: 1) Distribute it globally in your tenant to all users via the Microsoft 365 Admin portal or 2) install it locally in the Microsoft Teams client. In case of the 2nd option this must be allowed by the Microsoft 365 Administrator.

Option 1: Install locally

1. Login into https://admin.roger365.io with you Microsoft 365 account
2. Go to [Toastar for Teams]
3. Download the TOASTAR.ZIP manifest file
4. In the Microsoft Teams client, click on the [Apps] icon. You can find it at the left bottom of the navigation bar
5. Select the option [Upload a custom app] (If you don’t see this option, then this must be allowed by the Microsoft 365 Administrator) and select the TOASTAR.zip file

Option 2: Install globally

1. Login into https://admin.roger365.io with you Microsoft 365 account
2. Go to [Toastar for Teams]
3. Download the TOASTAR.ZIP manifest file
4. As a global Microsoft Teams go to the Microsoft Teams admin center:
5. Choose [Teams Apps] > [Manage apps]
6. Choose upload new app and choose select file.
7. In the pop up browse and select the TOASTAR.ZIP file you and acknowledge the screen to get the upload started.

After a successful upload it could take op to 24 hours before the app is visible in the Teams Apps. You can also logout/login into Teams to have the TOASTAR app directly available.

‍

This next step is to create a Microsoft Teams user, configure it with a phonenumber and connect it to the BOT channel created in the previous step.

Create user via Powershell

• Install the Microsoft Teams powershell module (https://docs.microsoft.com/en-us/microsoftteams/teams-powershell-install).
• Use the command underneath and enter the username for your domain, and past the App ID copied from the previous step:

New-CsOnlineApplicationInstance -UserPrincipalName "username@yourdomain.com" -DisplayName "ToastarForTeams" -ApplicationId "App ID"

• To check if this command is executed correctly, run the following command:

get-CsOnlineApplicationInstance -Identity "username@yourdomain.com"

You’ll see an output like this:

RunspaceId        : 55f3d4ff-5606-430d-81f9-************
ObjectId          : bde233d4-79cf-4d0a-a2bd-************
TenantId          : 5b98daa7-3f0e-4fb1-9ad0-************
UserPrincipalName : username@yourdomain.com
ApplicationId     : da0c3ed1-eeec-41cd-acb8-************
DisplayName       : ToastarForTeams
PhoneNumber       : tel:+5555

• To sync between Azure and the Agent Provisioning Service run the command (you can find the ObjectId from the output of previous step):

Sync-CsOnlineApplicationInstance -ObjectId "<ObjectId>"

• Add a Microsoft Teams license to the user via the Office 365 Admin Portal – >User management tab. You can use the free ‘Virtual-User’ license. It could take up to 24 hours before this has been processed.

Assign phonenumber to the user

Assign a phonenumber to the user. This can be an internal number, it does not have to be an external phonenumber

For DirectRouting environment

Set-CsPhoneNumberAssignment -Identity user@domain.com -PhoneNumber +5555 -PhoneNumberType DirectRouting

for CallingPlan environment:

Set-CsPhoneNumberAssignment -Identity user@domain.com -PhoneNumber +5555 -PhoneNumberType CallingPlan

‍

Configure BOT Channel

Log into your Microsoft Azure environment by browsing to https://portal.azure.com. Make sure you have an Azure subscription active.

• After logging in, choose [+ Create a resource]

• Type [Azure Bot] in the search field and press [ENTER]
• In the Azure Bot channel detail page, click [Create]

• Enter a name for the Bot handle. You can choose your own name, but it has to be unique within Azure
• Choose your Azure Subscription
• Create a new Resource Group and select it
• Choose a location near you
• Select your pricing choice
• Select Type of App ‘Single Tenant’
• Select ‘Create a new Microsoft App ID’
• Click [Review & Create]
• In the next page click ‘Create’

After creation, go to the new created Azure Bot Channel by clicking ‘Go to Resource’.

• Click [Channels] in the left menubar
• Click the Microsoft Teams button

• Select your Microsoft Teams Edition

• Select the tab [Calling]
• Enable the checkbox [Enable Calling]
• Fill in at [Webhook] https://api.roger365.io/v1.0/tft/calls?id=<yourcompanyname>
• Click [SAVE]

• Select [Configuration] in the left navigation bar
• Copy the string at Microsoft App ID. (You’ll need it in the next step ‘Create Teams Virtual User‘)

• Next to [Microsoft App ID] click [Manage]
• In the left navigation bar click[API Permissions]
• Then click [+ Add a permission] and click the section [Microsoft Graph] on the right side

• Choose [Application Permission]

• Scroll down to [Calls] and check all options
• Click [Add Permissions] to save

Last step is to grant the permissions by clicking the ‘Grant admin consent for….’ button:

‍

We have created a Powershell example which uploads a CSV file to a custom replicator. In the example you have to fill in your connection URL and Authorization key in the Powershell script. In the attached CSV file you can add your own CRM data. Make sure you keep the correct columns in the example.

Click here to download the example ZIP

Step 1: Get connection URL and Authorization key

First you have to create a Custom CRM Replicator. How to do this, you can find here. In the configuration of the custom CRM replicator you can find the [connection url] and the [authorization key]. Copy these two values, you’ll need this in the next step.

Step 2: Change Powershell script

After you’ve downloaded the ZIP file above, extract in a directory (for example c:\temp\)  you’ll get two files:

• SampleUploadCSV.ps1, this is the PowerShell script
• SampleUploadCSV.csv, this contains a few example contacts

Open the file ‘SampleUploadCSV.ps1’ in Notepad or another script editor. Paste the Connection Url at:

$Url = “<paste your url here>”

and paste the Authorization Key at:

$AuthorizationKey = “<paste your key here>”

Save the file

Step 3: Fill the CSV file

In the file “SampleUploadCSV.csv” we have added a few dummy contacts. You can replace this list with an export of your own contacts, make sure to keep the columns and columns names. Keep in mind to define labels in the ROGER365.io admin portal for the CRM action buttons, in case ActionLink0-3 properties has been configured.

Step 4: Run the PowerShell script

Follow these steps to run the script and upload the contacts:

• Open a command box
• Go to the directory where you’ve extraced the files
• Type: “Powershell” and press [Enter]
• Type: “.\SampleUploadCSV.ps1” and press [Enter]

.\SampleUploadCSV.ps1

If the command was executed successfull you’ll see this:

-> If you get this errormessage after executing the powershell script:

.\SampleUploadCSV.ps1 : File C:\temp\SampleUploadCSV\SampleUploadCSV.ps1 cannot be loaded. The file
C:\temp\SampleUploadCSV\SampleUploadCSV.ps1 is not digitally signed. You cannot run this script on the current system.
For more information about running scripts and setting execution policy, see about_Execution_Policies at
https:/go.microsoft.com/fwlink/?LinkID=135170.
At line:1 char:1
+ .\SampleUploadCSV.ps1
+ ~~~~~~~~~~~~~~~~~~~~~
+ CategoryInfo : SecurityError: (:) [], PSSecurityException
+ FullyQualifiedErrorId : UnauthorizedAccess

then you can first execute this command:

Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass

and answer [Y] to allow the script to run.

If then the script has ran succesfully you’ll see that the contacts are uploaded to the platform:

‍

Phonebook

With the phonebook feature you can search your CRM Replicators for contacts and call, email or chat them directly from Teams. To use the feature you must have a business or enterprise subscription.

The following components are needed:

• Add the ROGER365 app to your Teams environment
• Create a CRM Replicator with your CRM system
• Configure the phonebook settings at the CRM Replicator (see underneath)
• Social Messaging – Create and configure Toastar for Teams Profile (check documentation page)

Configure phonebook settings

When you have succesfully created a CRM Replicator, click on the ‘edit’ icon next to the CRM Replicator. On the bottom you find the Phonebook Configuration. Here you can change the default labels of the replicated phonenumbers, these will show up in the search results and at the CRM contacts. If you enable the checkbox ‘searchable’ then this number is also used to lookup a contact for incoming phonecalls (Toastar / CRM Calling feature).

Contact Center

Now, searching for the CRM contact which is part of your replicator will make the labels for phoneNumber fields to be shown:

When searching for CRM contacts, the search will by default return results that begin with the text you enter. If you add an asterisk (*) as a wildcard, you can broaden the search to match your text anywhere within a result.

Place the asterisk at the beginning to include anything that comes before your search term, or at the end to include anything that comes after it.

Social Messaging

When you now click the ROGER icon in Teams, you can select ‘Search contact’.

Then you can search for a contact, company name or phonenumber. You can also use a wildcard like *smith or *12345

‍

Prerequisites

Before continuing make sure you have ROGER365.io connected to a Microsoft Teams channel and you have enabled Webchat TouchPoint.

Result

At the end of this document you have created your first ROGER365.io flowchart that handles webchat messages from a website to Microsoft Teams and back.

Create Webchat to Teams flowchart

1. Login into ROGER365.io admin portal, https://admin.roger365.io with your Microsoft 365 credentials

2. Choose in the left menu, [Social Messaging] > [- Messages] and choose option [+ Add Message] and click [Save]

For the name use the text below

Send Welcome Message

Enter this at the EN language

Hi! I’m Roger, how can I help you?

‍‍3. Choose in the left menu, [Social Messaging] > [Messageflows] and select option [+Add Flowchart]

4. Fill in flowchart properties in the right panel;

For the name use the text below

Webchat

Enter this for the description

My first flow

‍

‍5. Choose the option [Variables] in the tool bar and create the variables below and press [Save]:

‍

6. Click the “Root” node and fill in the following “Node Properties”:

Webchat

‍

‍

1. Choose [SendMessageTo] action in the left panel, drag & drop it into the flow area and fill in the “Node Properties”;

For the name use the text below

Send Welcome Message

Enter this for the Message

PredefinedMessage["Welcome"]

Enter this for the touchpoint

Webchat

‍2. Connect both “Root” and “Send Welcome Message” actions with a link

3. Choose action “WaitInput” in the left panel and drag & drop into the flow area.

4. Set the following “Node Propeties” and link this node with the “Welcome” node;

For the name use the text below

Send user message to Teams

Enter this as the User input destination variable

UserMessage

‍5. Choose action “SendMessageTo” and drag & drop, fill in the following “Node Properties” and connect “WaitInputUser” with “SendToAgent”;

For the name use the text below

Send user message to Teams

Enter this as the User input destination variable

UserMessage

Enter this for the touchpoint

Teams

‍6. Choose the link between “Wait input User” and “Send user message to Teams” and make sure the “Link Properties” is set to “Input”.

7. Create a link from “Send user message to Teams” back to “Wait input User”

Create Teams to Webchat flowchart

1. Add action “Start” action into the flow area and give the following “Node Properties”;

• Touchpoints: Teams
• Add action “WaitInput”, fill in the “Node Properties” and link “Start” Teams with this action;

For the name use the text below

Wait input Agent

Enter this as the User input destination variable

AgentMessage

2. Add action “SendMessageTo” and fill in the following “Node Properties”;‍

For the name use the text below

Send agent message to Webchat

Enter this as the User input destination variable

AgentMessage

Enter this for the touchpoint

Webchat

‍3. Link “Wait input Agent” with the “Send agent message to Webchat” and make sure the “Link Properties” type to “Input”

4. Create a link from “Send agent message to Webchat” back to “Wait input Agent”

5. Add action “SendMessageTo” and fill in the following “Node Properties”;

For the name use the text below

Send error message

Enter this as the User input destination variable

LastErrorMessage

Enter this for the touchpoint

Teams

‍6. Create a Error link between the “Root Webchat” node and “Send error message”

7. Create a Error link between the “Start Teams” node and “Send error message”

8. Add action “End” and create a link between “Send error message” and “End”

9. Click [Save] to store the created flow

Assign new flow to Touchpoint

1. Go to [TouchPoints], and choose the webchat touchpoint you want to connect the created flow.
2. Choose option “Assign a flowchart…” and choose the just created flow.
3. Click [Update] to confirm changes.

‍

‍

‍

‍

‍

‍

‍

‍

Prerequisites

Before continuing make sure you have created a Whatsapp for Business Touchpoint  or configured a Test-Sandbox TouchPoint Also you should have configured a Microsoft Teams channel and installed the ROGER bot.

Configure flow between WhatsApp and Microsoft Teams

Log into the ROGER365.io admin portal. In the left navigation bar select [Social Messaging], [Messageflows]. On the top click [Add Flowchart from Library]:

1. Enter a name for your Flow
2. Enter a description
3. Start TouchPoint: Select your earlier configured WhatsApp TouchPoint
4. Details:
   1. Select a flow template, for example a simple ‘Whatsapp to Teams’
   2. Select Team TouchPoint: Select the earlier created Teams TouchPoint where the messages should be delivered.
5. Click Create

After clicking create, a flow is designer for you. You can simply press Save.

Select flow on WhatsApp TouchPoint

Last step is to select the flow you just created on the WhatsApp Touchpoint. In the left navigation bar, browse to [TouchPoints], and select your earlier created WhatsApp TouchPoint.

On the top at ‘Assign a flowchart’, select the flow you’ve just created. Click UPDATE

Now you’re ready

‍

When using a Facebook or Webchat TouchPoint, you don’t always get the user’s information to perform CRM recognition. In case of webchat you could add a pre-form webchat to ask the user for an email address, but another solution is to automatically ask a user for their email address when they start a conversation.

You can use a regular expression to filter or validate the email address and then use the SetParticipant action in our flow designer to identify the customer. Underneath you can find a screenshot how this could look within a flow and the Regular Expression (REGEX) which you can use for validation:

Regex which you can use in the IF statement:

System.Text.RegularExpressions.Regex.IsMatch(emailRaw.Text, "(?:[a-z0-9!#$%&'*+/=?^_`{|}~-]+(?:\.[a-z0-9!#$%&'*+/=?^_`{|}~-]+)*|"(?:[\x01-\x08\x0b\x0c\x0e-\x1f\x21\x23-\x5b\x5d-\x7f]|\\[\x01-\x09\x0b\x0c\x0e-\x7f])*")@(?:(?:[a-z0-9](?:[a-z0-9-]*[a-z0-9])?\.)+[a-z0-9](?:[a-z0-9-]*[a-z0-9])?|\[(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?|[a-z0-9-]*[a-z0-9]:(?:[\x01-\x08\x0b\x0c\x0e-\x1f\x21-\x5a\x53-\x7f]|\\[\x01-\x09\x0b\x0c\x0e-\x7f])+)\])")

You can present webchat users a pre-webchat form. In this form it’s possible to ask the user to enter his/her email address. This information then can be used to perform a lookup in the CRM Replicator and show the webchat user data when the conversation is routed to a teams channel.

Start with creating a flow and add ‘GetFormInput’ from the menu on the left.

Click the ‘Edit Mappings’ button on the right.

Here you can create & assign variables to the inputboxes (Field1, Field2 etc.) from the pre-webchat form. Now that we have the webchat-user data in variables, we can set these variables to set the identity. For this, add the ‘SetParticipant’ component.

In this example we have inserted the email address into the variable ‘Form2’. By adding this variable at the property ‘Contact Uri’ and setting the ‘Type of Uri’ to mail we have assigned the email address to the user. We can also assign the name, which is in variable ‘Form1′, by entering that variable at Form1.

After assigning the variables to the webchat user, ROGER365.io will try to match the info with the data from a datasource in the CRM Replicator and will show the customers’ full information in the Teams channel.

‍

This feature is only available in the Business and Enterprise subscriptions.

‍

Description

‍

Node Properties

‍

‍

Mapping

‍

Note: The value of CallId property can be used to terminate callback record from the Queue using the API (section – Delete).

‍

Before using this action in your Messageflows, the CRM Datasource(s) has to be assigned on your Social Messaging TouchPoint. (e.g., Webchat, WhatsApp, etc.).

Description

‍

Node Properties

‍

‍

‍

Description

‍

Node Properties

‍

‍

This feature is only available in the Enterprise subscription.

‍

Before configuring this action you have to create Translator resource in the Azure portal, you can find description how to do it in this Microsoft article – Create a Translator resource.

Be aware that images cannot be send through this action, translations can be not precise.

Description

‍

Node Properties

‍

‍

Mapping

‍

Edit JSON Content overview:

This feature is only available in the Enterprise subscription.

‍

Before configuring this action you have to create Azure OpenAI resource at Microsoft Azure, you can find description in this Microsoft article – Create and deploy an Azure OpenAI Service resource.
Most of the configuration part must be done in Microsoft Foundry portal.

Microsoft Azure OpenAI is a preview feature, please be aware of unexpected results, and test thoroughly before implementing to production environment.

Description

‍

Node properties

‍

* Variable in JSON Content can be used when Evaluate Jsoncontent option is enabled.

Mapping

‍

Setting up variable for result property is mandatory for action to work!

Description

‍

Node Properties

‍

Link Properties

‍

Mapping

‍

‍

Description

‍

Node Properties

‍

Link Properties

‍

Example

This POST returns the feedback from a user (using the Choice action) to the chatbot – variable (string)

"https://r365-core-webapiconfig.azurewebsites.net/api/FaqResults/Feedback?faqRequestId="+FaqRequestId+"&answerIndex=1&feedback=true"

‍

This returns the filled in values from a pre-webchat form. The values (Field1, Field2, Field3, etc) can be mapped to a variable – variable (string)

"https://r365-core-webapiconfig.azurewebsites.net/api/SiteVariables/ByConversation/" + Conversation.Id.ToString()

‍

‍

Description

‍

Node Properties

‍

Link Properties

‍

Description

‍

Node Properties

‍

Link Properties

‍

* The expression evaluator is based on C# and supports a subset of C# functions, which we regulary extend. If you need help, or a function is not supported, please contact your partner.

Description

‍

Node Properties

‍

Link Properties

‍

Mapping

‍

Description

‍

Node Properties

‍

Link Properties

‍

Mapping

‍

‍

Description

‍

Node Properties

‍

Link Properties

‍

‍

Description

‍

Node Properties

‍

Link Properties

‍

‍

‍

Description

‍

Node Properties

‍

Link Properties

‍

Mapping

Description

‍

Node Properties

‍

Link Properties

‍

Mapping

‍

Description

‍

Node Properties

‍

Link Properties

‍

‍

‍

Description

‍

Node Properties

‍

Link Properties

‍

Mapping

‍

‍

‍

‍

‍

Description

Description

This action is only possible with Webchat TouchPoints

‍

‍

Node Properties

‍

Link Properties

‍

‍

‍

Description

‍

Node Properties

‍

Link Properties

‍

‍

Description

‍

Node Properties

‍

Link Properties

‍

* The timeout link is special, because the flow itself will not continue from there. Use the timeout to trigger an interrupt on an action that supports interrupts.

** If a timeout is set on the action and expires and an interrupt is present, this will be followed. IE, you don’t have to draw a timeout arrow to the own action. If there is no interrupt link from the action, the timeout will only execute the Timeout link, but the flow remains on the WaitInput action.

‍

Next to various types of actions, there are a couple of basic elements you must or can use in every flow. Those are described underneath using this example.

1. Root Node

Every flow starts with a ‘Root’ node and it’s by default shown when you create a new flow. This node represents the starting point of the conversation flow, for example a webchat TouchPoint. On the root properties panel you can select the starting TouchPoint. From the node you can create three types of links:

1. Any: This link is followed by default
2. Closed: This link will be followed when a close event (for example user closed the webchat or a Teams agent closes the conversation) occurs.
3. Error: This link will be followed when conversation broke due to issue in configuration.

2. End Node

When the conversation of flow is completed, the end node is reached. This is also a mandatory element to end the active flow. Sometimes the End node must be connected to multiple other actions, in such case you can use Merge action.

The End node allows you to configure AI-powered summarization based on the transcript of a social conversation. Available summary aspects include Recap, Narrative, Issue, Resolution, and Follow-up Tasks (availability depends on the API version; not all aspects may be supported). An administrator must complete the configuration using a previously created AI Service endpoint that supports AI summarization. A language code is required; if none is provided, Dutch will be used by default. Once the configuration is complete, set the “Enable Conversation Summarization” property should be changed from “false” to “true”. The generated AI summaries are available in the corresponding social interaction record in Agent Assist.

3. Start Node

A conversation is always initiated in the Root Node, but may touch (multiple) other TouchPoints. For example: the Root starts at a Webchat TouchPoint but can be routed to a Microsoft Teams Channel. This Channel/TouchPoint then will be a Start Node. You could have multiple Start Nodes, for example when you use a choice action to let the user select if he/she want to chat with a salesteam channel or a supportteam channel.

4. Error Node

When an action experiences an error, it will first look for a local error link. All actions can have an individual error link. If such a link is not found, it will look for a global error link (from the root/start node). Always define a global error as shown in the example above.

If you connect a SMS number, you have to configure a Webhook in the portal of your SMS Provider. Underneath you’ll find the URL which you have to use:

‍

The Query string is mandatory in the Webhook’s URL. Replace “SMS Identifier” with the one created at SMS TouchPoint.

‍

The Webhook URL should look like this:

https://api.roger365.io/api/sms/SmsListener?cid=003112345678901

‍

If you connect a WhatsApp for Business number, you have to configure a webhook in the portal of your WhatsApp Business Provider. Underneath you’ll find the url which you have to use:

With the WhatsApp OUT functionality you can send ad-hoc WhatsApp messages to a user instead of only replying on an incoming WhatsApp messsage from a user. You can only send pre-defined template messages (WhatsApp charges you for sending a template). Directly after sending a message, a new conversation is created in the MS Teams channel. When the WhatsApp user responds to the template message, you can continue the conversation.

How you can use this feature is decribed here.

Getting a OPT IN

In order to initiate a WhatsApp message to a person, you must first receive opt-in permission confirming that they wish to receive future messages from you on WhatsApp. The opt-in must (a) clearly state that the person is opting in to receive messages from you over WhatsApp and (b) clearly state your business’ name. You may only initiate chats if you are sending a transactional notification, and only via approved Message Templates (as defined in Facebook’s documentation), subject to applicable pricing. Any Message Templates must comply with their terms and policies, and only be used for their designated purpose.

The ROGER365.io platform offers a solution to track the opt-in of a contact. When you try to send a WhatsApp message to a user, you’ll be notified if your have an opt-in of that user. You, as a business, are responsible to ask the user for permission. You can choose yourself how you ask this permission, as long as it complies with the regulations from WhatsApp (Facebook).

The ROGER365.io provides an API which you can use to add and delete opt-in confirmations from customers. Here you can find the description how to use it

Configure Message Flow

When you send a whatsapp out message, the system will send the contents of the templated message back, as if it was retrieved from the enduser. That way flows will be able to detect a message was send and a conversation in the Teams channel can be created. When using features like ‘autoreply’, additional steps must be taken to avoid the generated message will trigger additional messages to the whatsapp user.

Therefore, after the Root node of whatsapp, add an IF-statement, with SystemInitiated. If true, the message should be handled as the generated message; skip all auto reply features and move to the steps where the message is send to the Teams channel:

Configure Templates

You can only send ‘templated’ messages if you initiate a conversation with a WhatsApp user. You can find the description how to create/add templates here.

Enable WhatsApp OUT

Before you can use WhatsApp OUT, you have to configure a few settings. The country code is used to do the correct number formatting if you manual enter a number, and you can select a data source which is used to search your CRM for contacts.

1. Open web browser and go to: https://admin.roger365.io
2. Login with Microsoft 365 credentials
3. Go to [Social Messaging], [TouchPoints]
4. Click the edit button next to the WhatsApp TouchPoint
5. Go to the section named ‘WhatsApp Out’
6. Select your (default) country code
7. Select the Data Sources for contact lookup
8. Click [Update] to save

‍

Templates are predefined messages which you can sent to a WhatsApp user to send a notification (like ‘your item has been shipped’) or to trigger a user to send a message back. An example of a template message looks like this:

Hi, we would like to update you about {{1}}. Can you send me a message back, so we can continue our conversation?

In this example you can send the text above, with adding the subject at the {{1}} placeholder. Multiple templates in multiple languages can be added. You can also insert Rich media headers like icons or a document, or add buttons in the message as shown in the screenshot on the right.

‍

The customer care window ends 24 hours after the last message is sent by the end-user. After 24 hours, you will only be able to send a notification via Message Templates. Each time the customer sends you a message, whether this is the conversation initiating message or another reply in the conversation, the 24hour customer care window will restart. So, from the last message your customer has sent you, you have 24 hours to send custom messages, free of charge unless you exceed your package.

‍

Creating a template

Login at your WhatsApp for Business provider (for example cm.com) and create a template.

After creating the template it will be submitted to WhatsApp for approval. Once approved, you’ll get a ‘Facebook Namespace’. This will be the unique identifier which you have to use in the next step. You can create a single template in multiple languages, the namespace will be the same.

To add the template to ROGER365.io follow these steps:

1. Open web browser and go to: https://admin.roger365.io
2. Go to [Social Messaging], [TouchPoints]
3. Click the edit icon right to the Whatsapp TouchPoint
4. Go to the section named [Manage templates]
5. Add the languages you would like to support
6. Click ‘+’ in the titlebar of [Manage templates]
7. Fill in a [Display name] for the template
8. Fill in the [Name] exactly like in the portal of your WhatsApp Provider (like cm.com)
9. Copy/Paste the [Namespace] which you obtained at your WhatsApp provider
10. If you configured a header then select the type of header from the list and the required fields.
11. Create the variable fields which you’ve created in the body, with the correct Type and Index
12. If you’ve added buttons with a variable, then create the variables
13. Finally insert the body text including any references to variables, for each selected language.
14. Click [Add] to finish

This is an example of a configured template:

‍

If you connect a WhatsApp for Business number, you have to configure a webhook in the portal of your WhatsApp Business Provider. Underneath you’ll find the url which you have to use:

How to customize the default layout to your own design

Download customWebchatOptions and change the extention into .js.

This is a copy of defaultWebchatOptions.js file and allows users to overwrite the styles by adding the ones they want.

The file must be hosted on your server and included before the defaultWebchatOptions.js file, like a JS script:

‍

<script src="{customerBaseURL}/customWebchatOptions.js"></script>

Also, you can modify the file depending on how many facilities you want to give them. E.g. you can remove completely BotStyleOptions block meaning you will be able to change title and subtitle (text next to logo) and the look & feel for start and open buttons.

‍

• BotStyleOptions contains style set options provided by Microsoft BotFramework-WebChat and must be updated following the instructions from official documentation https://docs.microsoft.com/bs-latn-ba/azure/bot-service/bot-builder-webchat-customization?view=azure-bot-service-4.0 , https://github.com/microsoft/BotFramework-WebChat/blob/0465cbff4e7b834accc079b056ed6d6ffd67bb1e/packages/component/src/Styles/defaultStyleOptions.js#L3
• LayoutOptions contains options for title, subtitle and look & feel of start and open buttons

‍

Tips how to complete the file for few items:

• change title: uncomment title property and add the text wanted
• change start button: there are 2 ways to do that
  • add an image: uncomment startWebchatImage property and add the absolute path of image
  • add HTML code: uncomment startWebchatHtml property and add your HTML code

(if both are filled in the image has priority)

• change botAvatarImage: uncomment botAvatarImage property and add the absolute path of image
• add initials instead of avatar: set the following options

botAvatarBackgroundColor: '#colorCode' //e.g.#FF0000
botAvatarImage: '', 
botAvatarInitials: 'AB', //only 2 letters

• change look&feel for bubbleText: set the following options

bubbleFromUserBackground: '#BackgroundColorCode',
bubbleFromUserBorderRadius: 20,
bubbleFromUserTextColor: '#TextColorCode',

The image below shows which elements can be customized via customWebchatOptions.js file. All other components can be customized via CSS, for which you must write your own CSS files.

Customize Webchat Form

If you have added a pre Webchat form, you can also change the default layout. Underneath you’ll find the used CSS:

#wechatForm = preform
#webchatForm .message = headermessage
#webchatForm labels = input labels
#webchatForm Input = input veld
#webchatForm .button = send button

/* WebchatForm*/
#webchatForm {
    margin: 20px;
    border-radius: 15px;
    border: solid 1px grey;
    display: flex;
    flex-direction: column;
    background-color: rgb(240, 240, 240);
}

    #webchatForm .message {
        padding: 10px;
        margin-bottom: 5px;
        background: rgb(45,54,145);
        background: -moz-linear-gradient(135deg, rgba(45,54,145,1) 0%, rgba(77,214,176,1) 100%);
        background: -webkit-linear-gradient(135deg, rgba(45,54,145,1) 0%, rgba(77,214,176,1) 100%);
        background: linear-gradient(135deg, rgba(45,54,145,1) 0%, rgba(77,214,176,1) 100%);
        color: white;
        border-top-left-radius: 15px;
        border-top-right-radius: 15px;
        font-size: 14px;
    }

    #webchatForm .button {
        align-self: center;
        margin-top:auto;
        margin-bottom: 10px;
        font-size: 14px;
        padding: 6px 12px;
        color: white;
        display: inline-block;
        text-decoration: none;
        text-align: center;
        white-space: nowrap;
        vertical-align: middle;
        -ms-touch-action: manipulation;
        touch-action: manipulation;
        cursor: pointer;
        -webkit-user-select: none;
        -moz-user-select: none;
        -ms-user-select: none;
        user-select: none;
        background-image: none;
        border: 1px solid transparent;
        border-radius: 5px;
        background-color: #4dd6b0;
    }

    #webchatForm .formInput{
        margin: 0px 10px;
    }

‍

‍

Status message

On the top of the webchat window, you can show users a status message. This can be a persistent message, which you can change on-the-fly, but also a message based on a schedule. Then you can show a message based on the schedule ‘open’ or ‘closed’.

You can configure messages in multiple languages. The language is automatically detected based on the language setting of the users’ browser. If the language cannot be detected, the default message will be displayed.

To configure the status messages:

1. Open web browser and go to: https://admin.roger365.io
2. Login with Microsoft 365 credentials
3. Go to [Social Messaging], [TouchPoints]
4. Click the edit icon right to website touchpoint (by default you have 1 Webchat TouchPoint available, but you can add more if you like)
5. Select [+ Status Messages for Webchat] and select the Language
6. Enter the Status Messages during closed hours en open hours.
7. Click Save and Close

-> If you want to use a schedule to show an ‘open’ or ‘closed’message, make sure you have created a schedule and assign the schedule to the Webchat TouchPoint. You can select the schedule on the Webchat configuration page under the option “Assign a schedule…”

‍

‍

‍

Load options

There are a couple of load options available for the webchat. To enable, declare an object on the page: “r365_webchat_options”. This object has the following properties:

fullScreen: boolean.
Opens webchat area with bigger height if true. False is default height.

conversationVariables: object.
When defined, this object is uploaded as conversation data for this conversation. This can be usefull when no startform is displayed, but you still want to have data from the client available in the flow.

startDirect: boolean.
When true, the startchat button is not displayed, but instead the chat proceeds to the next stage as if the enduser clicked the start chat button.

formCustomization: object (formCustomizationOptions).
Options to extend the pre chat form as you please.

Form Customization Options

addInputs: function(area).
Area is a jquery object to append the inputs to. This function is called when the prechatform is generated.

onValidate: function(conversationVariables).
This function is called when the user clicks the accept button on the prechatform. Use this function to validate your customizations. Return true if all validations pass, return false if not.

Also, update the conversationVariables object with the values you want to have available in your flow. Its your responsibility to generate error messages etc, if applicable.

Example load options

var r365_webchat_options = {
  'fullScreen': true,
  'conversationVariables': {'UserId': 'someusername'}
  'startDirect': true,
  'formCustomization': {
    'addInputs': function(area){
      area.append("<input id="myPrechatFormInput"… />");
    },
    'onValidate': function(conversationVariables) {
      var input = $('#myPrechatFormInput').val();
      if (input) {
        conversationVariables.myPrechatFormInput = input;
        return true;
      } else {
        //update your UI with an error message
        return false;
      }
    }
  }
};

‍

Cookie Consent

To respect cookie consent, declare a function on the page: “r365_cookie_consent”, that returns either true (consent given) or false (consent not given)

example cookie consent

function r365_cookie_consent() {
  if ( /*cookies accepted*/ )
    return true;
  else
    return false;
}

‍

What is an Adaptive Card?

When a new conversation arrives in a Microsoft Teams channel information about the conversation and the available action buttons are shown in an adaptive card. Based on your subscription you can configure which elements should be available on the card. This includes features but also in which datasource must be search to display CRM data in the adaptive card.

‍

Create and assign an Adaptive Card

1. In the ROGER365 portal go to [Social Messaging], [Microsoft Teams] and choose [-AdaptiveCards]
2. On the right-top choose [+ Add AdaptiveCard]
3. Choose the configured TouchPoint (available features will be displayed)
4. Type in a description
5. Select de desired features/components
6. Click Save

Now assign the adaptivecard to the Microsoft Teams TouchPoint

1. Go to [Social Messaging], [Microsoft Teams] and find your channel
2. Edit the channel by clicking the blue edit button
3. Select your created AdaptiveCard in the menu item named “Assign a AdaptiveCard…”
4. Click Update to save the new setting

‍

Connect Microsoft Teams channel to the ROGER365 Teams TouchPoint

You can watch the steps described underneath in this tutorial video

NOTE:The admin portal menu has recently been updated and the video underneath doesn’t reflect all new changes

‍

1. Login into https://admin.roger365.io with your Microsoft 365 account
2. Go to [Social Messaging], [Microsoft Teams] and choose +Add Teams Endpoint

Copy the Activation code to the clipboard by clicking the green square icon

In Microsoft Teams select the created channel and type [@ROGER] followed by a space (make sure ROGER changes into blue) and paste the Activation code and press [ENTER]. ROGER will answer you that the channel is connected succesfully and gives you the name of the TouchPoint.

Now go back to the ROGER365 .io admin portal to [Social Messaging],  [Microsoft Teams] and choose the new created Teams channel
Give the new channel a friendly name by selecting the blue edit icon from the Actions column
Click [Update] to confirm

‍

Add ROGER365.io App to Microsoft Teams

There are two options to add the ROGER App to the Microsoft Teams client: 1) Distribute it globally in your tenant to all users via the Microsoft 365 Admin portal or 2) install it locally in the Microsoft Teams client. In case of the 2nd option this must be allowed by the Microsoft 365 Administrator.

You can watch the steps described underneath in this tutorial video

Option 1: Install locally

1. Login into https://admin.roger365.io with you Microsoft 365 account
2. Go to [Microsoft Teams] and choose [Add Teams Channel]
3. Download the ROGER.ZIP manifest file
4. In the Microsoft Teams client, click on the [Apps] icon. You can find it at the left bottom of the navigation bar
5. Select the option [Upload a custom app] (If you don’t see this option, then this must be allowed by the Microsoft 365 Administrator) and select the ROGER.zip file
6. Then (important!) don’t click Add, but select the arrow to see the pulldown menu and select ‘Add to Team’. Select the Team or channel which you would like to use for ROGER365

‍

‍

Option 2: Install globally

1. Login into https://admin.roger365.io with you Microsoft 365 account
2. Go to [Microsoft Teams] and choose [Add Teams TouchPoint]
3. Download the ROGER.ZIP manifest file
4. As a global Microsoft Teams go to the Microsoft Teams admin center:
5. Choose [Teams Apps] > [Manage apps]
6. Choose upload new app and choose select file.
7. In the pop up browse and select the ROGER.ZIP file you and acknowledge the screen to get the upload started.
8. Then go to your Microsoft Teams client, click on the [Apps] icon. You can find it at the left bottom of the navigation bar
9. Search for the ROGER app and click it
10. Then (important!) don’t click Add, but select the arrow to see the pulldown menu and select ‘Add to Team’. Select the Team or channel which you would like to use for ROGER365

After a successful upload it could take op to 24 hours before the app is visible in the Teams Apps. You can also logout/login into Teams to have the ROGER app directly available.

‍

‍

Updating the app

If you have to update the ROGER app, go to the Teams admin portal, find the ROGER app and click update.

‍

This feature is only available in the Enterprise subscription.

‍

The AI Flow feature for Contact Center application of ROGER365.io enables powerful AI actions such as Transcript, Summary, Quality Assurance or Classification into your daily work. Recorded calls will be later processed by various AI action in the configured by portal admins manner.

The Audio Provider and Transcript Actions are mandatory to use AI Flow feature.

‍Disclaimers:

• ROGER365 platform does not use its own bots in this feature, you can only use AI bots created by Microsoft (OpenAI and other AI services)!
• All Microsoft AI services do have a cost aspect (including transcription).
• ROGER365 does not have an influence on the quality of used model. Results may vary.

‍ There might be situations where your AI model will be blocked by Microsoft if the conversation contains violent content or other factors that violate the Microsoft Terms of Service.

‍

To use AI Flow feature, portal administrators must first:

1. Create Audio Provider Action – Audio Providers
2. Create Speech to Text Endpoint for Transcript Action – AI Endpoints
   1. Optionally you can create endpoints for Summary, Sentiment, Quality Assurance and Classification Actions – AI Endpoints.
   2. (optional) Create Quality Assurance Form – Quality Assurance Forms
   3. (optional) Create Classfication Action Config – Classfication Action Configurations
3. Create AI Flow – Flows

Processing of each AI Flow action is starting up around 1 minute after call has ended.

‍

AI Flow errors are logged on the Logging admin page. If an AI Flow instance fails, it can be rerun on the Flows admin page. After three unsuccessful attempts the record is marked as Abandoned and won’t be rerun again.

You can retrieve results of AI Flow and/or each Action, using AI Flow API documentation.

Note: Results of Summary action can be seen (and/or edited) in Customer Jounery tab inside of Agent Assist’s Last Interaction section.

Prerequisite: Deploy text embedding model within created earlier Foundry resource group (used in Azure OpenAI Endpoint). Keep in mind the cost aspect for using text embedding model.

The Classification Action Configurations (part of the AI Flow enterprise feature) enables portal administrators to define clear instructions (descriptions) for selected Classification items (labels). These instructions are passed to created earlier text-embedding model, which evaluates completed call transcripts and automatically assigns the appropriate classifications. This AI powered action reduces the need for manual call classification, shortens processing time, and helps teams stay focused on their primary responsibilities by automating repetitive labeling tasks.

The AI Classification configuration must be assigned to created earlier Flow.

To create configuration for Classification action, navigate to “Classification Action Configurations” page in ROGER365.io admin portal and press “+Add” button.

• Name: Provide a name for the AI Classification config which will be visible at Flows admin page.
• Azure Open AI Endpoint: Select from list created earlier Azure Open AI Endpoint.
• Deployment Name: Provide deployment name of the text embedding model (unique within your Foundry resource group). If not altered, it’s the name of the text embedding model, for example – “text-embedding-ada-002”.
• Api Version: Provide API version generated by the deployed model, version can be found in Endpoint’s URL within Microsoft Foundry portal.
• Options: This section allows to select created earlier Classification items, upon selection portal admins can provide instructions (descriptions) for each Classification item.
  • Providing clear and relevant details is crucial for the AI model to accurately classify the customer call interactions!

For maximum accuracy of the model, portal administrators must avoid high similarity levels between each classification item. Alike descriptions will be pointed out at the top of the configuration window, also marking relevant text in red for quicker identification.

After creation, it is possible to test the accuracy of the model (according to its instructions) directly from the same admin page using “Test” button. It allows for even greater tuning of your text embedding model.

You can check whether specific call interaction has been automatically labeled by AI Flow’s Classification action via:

• Agent Assist
  

• Classification API

Important Notes:

• Manual Classification by an agent does not trigger the AI Flow’s Classification action.
• Accuracy of the AI Classification action is fully depended on – received call transcript, quality of text-embedding model and given instructions. The ROGER365.io platform does not have direct influence over received results.

The Quality Assurance Forms page allows you to create QA Forms used for AI-powered evaluation of conversation transcripts. These forms are part of an optional Quality Assurance Action and utilize the Chat Endpoint type of AI integration.

How it works

Each QA form includes:

• AI Instruction: A prompt that tells the AI model how to handle the transcript and questionnaire. We provide a example instruction to get you started easily, please note that you might need to change this instruction in order to work in specific customer scenario’s.
• Example Questions and Answers: Used to guide the AI in scoring the conversation.

The built-in example AI instruction expects transcripts to contain meaningful dialogue (e.g., real customer-agent scenarios). If the transcripts are of good quality, the AI should return a structured response as expected.

Note: The AI instruction should always be tailored to your use case. ROGER365.io includes a ready-to-use example form to help you get started, but customizing the AI instruction is essential for best results. ROGER365.io support does not cover writing or optimizing AI instructions.

Steps to Create a QA Form

1. Go to the Quality Assurance Forms section in the Admin Portal.
2. Click + Add to create a new form.
3. Provide a Name and Description.
4. Enter your custom AI Instruction.
5. Add one or more Example Questions with sample answers to guide the AI model.

Use the provided example form as inspiration, then adjust it to fit your own QA criteria and business context.

‍

Audio Provider Action is mandatory for AI Flow feature to work.

This action is responsible for getting the recording of a call which will be then later be processed by configured AI Actions. There are two available Audio Provider types – Oak and Tollring.

To create Audio Provider configuration navigate to “Audio Providers” admin page and press “+Add”.

Fill in the description (name) of your Audio Provider configuration; the rest should be filled in according to the OAK Clarify Voice Recording Integration or Tollring Compliance Recording solution section of the General documentation page.

Oak

Tollring

Depending on aquired license, the Tollring can provide their own call trascription that can be used in the Transcript AI Flow action. Check your Tollring license before enabling “Provides Transcript” option. The Tollring Audio Provider can be used with the transcription service provided by the Microsoft.

‍

The Flows page lets you create AI Flow configurations that are used in agent calls. From here, you can also rerun failed AI Flow instances. Each AI Flow has a rerun button showing the number of Failed instances, alongside counters for how many records are currently Queued, In Progress, or Abandoned. After three unsuccessful rerun attempts, the record is marked as Abandoned and won’t be rerun again.

After configuring mandatory Audio Provider & Transcript actions (or including optional actions), navigate to “Flows” page at ROGER365.io admin portal:

• Description – name of your AI Flow configuration
• Direction – for which call direction the AI Flow will be initiated
• Webhook Trigger Condition & Webhook URL – defines when the webhook is triggered during the call, and the endpoint (URL) to which the webhook request is sent

After this step, the Audio Provider and Transcript configurations must be assigned in the Transcript Action panel. If configured, assign optional configurations to the corresponding panels.

Transcript, Summary, and Sentiment actions have optional settings available. More information can be found in the Microsoft documentation.

• Transcript Action – Microsoft reference
• Summary Action – Microsoft reference
• Sentiment Action – Microsoft reference

Beside of that, it is possible to decide whether results of each actions supposed to be saved (in order to retrieve them later via APIs).

Each AI Action does have certain functionality:

• Transcript Action – responsible for transcribing the recorded call based on the recording delivered by the selected Audio Provider configuration.
  • Notes: The Tollring Audio Provider can provide its own Transcription service. Before enabling it, make sure your Tollring license includes this feature and the “Provides Transcript” option is enabled in the Tollring Audio Provider configuration. The Tollring Audio Provider can be used with the transcription service provided by the Microsoft.
• Classification Action – is responsible for automatic Classification of the call based on defined instructions in created earlier Classification Action Configurations.
• Summary Action – is responsible for summarizing entire call.
• Sentiment Action – is responsible for assesing sentiment score based on tone of the conversation.
• Quality Assurance Action – is responsible for evaluating call interactions with customers based on various criteria defined in created earlier Quality Assurance Form.

Created AI Flow must be assigned. This can be done in two ways:

1. Assign AI Flow to created earlier Agent Profile. This will enable an AI Flow for every agent assigned to Agent Profile.
2. Assign AI Flow to specific agent via Agents admin page.

Portal admin does have option offering an agent a button in Agent Assist tab to opt-out AI Flow for the current call, this option may be useful when certain conversation should not be processed by various AI actions. More information on Agent Assist documetation page.

‍

This feature is only available in the Enterprise subscription.

‍

Description

‍

Node Properties

‍

Mapping

‍

‍

‍

This feature is only available in the Business and Enterprise subscriptions.

‍Disclaimer: Before proceeding make sure that created earlier SMS TouchPoint can send messages out.

‍

Description

‍

Node Properties

‍

Disclaimers: Usage of DTMF action is required. To trigger this action only for the calling side, create check via WaitForEnd and If actions.

This feature is only available in the Business and Enterprise subscriptions.

Feeback scores can be reviewed in the Agent Assist, or be exported via FeedbackExport API.

Description

‍

Node Properties

‍

This feature is only available in the Enterprise subscription.

Description

‍

Node Properties

‍

‍

‍

This feature is only available in the Enterprise subscription.

‍

Description

‍

Link Properties

‍

‍

Mapping

‍

This feature is only available in the Enterprise subscription.

‍

Description

‍

This feature operates only with Set Variable connected underneath, variable used in SetAction must be the type of boolean. In screenshot example, expression for “Set IsVirtualCall” is true.

This action can only be used inside an active Queue action, meaning the caller is already waiting in the Queue’s loop for an available agent. The Queue and Make Call Virtual actions must both be part of the same callflow. If an agent has already been invited to the call, Make Call Virtual won’t work – even if that agent has not answered yet.

Edit JSON Content overview:

This feature is only available in the Enterprise subscription.

Before configuring this action you have to create Azure OpenAI resource at Microsoft Azure, you can find description in this Microsoft article – Create and deploy an Azure OpenAI Service resource.
Most of the configuration part must be done in Microsoft Foundry portal.

Microsoft Azure OpenAI is a preview feature, please be aware of unexpected results, and test thoroughly before implementing to production environment.

Description

‍

Node Properties

* Variable in JSON Content can be used when Evaluate Jsoncontent option is enabled.

‍

Mapping

‍

Setting up variable for result property is mandatory for action to work!

‍

This feature is only available in the Enterprise subscription

‍

Description

‍

Node Properties

‍

‍

Mapping

‍

‍

This feature is only available in the Enterprise subscription. To use this action, you have to create a Azure Cognitive Service resource in your tenant. You can find the description Create Azure Speech to Text Resource v1.0, language and voice support for speech service is described in this Microsoft article.

Description

‍

Node Properties

‍

This feature is only available in the Enterprise subscription. To use this action, you have to create a Azure Cognitive Service resource in your tenant. You can find the description Create Azure Speech to Text Resource v1.0, language and voice support for speech service is described in this Microsoft article.

Description

‍

Node Properties