Use Catchr MCP to ask ChatGPT or Claude questions about connected Braze data. Review campaign delivery, compare Canvas variant conversions and investigate card engagement as you prepare your client report.
For freelancers and agencies, it gives lifecycle reporting a practical starting point: select the client workspace, define the reporting period and ask a focused question.
“Compare paid media performance across my client accounts and flag the biggest changes.”
Braze
Google Ads
GA4
Used by the world's leading companies
Build your analysis around campaigns, Canvas journeys and audience activity. Keep message results, variant results and daily user measures separate so each number has a clear meaning in your client report.
Review delivered messages and delivery failures by campaign and message channel. Use campaign names, tags, teams and scheduling types to organize the results around your client's lifecycle activity.
Check conversion window counts alongside the campaign's conversion behavior and window length. Keep send-time attribution separate when explaining how results were measured.
Compare Canvas variants using entries, conversions and attributed revenue. Review conversions attributed to entry time separately to keep the comparison consistent.
Canvas details also cover variants and steps, giving you context for journey reviews. Canvas analytics requests are limited to 14-day windows; plan longer reporting periods around that request limit.
Track event counts by event name and time, and review daily active users and new users as separate measures of audience activity.
For cards, compare clicks, unique clicks, impressions and unique impressions. These measures help you describe card engagement without treating card interactions as campaign message results.
A focused Braze investigation, a recurring client report and a question through Catchr MCP serve different needs. Choose the approach that matches the deliverable, using the same clearly defined entities and reporting periods.
Use source-focused analysis to investigate a lifecycle question, assemble a report for recurring reviews, or query connected Braze data through Catchr MCP for a specific client discussion. In every case, label the conversion basis and keep daily user measures separate from message and variant totals.
Connect Braze in Catchr using your workspace's REST endpoint and a Braze REST API key. Use the endpoint from Braze API settings, rather than the dashboard URL, and ensure the key has read access to the entities you want to report on.
Then connect your AI destination through Catchr MCP, select the relevant client workspace and start with a question about campaigns, Canvas results or audience activity. Confirm access to the required data before building a recurring reporting workflow.
Replace the bracketed values with your client's workspace, reporting dates and review thresholds. Each prompt asks for a defined analysis you can check before sharing with a client.
Using Catchr, review Braze campaigns in [workspace] from [start date] to [end date]. Show delivered messages and delivery failures by campaign ID, campaign name and message channel. Flag rows with delivery failures above [failure count threshold], sort them by delivery failures and include campaign tags where available without duplicating metric totals.
Copy promptUsing Catchr, compare variants for [Canvas] in [workspace] from [start date] to [end date]. Show entries, conversions, conversions attributed to entry time and attributed revenue for each variant. Keep the two conversion measures separately labeled, account for the 14-day Canvas request window limit and identify missing data before summarizing the differences.
Copy promptUsing Catchr, prepare a daily Braze activity review for [workspace] from [start date] to [end date]. Show daily active users and new users, plus a separate daily count for [event name]. Flag days when the event count falls below [event count threshold]. Do not sum daily active users into a period-level unique-user total or infer that the event caused a change in user activity.
Copy promptUsing Catchr, review Braze cards in [workspace] from [start date] to [end date]. Show clicks, unique clicks, impressions and unique impressions by card. Calculate clicks divided by impressions where impressions are greater than zero, label it as a calculated ratio and flag cards with at least [minimum impressions] impressions whose ratio falls below [ratio threshold]. Keep unique counts labeled with their reporting time buckets.
Copy promptYou need the workspace's Braze REST endpoint and a REST API key with read access to the relevant campaigns, Canvases, events, cards, segments or KPI series. Copy the REST endpoint from Braze API settings; the dashboard URL will not work for API requests.
Canvas variant data includes conversions and conversions attributed to entry time. Campaign message data includes conversion window counts and send-time measures. Compare matching definitions, and check the campaign's conversion behavior and window length before interpreting differences.
Canvas analytics requests are limited to 14-day windows. For a longer reporting period, account for that limit and check that the returned results cover your requested dates. The request window does not establish how much historical data is available.
No. Daily active users measure activity within each daily bucket. Adding those values can count the same person on multiple days. Use them to show daily trends, and present new users and event counts as separate measures.
Connect your sources to Catchr, add the MCP server to your AI assistant, and ask about the work that matters now.
