Improve Sentiment

Do you want to increase positive sentiment Overall mood or result of the interaction as determined by analysis of words, phrases, and context of the transcript. during interactions The full conversation with an agent through a channel. For example, an interaction can be a voice call, email, chat, or social media conversation.? Follow this success guide to design a workspace Named view of one or more widgets. focused on agent or client sentiment. Improving sentiment can benefit your organization in the following ways:

• How agent behavior impacts customer sentiment.
• How clients feel about products, policies, procedures, and so on.
• How to improve customer satisfaction and agent training.

Interested in a more visual approach? Watch this short training on customizing a workspace for sentiment.

Set Up a Sentiment Workspace

These instructions are a recommendation for how to build a workspace Named view of one or more widgets.. They're meant to help you get started. You can customize workspaces in many other ways to suit your organization's needs. You can configure a dataset workspace using other widgets Graphical representation of data that meets specified filter criteria., metrics Statistics you can measure to keep track of your contact center, and filters not mentioned here.

Create the Workspace

1. Go to My Data and open the dataset you want to analyze.
2. Click the workspace panel if it is not already open, then click Add Workspace.
3. Click the Widget Drawer .
4. Select these widgets to add them to your workspace:

   • Overview Bar, under the Volume section. Select this widget a second time, so that two appear in your workspace.

   • Top Categories, under the Volume section.

   • Silence Percentage, under the Volume section.

   • Sentiment Trend, in the Sentiment section. Select this widget a second time, so that two appear in your workspace.

   • Frustration Trend, in the Frustration section.

   • Top Company Profile Terms, under the Volume section.

   • Metrics, under the Volume section. Select this widget a second time, so that two appear in your workspace.

   Some of these widgets require more configuration after you add them to the workspace. This is so you can view specific data related to sentiment. Other widgets use the default settings.

5. You can edit any of the widgets to adjust their size or position.

Configure Overview Bar Widgets

Configure one Overview Bar widget for client sentiment (the default) and the other for agent sentiment.

1. Click more options > Settings in the top right corner of the widget. Change the widget Title to Client Overview. That's it for configuring the client sentiment widget.

2. To configure the agent widget, click more options > Settings. Change the widget Title to Agent Overview.

3. Click the Widget Options drop-down and for Participant, select Agent.

4. Clear the checkbox for Display Interactions. The client Overview Bar widget already displays this information.

5. Click OK.

Configure Sentiment Trend Widgets

Configure one Sentiment Trend widget for client sentiment (the default) and the other for agent sentiment.

1. Click more options > Settings in the top right corner of the widget. Change the widget Title to Client Sentiment Trend. That's it for configuring the client sentiment widget.
2. To configure the agent widget, click more options > Settings. Change the widget Title to Agent Sentiment Trend.

3. Click the Widget Options drop-down and for Participant, select Agent.

4. Click OK.

Configure Metrics Widgets

Configure one Metrics widget to display the top performing teams and the other for top performing skills.

1. Click more options > Settings in the top right corner of the first Metrics widget.
2. Click the Widget Options drop-down and for Field, select the Team Name.
3. For Number to Show, enter a number from 1–25, the number of teams you want the widget to display data for. The default is 10.
4. You can customize the bar color or order in which the bars display in the widget.
5. You can give the widget a new Title, such as Top Teams.
6. Click OK.
7. Repeat these steps for the second Metrics widget but for Field, select Skill Name. You can make the same customizations and give the widget a new Title, such as Skills.

Filter Workspace by Sentiment

You can filter your workspace Named view of one or more widgets. by different sentiment Overall mood or result of the interaction as determined by analysis of words, phrases, and context of the transcript. types. You can also analyze overall sentiment, beginning sentiment, or end sentiment. Filter by each sentiment type to establish a baseline for interactions in your dataset Set of parsed interaction transcripts that spans a specific period of time and match your specified filter criteria..

1. Click in the top right corner of your workspace.

2. Click Add Filters.

3. Expand the Sentiment drop-down and click the sentiment you want to view: Positive, Negative, Neutral, or Mixed.

4. Define when the sentiment must have happened in the interaction. Choose from one or a combination of:

   • Overall Sentiment: The sentiment for the whole interaction.

   • Beginning Sentiment: The sentiment at the beginning of an interaction. This is determined by the first 400 words, or first 30%, of the interaction, whichever occurs first. It can help you identify common reasons or problems clients call about. It can also help you compare against end sentiment. If the client or agent sentiment changed over the course of an interaction, you can explore why. Mixed sentiment is not available for beginning sentiment.

   • End Sentiment: The sentiment at the end of an interaction. End sentiment is determined by the last 30% of an interaction. It can help you compare against beginning sentiment. If the client or agent sentiment changed over the course of an interaction, you can explore why. Mixed sentiment is not available for end sentiment.

The widgets in the workspace update to only display data for the sentiment filters you selected.

How to Use a Sentiment Workspace

Start by developing a theory about sentiment Overall mood or result of the interaction as determined by analysis of words, phrases, and context of the transcript.. Consider agents, teams, and topics you think may have strong positive or negative sentiment. Each widget displays details to help you test your theory. Compare widgets side by side to see specific issues. Your observations help you determine what filters can identify areas for improvement. The widget data that displays in the workspace Named view of one or more widgets. depends on the dataset parameters set for the dataset.

Choose one sentiment type to investigate first. Then filter by a different sentiment type to compare data and find areas where you can work on improving sentiment.

Overview Bar Widget

The Overview Bar widget displays big picture information about the data in the dataset Set of parsed interaction transcripts that spans a specific period of time and match your specified filter criteria.. You can include one or more of these three sections in the widget:

• Interactions: Number of interactions per minute, per hour, per day, and total in the dataset. Also shows increasing or decreasing trends in the number of interactions.
• Sentiment: Percentage of total interactions per sentiment Overall mood or result of the interaction as determined by analysis of words, phrases, and context of the transcript.. You can choose whether to show contact sentiment or agent sentiment.
• Volatility: How sentiment is trending over the time period in the dataset. Different interactions can display different sentiments. Sometimes sentiment swings occur within individual interactions themselves. This line chart helps you see how sentiment is trending throughout the entire dataset and time period being analyzed. The line trends upward when the data is considered volatile, that is, when there is an increasing amount of rapid change in sentiment. It trends lower when the data is more stable, or when there are fewer changes in sentiment.

Having these widgets at the top gives a general idea of how sentiment is trending throughout all interactions in the dataset.

Top Categories Widget

The Top Categories widget lets you see what categories Data groups that Interaction Analytics uses for parsed call transcript data. They make it easier to find trends. are most common. This is most helpful if you have categories organized by specific topics. View the Interaction Reason success guide for help creating custom categories.

Filter the widget to see which categories have the highest negative sentiment. This can help you identify interaction reasons to address.

Interaction Silence Widget

The Silence Percentage widget helps you identify interactions containing notable silence. In Interaction Analytics, silence is considered notable if it lasts three or more seconds. The widget shows the notable silence percentage for each channel. Total notable silence for an interaction is the sum of all notable silence periods.

Chats interactions tend to have a higher silence percentage than voice interactions. That's because there can be varying times between responses.

Sentiment Trend and Frustration Trend Widgets

View these two widgets together. Sometimes, negative sentiment Overall mood or result of the interaction as determined by analysis of words, phrases, and context of the transcript. and frustration Looks for cues to identify customer frustration. The cues include words and phrases like, "I'm very angry". Frustration isn't the same as negative sentiment. Frustration cues show that a customer is upset rather than just discussing something negative. don't mean the same thing. However, spikes in both negative sentiment and frustration usually indicate an issue. Frustration is only measured for the client, so you may want to put the Frustration Trend widget next to the Client Sentiment Trend widget. This makes it easier to see where negative client sentiment and frustration overlap. However, if you're interested in seeing how agents handled frustrated customers, place the Frustration Trend widget next to the Agent Sentiment Trend widget. Periods where negative feelings and frustration are high in all three widgets are good places to investigate.

If you see a spike in sentiment or frustration, you can select a different Date Range in the workspace filters. Focusing on interactions in a more specific date range can help you know what company profile terms and metrics to filter for. This can help you identify specific causes of negative sentiment.

Top Company Profile Terms Widget

This widget shows you the company profile terms mentioned most often. This can help you identify topics that clients typically complain about or are pleased with. To hide irrelevant terms, click the term and then Hide. To take a closer look at individual interactions where certain words appear, click the term and then Show Interactions.

Metrics Widgets

The Metrics Widget shows which teams, agents, or skills rank highest for each sentiment type. You can evaluate who is doing well and who may require further coaching. You can click more options > Settings in the widget to edit the display. You can also click any of the bars in the widget and then click Show Interactions to see more details.

Measure and Track Progress

A goal-focused workspace is most valuable when your insights are used to drive change. Be sure to measure and track your progress on a regular basis. This helps you see how you are improving and helps you set appropriate goals.

Run the Analytics reports to view your progress. You could view data on a monthly basis or after a coaching session to track progress. You can also monitor the Sentiment Trend widget often to track progress. Each time you run an Analytics report, you can compare that data to your baseline to measure your progress. Based on your improvements, you may also decide to use updated data to establish a new baseline to measure against. After you review insights from the workspace and Analytics reports, consider what milestones and supporting goals should come next to maintain a process of continuous improvement.

When you review report data, consider the effects of timing. Look for patterns that may occur on the same day of the week or time of year. For example, negative sentiment may be higher on weekends if wait times are higher. So, you might address that issue by having more agents available on weekends. Or positive sentiment may be higher after agents experience quarterly performance reviews. So, you may want to begin having evaluations or trainings more often to keep agents motivated and informed.