Add Threadseer
Open your report in a supported Power BI authoring environment and add Threadseer to the report canvas from the Visualizations pane using the distribution provided to you.
Quick-start guide
Getting started with Threadseer
Add one visual, map three fields, and see the paths your work actually follows. Your report and process data stay in Power BI and are not sent to us.
Release status: Threadseer has not been submitted to Microsoft Marketplace and is not yet available to install publicly. This guide applies to the current private pre-submission build.
Start in three steps
Map three fields
Add Case ID, Activity, and Timestamp. Apply any Power BI filters that should define the data you want to analyze.
Choose Analyze
Select Analyze this selection. Threadseer loads the complete filtered dataset, builds the process, and opens the workspace.
Threadseer does not require a separate platform, Threadseer account, Python runtime, or preprocessing application. Power Query is optional when your source data needs reshaping.
What your data needs
Each row should represent one event in a case, such as an order update, ticket status, or claim step. Your column names can be different from Threadseer's field names.
| Required field | Purpose | Example |
|---|---|---|
| Case ID | Connects events that belong to the same order, ticket, claim, or other case. | Order ID, Ticket Number |
| Activity | Names the step that happened. | Approved, Shipped |
| Timestamp | Places the events in time order. | Status Changed At |
If your milestones are stored in separate date columns, use Power Query to unpivot them so each milestone becomes a row. Use precise timestamps when sequence matters. When events share the same timestamp, Threadseer uses their source-row order and reports the condition in Data Quality.
Optional fields and what they add
| Optional field | What it adds |
|---|---|
| Activity Category | Category labels and colors on the process map. |
| End Timestamp | Separate processing-time and queue-wait measures. Without it, Threadseer does not guess which kind of delay occurred. |
| Event / Row ID | A direct connection to an individual Power BI source row when one is available. It does not change the calculations. |
| Resource | People, teams, or other resource context for investigation. It is not proof of utilization or capacity. |
| Outcome | Outcome-rate comparisons. |
| Value or Cost | Value and cost summaries. Threadseer uses the Power BI field's format for mapped-value labels and tooltips, including currency, decimals, percentages, and negative-number formats. |
| Additional Dimensions | Comparisons using up to three extra business fields, such as location, product, or priority. |
When your filters change
Threadseer analyzes only when you choose Analyze or Reanalyze. Resizing the visual, changing formatting, or moving between views does not start a new analysis.
After you change a mapped field or Power BI filter, Threadseer keeps the previous results visible and tells you they need refreshing. Choose Analyze new selection when you are ready to replace them. Selections you make inside Threadseer narrow the current results without rebuilding the full analysis.
If Power BI cannot load the entire filtered dataset, Threadseer stops instead of analyzing partial data. Read-only dashboard tiles may receive only the first part of a larger selection; when that happens, open or refresh the source report to load and analyze the complete selection.
Choose a view for your question
See the process
Process Map shows the activities and paths cases follow. Variants ranks the most common routes. Compare places two observed routes side by side. Conformance compares complete case routes with paths you mark as Accepted.
Find delays and rework
Drivers shows where queue wait, processing time, or rework is concentrated. Bottlenecks summarizes delay evidence. Rework finds repeated steps and return loops. Improve explores regularity, handoffs, resources, and shortcuts without making causal, staffing, compliance, or return-on-investment claims.
Investigate the evidence
Cases traces results back to matching cases and a selected timeline. Data Quality explains excluded, normalized, and retained input conditions so data problems remain visible.
How Conformance works
Mark observed paths as Accepted and choose one accepted path as Primary. A case conforms only when its complete activity sequence exactly matches an Accepted path. Primary changes the visual emphasis; it does not change the score.
The map distinguishes Primary, accepted-only, shortcut, repetition, unexpected-path, and deviation evidence. You can zoom, fit, pan, and use the keyboard to inspect it. Paths saved from another filter context remain manageable even when the current data contains no matching cases. Deviation details link back to contributing cases and Driver Analysis.
Working with larger datasets
- Start with a representative filtered dataset, especially when it contains many different activities, routes, dimensions, or cases.
- Community accepts up to 25,000 event rows in each analysis.
- Professional has a supported baseline of 100,000 event rows. Larger attempts depend on Power BI, WebView2, available memory, the device, and the shape of the data.
- If Threadseer stops with resource guidance, narrow the Power BI filters and choose Analyze again.
See Plans & Pricing and Known Issues and Limits for the complete boundaries.
Your data stays in Power BI. Threadseer has no external network permission and does not send event data, field names, report names, case identifiers, or process metrics to Wittenauer Software.