Creating report templates and adding notes just got easier. You can now copy text or images from anywhere and paste them right into the Template Editor or Annotations Editor — no need to upload or drag files around.
✨ Just use:
• Ctrl + V (Windows)
• Command (⌘) + V (Mac)
• Or right-click and select “Paste”
Whether you’re grabbing charts from another platform, screenshots, or snippets of text, it’s now simpler than ever to bring them into your reports.
Before, the only kind of data that you could assign to an Agent's Knowledge was datasets from our Data Cloud. We have released the ability to now assign PDF files to an Agent's Knowledge so an Agent can then take context of those files into account for any future conversations. Some examples of where this could be helpful: so that the Agent can understand company best practices when it comes to campaign optimizations or media strategy, customer goals and priorities, brand guidelines or team operating procedures.
To start, the Agent will only understand the text of those PDF files. It won't understand any charts or images in the files (if this is something that you need for a particular use case of yours, please reach out and let us know!)
Learn more here in our help docs, including limitations with this new feature: https://docs.ninjacat.io/docs/adding-data
We’ve upgraded the AI Insights widget to give you greater control, better prompt accuracy, and even deeper performance analysis. These changes are now live and ready to use in your reports.
✨ What’s New?
🎛️ Model Selector
You can now choose the AI model powering your analysis. Available options include:
• GPT-4.0, GPT-4.1, 4.1 mini, 4.1 nano, o3, o3 mini, o4 mini
🔁 Regenerate Button
Quickly rerun your analysis with the same prompt and settings. Found just beneath the “Analyze Performance” options, this feature makes it easier to refine results or explore new phrasing without starting from scratch.
🎯 Improved Prompt Adherence
We’ve fine-tuned how the system interprets your instructions, delivering sharper, more on-target responses even for highly specific prompts.
📊 Comparison Support
Generate side-by-side comparisons between campaigns, date ranges, or platforms—giving you clearer insights and more confident decision-making.
You can now add internal notes directly in the annotations editor! Notes are saved with the report instance and visible to all users on your team—making it easier to coordinate, document changes, and stay aligned. This feature is live now.
We’ve refreshed the AI Insights widget with a streamlined right-hand interface and improved formatting for AI-generated responses. These updates make it easier to navigate settings and deliver cleaner, more readable insights—helping you get to the point, faster.
There are two new options available in the Model dropdown at the Agent level: Claude Sonnet 4 and Claude Sonnet 4 -Thinking. According to Anthropic, "Claude Sonnet 4 is a significant upgrade to Claude Sonnet 3.7, delivering superior coding and reasoning while responding more precisely to your instructions."
Claude Sonnet 3.7 will continue to be an option at this time for your NinjaCat Agents, but we may retire it in the future now that this smarter version of the model is available.
For more information about Claude 4, please visit Anthropic's help docs: https://www.anthropic.com/news/claude-4
We're excited to share a behind-the-scenes upgrade to the AI Insights widget in your reports!
✨ What's New?
AI Insights now understands the full context of a report page. That means it can analyze all the widgets on the same page, helping it deliver insights that are more informed, relevant, and holistic.
🧠 More context = smarter insights
Whether you're using AI to summarize performance, suggest optimizations, or highlight anomalies, this enhancement gives the model a broader view of the data on your report page.
No changes to your workflow—just more powerful analysis.
As always, we'd love your feedback!
We're excited to introduce the Reports Workflow Manager — a streamlined way to track, manage, and take action on your scheduled reports.
With this new feature, you'll get real-time visibility into every stage of your reports:
Best of all, you can now rerun errored reports or trigger delivery manually — all from a single, intuitive interface.
No setup is required. The Reports Workflow Manager is already available in your account. Just head to the Reports section to explore the new workflow.
In the Export settings, there is now a Failure Condition available to optionally enable: Mark Failed if Zero Rows. This is a useful option if you would like better visibility into your scheduled Exports that may have run with zero rows. This failure would be seen in the Export Logs if the setting is enabled. Otherwise, your Exports with zero rows will be logged as a success.
We’ve upgraded the Shape platform to Facebook Ads API v22!
