If you've scheduled a Gumshoe report to repeat (weekly, biweekly, or monthly), you can view the history of all past and individual runs directly within the platform.
Where to Find Your Report Trends
The Trends page provides a visual overview of your report's performance over time, helping you spot trends, track improvements, and compare AI search visibility across past runs. While there's a smaller trends graph on your main report page, there's also a much more detailed Trends page.
To View the Trends Page:
- Go to any report in your Gumshoe account.
- Select "Trends" in the left-hand menu.
- Review the graphs to compare all previous runs for that report in one place.
- Change the page filters to see different information.
Each run shows the brand's AI visibility based on the conditions and model data at that time. This is an effective way to monitor changes and improvements over time, accounting for the variability in the AI model's responses.
Where to Find Previous Versions of a Report
You can view the entirety of any specific version of a report in a series.
To View Previous Report Runs:
- Go to your report's main landing page.
- Locate the timestamp near the top right (e.g., "Run 2 days ago").
- Click the dropdown arrow next to the timestamp.
- A list will appear showing all previously completed runs of that report.
- Click any entry to load and review the results from that specific run.
Notices and Event Markers in Trend Graphs
Trend graphs may include notice markers that indicate Gumshoe system-level updates or industry changes. These markers provide context for changes that could impact visibility results, such as prompt methodology updates, model upgrades, or provider version changes. When you see a notice, it means results may shift even if you haven't made any changes, and neither have your competitors. Use these markers to distinguish between changes driven by outside updates and changes caused by your own optimization efforts. This helps ensure trend analysis remains accurate and prevents misinterpreting normal system evolution as performance issues.
If you want to learn how to schedule a report to repeat, click here.