The purpose of this document is to briefly clarify and compare what is the added value of Talkdesk’s Contacts Data Model.
Watch these videos to learn more:
- New Data Model Explained
- New Data Model Explained: using Live
- New Data Model Explained: using Explore
What is the Contacts’ Data Model?
Talkdesk’s Contacts Data Model is a new data paradigm that focuses on the analysis of Contact data, instead of Call data.
When you analyze your contact center data based on Contacts, you have more detailed information and new measures and dimensions, which provide more flexible and powerful data exploration. With the Calls Model it was not possible to have this level of detailed data, with meaningful information to explore.
The powerful paradigm of the Contacts’ Data Model diverges from Talkdesk’s historical call reporting, allowing you to have more granularity with contacts data and its flows. Your organization now has metrics with more details, which can better support your data-driven decisions.
Your organization can improve the customer experience with the Contacts’ Data Model, because it will transform the data you need into valuable and meaningful insights you can use to drive your business goals.
The main objective of the Contacts’ Data Model is to increase organizations’ visibility into several metrics by presenting data related to Contacts.
Stages of an interaction (call) in the Contacts’ Data Model
The Contacts’ Data Model allows customers to track an interaction more accurately. The interaction is split into three distinct levels of detail:
Interactions are the macro object that encloses the pre call events (IVR or others), the handling piece of the call when the agent(s) and the contact person actually interact with each other and ends when the call records the last event of post-call Studio flow (like phone surveys or automations for data collection). The legacy data model is based on interaction data only, thus excluding the detail of each contact.
Interactions include all events from the time a call is connected to a flow to the end of the call, including After Call Work (ACW). Contacts are only created when an interaction reaches a queue, therefore interactions that don’t reach the queue don’t have contacts.
A Contact is a collection of segments that generally starts with a wait segment and ends with another wait segment, an After Call Work (ACW) segment, or the termination of the contact.
A Contact is only created once an interaction reaches a queue. No pre-call data is shown in contacts. Contacts are the primary object from which contact center metrics are calculated. Each segment of a contact, the duration of those segments, and the events within the contact are used to report the dimensions and metrics for contact center KPIs.
A segment is a small component of an overall interaction between the contact person and an agent. Examples include Queue, Ring, Talk, Hold, After Call Work (ACW), among others.
Segments can overlap for one interaction. For example, two segments coincide after the interaction is transferred: this will be evident for when we have the first agent ACW, where we’ll be able to have the time in ACW that will overlap with when the customer is placed into the new, post-transfer queue.
New Data Model Metrics
New Metric - Average Handle Time (AHT)
The New Data Model provides a new metric for Live - Average Handle Time (AHT). This is a new metric, so there is no comparison with the Call Data model, just an explanation of what it represents.
Average handle time (AHT) is a new metric that measures how long an agent is occupied while handling a contact for a given queue. Individual values for handle time are a sum of a contact’s talk, hold, and ACW time. The expected use case for this metric is staffing and forecasting.
The new calculation method (from the Contacts’ Data Model) of Service Level percentage is the number of contacts answered within your organization’s waiting time threshold within business hours by Inbound contacts (minus short abandoned contacts) The threshold value is configurable through Admin > Preferences as well as through numbers’ custom settings.
The new calculation considers the Abandoned contacts in the denominator something that was not recognized before. Before the service level could show a percentage of 100% while the Abandon Rate would also be high and that was not being considered in the calculation.
One other accuracy advantage of this calculation is that because it is made at contact level (and not call level as before) it returns the Service Level of each queue involved in an interaction.
The Contacts’ Data Model will calculate Service Level as follows:
[All contacts answered within threshold in Business Hours /
(All inbound contacts - Short abandoned contacts)] in Business Hours x 100
Note: Service Level is an exclusive metric for queue, this is because agents have no control over how long a contact waits in a queue, and no control over which contacts get routed to them.
An example of the Service Level in the Contacts’ Data Model
The Contacts’ Data Model has a new way to evaluate the Service Level and we provide an example to highlight how it works.
The use of Contacts instead of Interactions may either increase or decrease your Service Level (%). For example, consider:
- Wait time threshold = 60 seconds
- An interaction has two contacts:
- Wait time for contact 1 = 65 seconds
- Wait time for contact 2 = 30 seconds
- Legacy Data Model - Service Level:
- (65 sec + 30 sec) > 60 seconds so interaction is not within SL
- The queue (Ring Group) for contact 2 gets the impact in SL% because the data is associated to the final Ring Group of the interaction
- New Data Model - Service Level:
- 65 sec > 60 sec so contact 1 is not within SL; 30 sec < 60 sec so contact 2 is within SL
- Contact 1 queue (RG) sees a decrease in SL% because the wait time for that queue is finally being reported against that queue
- Contact 2 queue (RG) sees an increase in SL% because it’s only looking at the wait time for that queue
The example provided is just an overview of how Service Level works in the Contacts’ Data Model, because the changes can increase or decrease SL% and it’s largely dependent on the operations and specific use cases of each Contact Center:
- High or low abandon rate?
- High or low number of missed calls?
- High or low number of transfers?
When analyzing the Service Level, please keep in mind that:
1. Using contacts instead of interactions means it’s not always the last queue that gets tagged with the SL%. This means you could see an increase in the last contact of an interaction. You could see an increase or decrease in queues for contacts that aren’t the last contact of an interaction (see example above).
2. Including abandons means SL% is likely to decrease.
- The new definition for abandons, however, means it would not decrease as much as it would have with the legacy definition of abandons. Nonetheless, this also means your SL% will be a clear representation of what is going on on your contact center.
3. Counting missed as not within means SL% is likely to decrease.
4. The date filtering has an impact on SL% only from the perspective of which interval the contacts/interactions are counted in.
- You have a lot of contacts that start during hour 1 and end during hour 2.
- There are no transfers for those contacts.
- All of those contacts are answered within SL%.
- Legacy SL%:
- Hour 2 SL% will have the positive benefit of those interactions being answered within SL because that’s when the interactions ended.
- New SL%:
- Hour 1 SL% will have the benefit of those contacts being answered within SL because that’s when the contacts started.
Calls Data Model
Contacts Data Model
Data based on
End date of interaction
Begin date of contact
Only accounts for inbound and missed calls within the Service Level Threshold (meaning the waiting time until the call is answered or disconnected by the contact person after being presented to an agent without being answered).
Only contacts answered within Service Level Threshold are considered positively for the calculation. Abandoned contacts have a negative impact.
Sum of all wait time during the interaction
Wait time for the contact
Reporting Ring Group
Last Ring Group of the interaction
Ring Group of the contact