Set up Escalation Predictor to Preempt Tickets

Support agents usually develop a gut feeling for cases that are quick to solve and cases that carry with them a risk of escalation. Such agents are worth their weight in gold. If you are a support manager and find such an agent, hire them right away. But if you cannot find such an agent, try Escalation Predictor.

Like an old hand, Escalation Predictor forewarns when a case is about to go south so you can take appropriate action to thwart an escalation.


  • Escalation Predictor works only on cases which have the following fields indexed: Case Number, Created Date, Closed Date, Last Modified Date, Escalation Field, Subject, and Status, Priority or another categorical field.
  • Before connecting your case management system with Escalation Predictor, ensure that the system has at least one escalated case.


  1. Click Apps from main navigation, and then Install New App.

  2. Install Escalation Predictor.

Input Data

Escalation Prediction needs to access cases before it can start analyzing them. In this section, you will learn to connect the app with a case management system.

  1. Expand Apps, click Escalation Predictor, and then click Connect with a Content Source.

  2. Escalation Predictor analyzes the historical cases in your system in order to predict escalations. Select a content source. NOTE. As of November, 2020, only Salesforce supported.

  3. To establish a correlation between cases and escalations, SearchUnify uses the IsEscalated case field or a custom field set up by Salesforce admins. Select that field and set its value to True for IsEscalated or a value that represents an escalation in your custom field.

  4. Escalation Predictor uses an advanced model which trains on cases' data. The more the data, the more accurate the model. With Start Date, you decide how much data to keep. Pick a date and click Next. Only the tickets logged after Start Date are analyzed.

Transform Data

Escalation Predictor trains on cases to find those factors that lead to escalations. Depending on your case management system, the factors can easily run into dozens. The factors frequently include a case's title, a product's category, and the purchase amount. All of these factors (or case fields) are put into one of the three available categories in the app: Textual (used for sentimental analysis), Categorical, and Numerical.

  1. Drag all the text fields, such as case titles, case description, and numbers into Textual. The Categorical field is for fields like priority, product group, and category. Use the Numerical field for storing number of comments exchanged, purchase amount, and other fields with numbers. NOTE. Case number is always a Textual field and both Closed Date and Created Date are always Categorical fields.

The data after connecting Salesforce with Escalation Predictor.

Feature Classification

NOTE. The Feature Classification tab appears only when you click Start Classification in Input Data.

Based on the cases data in Textual, Categorical and Numerical, Feature Classification generates visual reports. Review them to spot patterns. For instance, for an organization which prioritizes A, B, C, D and E; if D is the largest then the cases with priority D are more likely to escalate. Also, the keyword cloud based on an analysis of case messages help you identify the triggers which suggest an imminent escalation.

The analysis run on cases today might or might not be relevant six months down the line. It's recommended to click Refresh, when case volume increases significantly. For instance, after inputting data of 100,000 cases, you might want to run Refresh when the number of cases increases to 110,000 or 120,000.

Scroll down to view them all.

Scoring Data

In Scoring Data, the metrics to be used in forecasting are picked. A rule of the thumb is to check the metrics with high correlation and uncheck the ones with low correlation and then click Retrain Model.

Example: In the next image the most accurate predictor of escalation is Days to Respond, which stands at 21%. As the Days to Respond number goes up, so does the probability of an escalation. It's better to keep it checked. However, case fields that have little correlation with escalations can be removed. In the next image, Solutions Area can be unchecked.

Escalations Dashboard

Escalations Dashboard predicts the likelihood of escalations on incoming cases. The probability is expressed through a number—where 0% means the case will not escalate and 100% guarantees an escalation—and one of these three emoticons: (irate user), (usual user), and (content user). Clicking the emoticon on a row expands the row, showcasing the snippets associated with an emoticon.

Expansion is a great time-saver for SU Admins who don't have to skim through entire case bodies to figure out why a particular customer is content or unhappy. Expansion highlights the snippets exuding customer reaction.

The other columns in the report (from left to right) are: # (Number), Case Number, Case Title, Created Date, Email, and Likelihood to Escalate. Out of these columns, Created Date is sortable.

The only critical setting on this screen is the Minimum Threshold box. Too low a number floods the screen with cases and by placing too high a number, you miss letting imminent escalations slip under the radar.

Email, Download, and Alerts

The data on Escalations Dashboard can be emailed and downloaded. You can also set up hourly, weekly, and monthly alerts on potential escalations. To send email Escalations Dashboard data, click .

From Select Date Range, pick a time period. The data for the selected period is emailed. In the second field, Email Address, write the email addresses of the people who are to receive the data potential escalations data. Finally, click Send.

A CSV file with the subject "SearchUnify: Escalation Prediction Analytics" is send to each person mentioned in Email Address. Click Download

Open the file in Excel, Sublime, or a similar program. The file ought to have five columns: Case Number, Case Title, Created Date, Contact Email, and Escalate Score.

Download works like Email. The only difference being that instead of sending data over email, it is downloaded to your device's hard-disk. To start, click .

Pick a time period from Select Data Range and click Download.

A CSV file is downloaded. Open the file in Excel, Sublime, or a similar program. The file ought to have five columns: Case Number, Case Title, Created Date, Contact Email, and Escalate Score.

Both Email and Download are great to find out potential escalations at any point in time. Alerts serves a different purpose. Instead of having to log into your instance, you get emails on set intervals, which can be Hourly, Daily, on Monthly. At the end of the set interval, an email is shot to subscribers sharing with them Escalations Dashboard data in a CSV file, just as it is the case with Email and Download. To start, click . Alternatively, you can email and download this data as well.

Configure the time interval in Set Frequency and in Email Address write the email addresses of the people who are to receive the alerts. Then, click Send.

People who receive emails are Subscribed Emails. There can be a maximum of five Subscribed Emails in an instance.

Last updatedMonday, August 2, 2021

Was this article helpful? Send us your review at