What’s a dataset in Talkdesk Explore™? It’s a set of dimensions and metrics or a single statistical matrix with dimensions and measures, organized in such a way that Explore users can easily get the answers to their performance queries and to make it as easy as possible for them to find the answers to their questions.
What’s the difference between Dimensions and Measures?
Dimensions are the contextual fields in your dataset, the independent variables, the x in your distribution.
Metrics or measures are what you want to assess on dimensions of behavior. For instance, a specific agent that you’ll identify by name is a dimension, but you want to know how many calls were answered by that agent in the last 30 days, so the number of calls is your measure.
For customizing your own reports and dashboards we have 6 distinct datasets so that you know exactly where to start and what kind of metrics and dimensions you can expect to find.
Agents Activity Analysis Dataset
As the name suggests, this dataset is dedicated to agents’ metrics. It’s here you find statuses, duration related metrics, as well as the speed of answer, calls answered, and missed calls.
The way that the Ring Groups and Team objects are currently stored in the database differ, meaning that you’ll have the information of the Ring Groups returning information based on the logic of “as was” in one hand and in the other, the Team information will be returned “as is”.
For example, Let’s assume I pick Team Support and Agent Ring Group support_en for the past month of May.
The results returned will show for all the agents that now belong to the Support team (as is) but for the Agent Ring Group support_en, it will show me all the Agents allocated to that Ring Group in May (as was).
Calls Analysis Dataset
As for Calls Dataset, this is the more detailed and complete dataset for calls dedicated info. It is similar to a dump of the information from our platform in a dataset format. In this specific dataset, you’ll be able to filter by almost all dimensions.
This is an exclusive capability of this dataset, most dimensions can be used as a filter. This particularity of the dataset along with the fact it is the most thorough and complete compilation of information regarding the record of each call, means that is the most flexible and complete dataset, providing a great level and freedom of analysis.
This can also be challenging and demanding, which is why a clear understanding of the dimensions and what each metric will return is advised, so we stress the importance of using the Data Dictionary.
Calls Dataset offers unique ways* to aggregate your information, anchored to the start time or end time of calls. The following aggregation options are available:
- Date (yyyy-mm-dd)
- Day of Month (1, 2, 3 . . . )
- Day of Week (Monday, Tuesday . . . )
- Hour (yyyy-mm-dd hh)
- Minute (yyyy-mm-dd hh:mm)
- Minute10 (yyyy-mm-dd hh:10, yyyy-mm-dd hh:20)
- Minute15 (yyyy-mm-dd hh:15, yyyy-mm-dd hh:30)
- Minute30 (yyyy-mm-dd hh:00,yyyy-mm-dd hh:30)
- Month (yyyy-mm)
- Quarter (yyyy-Q)
- Time (yyyy-mm-dd hh:mm:ss)
- Week (yyyy-mm-week starting day of the week)
* not available in the other datasets.
Let's look at the data entry 49 (third from top). You can see the AVG Talk Time is 1555.0 (approximately 25 minutes). This figure includes all calls that occurred Tuesday of the configured timeframe and that finished within the hour of 20:00.
Contacts is the historical piece of the New Data Model. From now on, you can access full details and information within a specific interaction. For instance, if you have a transfer on record and you wish to consult the handle time, the total time, or the agent assigned you will use the Contacts Dataset. Here you can add Measures and/or Dimensions, as the Contact ID, the Interaction ID, or the Service Level.
Ring Attempts Dataset
This dataset provides the ability to check under each contact and/or call the detail of the ring attempts. This means that you’ll be able to see who rejected a call, for instance, or to confirm if your agent ignored an inbound contact.
Numbers Analysis Dataset
Numbers Analysis Dataset is dedicated to everything phone number related. Here you can find the thresholds set for each of the numbers in your account. Both Service Level threshold and Short-Abandoned threshold details for each number can be derived from this dataset. Besides these configuration details you can also check the Ring Groups associated with Number and performance-wise, the volume of calls and also some time-based metrics.
Ring Groups (Queues) Analysis Dataset
The Ring Group Dataset is dedicated to providing metrics and dimensions related only with the Ring Groups’ activity and performance. For instance, to retrieve information regarding Service Level for a specific Ring Group, this is the dataset you’ll use.
You can find detailed information for this dataset by following the link to our Data Dictionary for the article dedicated to Ring Groups (Queues) Dataset.
For fully detailed information on each dataset available and for information on the default dashboards, reports, and metrics in Explore, we’ve compiled a comprehensive glossary that we’ve called the Data Dictionary.
The Data Dictionary is split into a Glossary of Call Types, Scheduled Reports, Dashboard Historical Metrics and Create Feature Datasets, with each section providing a thorough assessment or definition of all the terminology used in our Reporting product.
Using the Data Dictionary
Use the Search  function to enter a metric and the Data Dictionary will return all articles with matching entries.
The results are displayed in order of relevance  and you can display them all by clicking View All Results .
Alternatively, you can use the browser’s search facility  (CTRL+F). Any matching entries will be highlighted in both the left navigation menu  and in the content of the section your viewing .
You also have the option of simply scrolling to, and selecting a specific entry for the dataset or Metric you’re working on .
The detail is presented for each dataset in the following ways:
|Name of the field that can be a Metric/Measure or a Dimension
|Detail on what that Dimension or Metric represents
|Statistical nature of the outcome - number, sum, average, max, min
|Formats can be 0.00% - percentage0 - number at unit level00 - number representing a count[hh]:mm:ss - duration
|Within each Dataset what is the section (view label) the Metric or Dimensions belongs to
|If the Metric or Dimension can be used as a filter
|If the field listed is a Dimension or a Measure/Metric
Reports and Metrics
|Name of the Metric or measure
|Details of that particular Metric or Measure
|A breakdown of how the Report or Metric is calculated
|How the Metric is shown - as a percentage, a count, or a timestamp.
|If it’s an agent or a queue (ring group) metric