To ensure a successful implementation of Freeform Topics, it would help if you did some preparation to address use cases for your organization before proceeding to create Freeform Topics. We recommend you review the following first:
Learn more about Freeform Topics and its uses.
Have a general understanding of how to use APIs, webhooks, or AWS EventBridge.
Prepare Freeform Topics you want Agents to capture in Conversations
What Conversation attributes do you want to capture through Freeform Topics? Based on the kind of analysis or type of business you run, you might want to construct and capture attributes like:
Order Number
Courier
IATA Code
Locations
Codes
SKUs
Product ID
Proceed to create these as Freeform Topics.
Set up your analytic tool to receive Freeform Topics usage
Transmit Freeform Topic usage to your preferred analytics tool via API, webhooks, or AWS EvenBridge. See Analyze Freeform Topics Attributes for setup information.
Custom attributes power Freeform Topics through API
Custom attributes power Freeform Topics through API, the Conversations API, the Events API, and events used in Webhooks.
Determine structure for applying Freeform Topics
When applying a Freeform Topic attribute such as Order Number, Agents apply the same format as your actual order numbers, which are inherently unique to each order. The same applies to SKUs or IDs. For other Freeform Topics, such as Sentiment, Location, or Courier, it is recommended to provide clear specifications and options to Agents for applying these attributes to avoid variances.
Format examples
Location | Courier |
|---|---|
Store #123 | FedEx |
Store #456 | USPS |
Store #789 | UPS |
Store #1011 | OnTrack |
Using Courier as an example, it's best to provide specifications on how Agents should apply FedEx and not of FDX or USPS, not US Postal. For Location, Agents should format entries like Store #123, not Store123 or Store number 123.
Train Agents how to apply Freeform Topics
Once you create the Freeform Topics in Gladly and the API, webhooks, or AWS EvenBridge is configured, take the time to train Agents on how to use Freeform Topics.
Go over the formatting requirements when applying a Freeform Topic.
Freeform Topics formatting
Values can only be composed of letters, numbers, and symbols.
They cannot be more than 255 characters.
Values must be unique to a Conversation, meaning you cannot add the same value twice in one Conversation.
Values are case sensitive (PO1234 and po1234 will be considered different).
Help Agents understand that the Topics menu will contain Freeform Topics and their standard Topics list.
Train Agents to understand the difference between applying a Topic(s) and/or Freeform Topics to Conversations and how they could apply both. Show Agents how to tell the difference when applying a Topic or a Freeform Topic, with the latter exposing a freeform field to enter a relevant attribute.
Agents can also apply multiple Freeform Topics. For example, if multiple order numbers are relevant to the Conversation, Order Number Freeform Topic can be applied multiple times, and each is associated with a relevant order number.
As an additional benefit outside of analysis, highlight to Agents the benefit of obtaining more context about a Conversation based on Freeform Topics already applied to a Conversation.
Freeform Topics count as a required Topic
Applying one Freeform Topic satisfies the Topic Assignment setting, which requires at least one Topic to be applied to a Conversation.
Help Agents find where Freeform Topics and Topics appear for context
Topics and Freeform Topics applied to a Conversation appear in the Conversation Details card in the Customer Profile.
They also appear in the Conversation Timeline headers.
