What Is an Insight
An Insight is an alert automatically generated by the Tractian platform when sensors detect an anomaly in an asset's behavior. When the system identifies something outside the normal pattern (in vibration, temperature, or another monitored parameter) it creates an Insight with the detected failure mode, quantitative data, and a recommended action.
An Insight does not replace your team's technical assessment. It delivers the data and initial diagnosis; the failure confirmation happens through the Insight Check.
Each Insight includes the following information:
Field | What it shows |
Failure mode | E.g., Mechanical Looseness, Bearing Wear, Unbalance |
Criticality | Severity level of the detected anomaly |
Date and time | When the system detected the anomaly |
Asset and location | Affected equipment and area |
Quantitative diagnosis | Measured values vs. configured limits for the asset |
Prescription | Recommended inspection checklist or corrective action |
Before You Start
You need access to the Insights module in the Tractian platform.
To change an Insight's status or complete a check, you need Technician permission level or higher.
How to View Insights
In the top menu, click Insights.
On the listing screen, select the asset or location you want to analyze. Use the location selector at the top of the page to filter by plant, area, or specific machine.
Use the available filters to refine the view.
Click any Insight in the list to open the side panel with full details.
💡 Tip: Switch between Panel view (visual cards) and Table view (rows and columns) using the icons in the top-right corner of the listing. Table view makes it easier to compare multiple Insights at once.
Insight Status
The status shows which stage of the analysis workflow the Insight is in. You can update it manually as the investigation progresses.
Status | What it means |
Pending | The Insight was generated but has not been assessed yet. |
Under Inspection | An inspection has been started to investigate the anomaly. |
Check | The inspection is complete and is waiting for findings to be recorded in the platform. |
Awaiting Check | The physical inspection was completed, but the formal validation in the platform has not been done yet. |
Important: Manually changing the status does not replace a formal check. To record the analysis result, create or link an Event, and close the Insight cycle, you must complete the Insight Check.
Insight Panel Structure
When you open an Insight, you'll see the sections below. Understanding what each one shows is essential for correctly interpreting the diagnosis before any intervention.
1. General Information
Identifying data for the Insight:
Insight name and number
Asset and location
Linked inspection (when applicable)
Current status
2. Measured Values
Metrics extracted from the vibration signal, used primarily for bearing failure modes. Each indicator offers a different perspective on the severity of the problem.
Metric | What it represents |
Crest Factor | Ratio between the signal peak and the RMS value. High values indicate the presence of impacts. |
Defect Factor | Indicator of a localized defect in the component. |
K Factor | Sensitive to early-stage bearing wear, before peaks become visible in the spectrum. |
Peak-to-Peak | Total signal amplitude between the highest and lowest captured values. |
💡 Tip: These values are informational and complement the analysis. The Insight is generated from spectral analysis and the configured limits for the asset, not directly from these indicators.
3. Detection
Shows the specific parameter that triggered the alert:
Monitored parameter: e.g., BPFO (Ball Pass Frequency Outer race)
Measured value and percentage change from baseline
Defined limit for that parameter on the asset
Analyzed axis: Axial, Vertical, or Horizontal
Date and time of detection
4. Prescription
Customized inspection checklist for the detected failure mode. The items guide the technician during the physical inspection to confirm or rule out the root cause before the check.
To understand how prescriptions are configured, see Configuring Procedures for Insights.
5. History
Charts showing how parameters have evolved over time. Use this section to:
Verify whether the failure is progressing or stabilizing
Identify when the anomalous behavior started
Track the evolution of BPFI, BPFO, BSF, and FTF frequencies across Horizontal, Axial, and Vertical axes (available for Bearing Wear)
6. Harmonics
Visualization of the multiples of the detected failure frequency (e.g., BPFO1, BPFO2, BPFO3, BPFO4).
Use this section to confirm patterns characteristic of the failure mode. Multiple harmonics with increasing amplitude generally indicate that the failure is progressing.
7. Spectral Analysis
Comparison between three references to identify anomalous peaks and patterns:
Spectrum | What it represents |
Insight spectrum | Measurement captured at the moment the anomaly was detected |
Average alert spectra | Average asset behavior during alert periods |
Training spectrum | Healthy baseline condition of the asset, used as a reference |
Available in 2D and 3D views. Use the 3D view to track how peaks have evolved over time. For a detailed guide on interpreting spectral analysis, see Spectral Analysis.
8. Comments and Action History
Use this field to record information throughout the investigation process:
Observations and interim findings
Attached files: images, videos, audio, or documents
Mentions of team members using @name
All interactions are logged in the Insight history and can be reviewed at any time.
How to Check an Insight
The check is the step where you record the analysis result, validate the diagnosis, and close the Insight cycle. It can be completed with AI support (automatic assessment) or filled in manually.
With the Insight open, click Check.
Describe your observations in the text field. The more detailed the description (including photos, videos, or audio) the more accurate the AI-generated assessment will be.
Choose how to proceed:
Generate assessment: the AI analyzes your description and suggests the assessment type, failure mode, affected component, and problem status. You review before saving.
Fill in manually: you directly select the assessment type and complete the corresponding fields.
Click Save to record the check.
For the full step-by-step of the check, including all assessment types (Potential Failure, Functional Failure, Process Adjustment, and others), automatic Event linking, and what happens after saving, see Checking Insights.
How to Export or Print an Insight
With the Insight open, click the printer icon in the top-right corner of the screen.
The system generates the Predictive Analysis Report with all Insight data.
Save the file as a PDF or send it to print from your browser.
Frequently Asked Questions
Does the Insight disappear after it's checked?
No. The Insight remains in the history with the updated status and all check records. Insights assessed as "No Failures" are archived but can still be accessed.
Can I link an Insight to a maintenance Event?
Yes. During the check, you can create a new Event or link to an existing one for the asset. If there is already an unresolved Event, the platform may link it automatically. See full details in Checking Insights.
The spectral analysis panel is blank. What should I do?
The training spectrum is generated after the sensor's learning period. If the asset was recently installed, wait for the baseline period to complete. If the sensor has been installed for more than 30 days and the analysis still isn't showing, contact Tractian support.
Can I change the Insight status without completing the check?
Yes, the status can be changed manually at any time. However, changing the status does not record the analysis result or create/link Events. To formally close the Insight cycle, the check is required.