Feb 17 release now available

The February 2017 Release of Oracle Service Cloud (OSvC) is now available to the public and brings some cool new features and functionality that touch the Customer Portal (CP), the Contact Centre experience, the Knowledge Base (KB) and Oracle Policy Automation (OPA).

Experience/Skills-based routing for all channels

This is one of the features that many OSvC users have been waiting for. The Feb 17 release now supports Experience/Skills-based Routing not only for Chat (introduced on the May 14 release) but for all channels – email, social, web.

This means that the assignment of all incidents, based on agent’s skills will be possible, and allow you to optimise the routing of incidents – assigning it not only to the right team, but to the best person, increasing the likelihood of a positive outcome and first contact resolution.

To have this, make sure you ask Oracle to enable the “Smart Interaction Hub”.


Widget Inspector

The widget inspector aims to make life easier for developers and CP administrators. It helps you identify widgets and inspect its attributes and settings. This new feature allows you to access CP admin pages, inspect widgets and change them without having to touch any code.

You can highlight and inspect widgets on a CP page, while in development mode. When you select a widget (or more than one) the page shows the widget being highlighted with the widget name superimposed over the top of the widget. Then, by selecting the widget name you can see the attributes of that widget.

Knowledge Advanced

The new features in Knowledge Advance bring additional reports that will help you look into agent knowledge usage, allowing knowledge managers to differentiate usage of knowledge articles by customers on the Customer Portal, and by agents on the Agent Desktop.

These new reports provide knowledge manages additional insight and key metrics on agent behaviour and engagement, helping them improve the effectiveness of knowledge based articles.


Another new feature is related to Travel Ontology (industry dictionary). Oracle added the latest terms and abbreviations for the travel industry, including IATA airport codes to the dictionary, in 22 languages. This will improve search results via contextual recognition of IATA airport codes.

The Recently Viewed Answers widget has now also been extended to the Advanced Knowledge (it was already available for Knowledge Foundation), and will help improve navigation, UX and ultimately CX.


Lastly for Knowledge Advance, a new feature that will allow the automatic notification of knowledge managers and authors on changes or updates needed on answers assigned to them. The Author Task Notifications is now available to be configured and will make things much more pro-active and efficient.

Oracle Policy Automation

In Feb 17 release, Oracle introduced some Connection Enhancements which will allow various users to use one OPA interview, including anonymous users, logged-in customers, and staff/agents.

Another new feature coming to OPA is the ability to synchronise OPA HUB users with other systems, enabling automatic creation of user accounts using the OPA HUB REST API.

The OPA HUB REST API for deployments will also help to automate policy life-cycle tasks, like the promotion from Test to Production environments, and the ability to deploy, explore, activate, and retrieve policy models on an OPA hub.

Also, the new export project data model feature allows you to export the attributes that define an OPA project’s data model. The new capability enables you to review and analyse attribute metadata and detailed entity and relationship information.

REST API Enhancements

The Feb 17 release bring enhancements to the REST API allowing you to download multiple file attachments as a single compressed file. Additionally, the Campaign (Outreach) object is now available in CCOM to retrieve information on email outreach campaigns via Connect API.

High Volume Bulk Delete API

Another of the features many of us will be happy to hear about is the High Volume Bulk Delete API, which allows you to delete large volumes of incidents in a single API for incident archiving or purge.

Oracle Social Cloud Accelerator

Adding to the list of current “accelerators”, Oracle is releasing the new Social Relationship Management (SRM) accelerator. Allowing you to respond to SRM posts from Oracle Service Cloud, configure private and public channels (Facebook, Twitter, etc.), prevent duplicates with interactive contact matching and view full conversations in Oracle Service Cloud.

Note: One of the novelties Oracle Product Management Team introduced in the last few releases was a video where one of the guys presents and talks us through the new features. You can see it here. This time is Jeff Wartgow, Director of Product Management, who presents.

3 Options To Set Up Service Mailbox

As you know, one of the key strengths of Oracle Service Cloud (OSvC) is its email management capability, which handles email communication between customers and staff members. OSvC receives incoming email and manages all back-and-forth email replies between staff and customer until the incident is resolved.

For that you need to set up Mailboxes in OSvC, which are used to define the email accounts that OSvC should pull incoming email messages from, as well as outgoing email settings. There are two types of Mailboxes:

  • Oracle Managed Service Mailbox: Created, managed and monitored by Oracle.
  • General Service Mailbox: Created, managed and monitored by (your) Email Provider.

Even though this is a standard function of OSvC there is still a few questions and clarifications asked when it comes to the implementation. This post aims to expose options to set up, and hopefully answer a few questions.

Option 1 – General Service Mailbox (Intermediate)

With this setup, the flow would be:

  1. Customer sends email message to email@company.com
  2. Email message is received by the mailbox in the email provider servers
  3. Email message is redirected to an intermediate mailbox in email provider servers
  4. OSvC collects and deletes email message from intermediate mailbox
  5. OSvC creates Incident, and assigns to queue/agent
  6. Agent responds to customer’s email message
  7. OSvC sends response impersonating email@company.com

Oracle Service Cloud - Generic Service Mailbox Inter.png

Option 2 – General Service Mailbox (Direct)

With this setup, the flow would be:

  1. Customer sends email message to email@company.com
  2. Email message is received by the mailbox in email provider servers
  3. OSvC collects and deletes email message from email@company.com mailbox
  4. OSvC creates incident, and assigns to queue/agent
  5. Agent responds to customer’s email message
  6. OSvC sends response impersonating email@company.com

Oracle Service Cloud - Generic Service Mailbox Direct.png

Option 3 – Oracle Managed Service Mailbox

With this setup, the flow would be:

  1. Customer sends email message to email@company.com
  2. Email message is received by email provider servers
  3. And gets redirected to an Oracle-Managed Mailbox e.g. company@custhelp.com
  4. OSvC creates incident, and assigns to queue/agent
  5. Agent responds to customer’s email message
  6. OSvC sends response impersonating email@company.com

Oracle Service Cloud - Oracle Managed Service Mailbox.png

Oracle Tap – Service Cloud on Mobile

Mobility is key in today’s digital world, where information travels “at the speed of light” and businesses need to be always aware, and have real-time data feeds.

The Oracle Service Cloud Mobile Agent App – named Oracle TAP – was introduced back in 2013. Since then Oracle has enhanced its capability significantly.

Its aim is to allow Customer Service directors, managers and team leaders to be on top of things, and be able to see what is happening, remotely, from a mobile device.

It is important also to say that the Oracle Service Cloud Mobile Agent App is not a full-blown Agent Desktop, where users can see everything and anything.

The data available in the Mobile Agent App is only that related to the open Incidents. Meaning that you will only be able to see Organisations, Contacts, Tasks, etc. related to open Incidents (i.e. not “Solved”).

That said, the capability is actually outstanding. First, it is very easy to set up, as you just need to add the Server URL (https://YourSiteURL/cgi-bin/InterfaceName.cfg/php/tap/v1), and it is possible to have multiple sites set up.

20170127 Oracle TAP - login.png

The user can create his/her own home page with Charts, and other useful elements that can give an immediate overview of the customer service operation.

20170127 Oracle TAP - Charts.png

The main menu allows the user to access the list of staff (and drill down into the Incidents assigned to each of the staff members), Incidents, Contact, Organisation, Tasks, and even Answers in the knowledge base. The Incidents area, for example, shows “My Inbox”, but also the list of all open Incidents.


Tapping on one of the records (e.g. an Incident) allows the user to see the key details (e.g. Subject, Status, Queue, etc.) as well as the detail associated to it (e.g. Message Thread, Tasks, etc.).

20170127 Oracle TAP - Tasks.png

Integration: Connect Web Services and .NET API

One of the best, and most challenging, implementation projects I had the opportunity to be involved with as a Lead Consultant and Project Manager was in a company with 50 offices, 700,000 customers, a global sales & marketing organisation and 2 contact centres (Europe and Asia).

Oracle Service Cloud was chosen as the platform at the centre of this key strategic project which aimed to achieve transformation in Sales, Marketing, Customer Service, and back-office Operations.

As the main Customer platform, and to provide the users with a full 360-degree view of the customer, Oracle Service Cloud had to be integrated with legacy Oracle eBusiness Suite (on-premise), as well as a bespoke system (Amazon Cloud). Oracle Fusion Middleware was already in place to be the glue that would join up all systems.

To deliver the full 360-degree view of the customer, Oracle Service Cloud would have to allow staff members to see key customer information (demographics, contacts, etc.), customer details (roles, relationships, interactions), finance and other information, stored in Oracle eBusiness Suite.

Furthermore, the full 360-degree view of the customer in Oracle Service Cloud would also have to display company-specific information, associated with each customer, stored in a bespoke system, hosted in the Amazon Cloud.

The challenge was not only to have a synchronisation of key customer information, and service requests (created by legacy processes in Oracle eBusiness Suite), but also display the additional data (from Oracle eBusiness Suite and bespoke system) without holding it in the Oracle Service Cloud database.

Two types of integrations were developed:

  • Near Real-Time type integration (in grey) to synchronise key customer information and service requests, between Oracle eBusiness Suite and Oracle Service Cloud databases;
  • Surfacing type integration (in red) to get customer details, finance and other information from Oracle eBusiness Suite and the bespoke system, and display it in a custom screen embedded in the Contact workspace.

Below is a brief overview of the architecture:


These two integrations were developed, leveraging the Oracle Service Cloud Connect Web Services for SOAP (CWS SOAP), and the .NET API that allowed our team to build the integration and the custom .NET components, controls, and applications needed for the custom screen.

The project go-live was a few months ago. After a period of stabilization this post is not only to share the experience, and show the power and art of the possible with Oracle Service Cloud, but also to celebrate the effort of a fantastic team (made of Capventis and Client resources).

6 Must-Haves for Agent-Facing Knowledge Base

Every contact centre manager wants – in all honesty, desperately needs – to help agents deliver a much better customer experience. But unfortunately there hasn’t been much originality in the solutions and practices to try and gear them up.

Soft skills training and call scripting are usually chosen to try and make agents deliver a better service, but it hasn’t worked out well. Fresher and smarter approaches are needed to replace these traditional ways. Contact centre managers need to invest in agent-facing knowledge and contextual guidance. Technology is crucial to help enable this.

Most companies have CRM systems able to manage customers, incidents or interactions, but often miss a knowledge base or, even better, a CX platform that would allow the Customer and Case Management capability to be tightly joined up with the knowledge base.

I’ve seen contact centres use Dropbox to store troubleshooting guides, MS SharePoint to build wikis, Google Drive to hold how-to guides, and even gigantic MS Power Point documents with product FAQs and answers. Those tools can give the illusion of an internal and centralised knowledge base, but they are not even close to what needs to be deployed to actually give agents what they need.

A true agent-facing knowledge base needs to have certain capabilities that the aforementioned tools cannot deliver – because they were not actually developed to do so! And the end goal must be to increase agent productivity, and improve key performance metrics (such as FCRR or response time), leading to increase in key customer metrics (such as NPS and CSAT scores).

Here are 6 of the capabilities that a true agent-facing knowledge base must have:

  1. Content creation and authoring – to allow the person responsible for the content (e.g. knowledge rep or manager) to manage content. Create, update or retire. This will enable an easier, faster and controlled deployment of knowledge throughout the organisation.
  2. Categorisation and scoring – for content to be organised, structured, and linked to interaction drivers, as well as dynamically ranked by usefulness. This will enable presentation of the most relevant and effective content first, improving efficiency.
  3. Self-Learning – automatic linking between answers based on search and usage, as well as manual association (by the knowledge rep or manager) of sibling or related answers will help agents find the right answer quicker.
  4. Step-by-step guidance – embedded in knowledge base articles, step-by-step guides give agents the ability to easily and quickly reach the information they need by selecting responses from question branches in guides. Or even, for example, guide customers through a troubleshooting issue.
  5. Usage and management – reporting on the usage and usefulness of each article, as well as the gaps (e.g. keywords search without a match) will allow the knowledge rep or manager to keep a fresh, ever-green, and up-to-date knowledge base.
  6. Keyword, intent and category search – a keyword Google-type search (potentially enhanced by semantic search technology, linguistic dictionaries and advanced algorithms) will match searches to the most relevant content and deliver the right answers immediately.

Configuration Settings – Tips II

Oracle Service Cloud has circa 500 configuration settings, which control the platform’s functions. Some of them are commonly used but many are not that well known. Below are a few that you might want to be aware of.


This configuration setting will enable the links/URLs sent by customers to agents, during a chat interaction.

By default, and for security reasons, this configuration setting is set to “No”, meaning that links/URLs sent by customer would not be actionable and agents would not be able to just click and open a browser.


This configuration setting is related to the Incident Status, and sets the time an incident is on “Waiting” status until the agedatabase utility automatically closes it.

The minimum value is 6 (hours) and the default is 48 (hours). In case you would like to de-activate this you should set it to 0 (zero).


This configuration setting is directly linked to the one above (CI_HOURS), as it defines which days of the week are taking into account when CI_HOURS is calculating how long the incident was on “Waiting” status.

By default all days of the week are considered: Sunday, Monday, Tuesday, Wednesday, Thursday, Friday, Saturday. And that is why you see the code “SuMTuWThFSa” in the setting.

If you leave the configuration setting blank, all days of the week will still be considered.


There are a few configuration settings (see list below) that control the access to various Oracle Service Cloud areas, based on domain names or IP addresses. When you set these configuration settings you are limiting access only to users or customer who are logging in from the defined domains or IP addresses.

The values you can put on the configuration setting are: domain names (e.g. *.oracle.com), IP Addresses (e.g., or IP Subnet Masks ( Since Aug 2014 release, these settings also support IPv6 addresses (e.g. FE80:0000:0000:0000:0202:B3FF:FE1E:8329)

  • SEC_VALID_ADMIN_HOSTS – access to Admin interface (Console)
  • SEC_VALID_CHAT_API_HOSTS – access to the Chat SOAP interface
  • SEC_VALID_ENDUSER_HOSTS – access to the end-user interface (Customer Portal)
  • SEC_VALID_INTEG_HOSTS – access to the integration interface

[Guest Post] Co-Browse & Reports

This is a guest post from Colin Campbell. My colleague at Capventis. A seasoned Oracle Service Cloud expert, with c. 10 years experience. To whom I thank for this post, as well as  for his invaluable expertise, which makes our implementation team even stronger.

I have to admit it, I love writing reports for service cloud, and whenever I have someone saying that they have an issue with a report the first thing I look at is the report definition.

This can be a bit confusing for some people but recently this immediately let me see that an out-of-the-box report (“Co-Browse Summary”) had a basic flaw.

Co-Browse 193 Report Definition.PNG

The first thing I noticed was the column definition for the Source:

if(cobrowse_sessions.tbl=1, msg_lookup(7475), msg_lookup(7494))

This gives a simple Binary result, if the co-browse session was initiated through a Phone Call or a Chat Session. You can see in the tables that the Interactions (what Incidents are called in this environment) is linked to the incident ID if the Co-browse session table = 1 (an Interaction).

This immediately made me think of the following situation, if a chat request comes in which includes a co-browse session, and then gets converted to an incident, it will have an incident ID, and as a result would be shown as a Phone Call rather than a Chat.

When I finally got round to running this report I saw exactly that. In the image below you can see Luis has had a phone call on the 18 Nov (row ID #9) and has an Incident Reference number next to it.

Co-Browse 193 Report Running.PNG

Opening the incident associated to this entry, you can see that the incident was actually originated from a Chat session (Channel = Chat).

Co-Browse phone call.PNG

So this could present an issue to the analysis of the data by whomever is using co-browse, as the out-of-the-box report is not showing the true channel of the incident and you could end up inflating the number of phone calls ending with a Co-Browsing session, which can then impact your resourcing and potentially having the wrong skilled staff waiting when it should be chat skilled staff. This increased wait will ultimately impact the Customer Experience.

It is not all doom and gloom as this is actually quite an easy fix for your reporting, ant you would just need to edit one table of the report.

Co-Browse Initial Modification Report Definition.PNG

The first change I have made is to the Source Column, rather than having the two message lookups I have edited the true condition to look at the incidents channel ID:

if(cobrowse_sessions.tbl=1, incidents.chan_id, msg_lookup(7494))

Using Luis’s incident above will mean that this co-browse session should now show it is a Chat session and not a phone call (notice again row ID #9).


You can see here that my assumption is correct but it has shown something else. There are now some sources which are a “No Value” (and one as an “Email”).

The “Email” source itself is easy to explain and although co-browse is intended to originate from only two channels Phone and Chat, there can be occasions where a customer will phone about an e-mail, social media post, letter, etc. that they initially contacted you through with the phone call resulting in a co-browse session.

The “No Value” on the other hand is not, this is where a co-browse session has been initiated but it has not been associated to an incident, although you need to have the incident or chat workspace open for co-browse if you don’t save it, there is nothing to record it against.

To report on this I would look at updating the definition once again to include an additional IF statement against the source column so it reads as…

if(cobrowse_sessions.tbl=1, if(incidents.chan_id IS NULL, ‘Unlogged Phone’, incidents.chan_id), msg_lookup(7494))

With this last edit you can now define all of the channels and friendly terms that you need to aid you in your reporting and analysis.

Co-Browse Final Modification Report Running.PNG

Using this report as a base you can also add either a Contact ID or Incident ID/Ref No. Filter and embed this as a report within a Contact or Incident Record showing the history of the Co-Browse sessions.