Connect Taboola through Catchr MCP to ask ChatGPT or Claude about your client’s advertising performance. Review campaign ROAS, compare click-through and view-through conversions, and investigate viewable impressions before your next client call.
For freelancers and agencies, it brings Taboola questions into the same AI workflow you use to prepare analysis and explain results.
“Compare paid media performance across my client accounts and flag the biggest changes.”
Taboola
Google Ads
GA4
Used by the world's leading companies
Start by discovering the fields available for your connected Taboola account, then select the columns that answer your question. Use supported combinations of campaign, conversion, and delivery data to keep each analysis focused.
Review campaign names and IDs alongside available ROAS, CPA, conversions value, and conversion counts. Identify campaigns that need a closer look before you make recommendations to a client.
Where available together, add campaign status or learning state to give the performance review context. Select those details when they help answer the question.
Compare conversions attributed to clicks with conversions attributed to views. Taboola fields include separate conversion counts, CPA, conversion rates, and ROAS for each attribution type.
Keep the attribution labels visible in client reporting. Conversion look-back windows are configurable per conversion rule, so confirm the relevant settings before interpreting differences.
Use impressions, visible impressions, clicks, CTR, and VCTR to examine delivery and engagement. CPM and VCPM provide separate views of cost per thousand impressions and cost per thousand viewable impressions.
Discover which breakdowns support your selected metrics, then explore available site IDs, countries, regions, platforms, or audiences. Start with one breakdown to make the results easier to interpret.
A focused campaign review, a recurring client report, and a follow-up question each need a different workflow. Catchr supports source-focused analysis, reporting destinations, and questions about connected Taboola data through MCP.
Use a report for a consistent Taboola performance view and Catchr MCP for the questions that follow. A change in CPA can become a focused campaign or attribution comparison using the fields available to your account.
Connect Taboola in Catchr using its OAuth authorization flow and select the relevant client account. Choose ChatGPT or Claude through Catchr MCP, then ask which Taboola fields and breakdowns are available before selecting your reporting columns.
Replace the bracketed values with your client account, dates, and review thresholds. Each prompt starts with field discovery so the analysis uses supported columns and combinations.
Using Catchr, discover the available Taboola campaign fields for [account]. For [start date] to [end date], use the supported combination of campaign name, campaign ID, ROAS, conversions value, and conversion count. Show the available columns for me to confirm before querying. Rank campaigns by ROAS and flag those below [ROAS threshold]. Keep percentage units explicit and distinguish missing values from zero.
Copy promptUsing Catchr, discover the available Taboola attribution metrics for [account] for [start date] to [end date]. Let me select supported columns for click-through and view-through conversion counts, CPA, and ROAS. Compare the two attribution types in a labeled table and draft a short client explanation. Identify which conversion-rule look-back settings I should confirm before drawing conclusions.
Copy promptUsing Catchr, discover the available Taboola site breakdown for [account] for [start date] to [end date]. Let me confirm a supported site identifier and available columns for clicks, impressions, visible impressions, CTR, and VCTR. Compare sites with at least [minimum visible impressions] and highlight differences between CTR and VCTR. Keep this as a period-level site review without campaign details, and explain any unavailable field combinations.
Copy promptUsing Catchr, discover which Taboola country or region breakdowns and performance metrics are available for [account]. Let me choose one geographic breakdown and supported columns from clicks, conversion count, CPA, and ROAS. Compare [first period start] to [first period end] with [second period start] to [second period end]. Show the changes, flag rows below [minimum clicks] for cautious interpretation, and draft client meeting talking points without claiming a cause the data cannot establish.
Copy promptAuthorize Taboola in Catchr through OAuth, select your client account, and choose ChatGPT through Catchr MCP. Start by asking for the available Taboola fields, then select the columns and reporting period for your first question.
CTR measures clicks relative to all impressions. VCTR measures clicks relative to visible impressions. Comparing them alongside both impression counts helps explain how the choice of denominator changes the engagement rate.
Taboola fields include separate click-through and view-through conversion counts, CPA, conversion rates, and ROAS. Discover which are available for your selected breakdown, and confirm conversion-rule look-back windows before interpreting the comparison.
Discover available fields first and select only the columns needed for the question. For site reporting over a long period, start with fewer dimensions or a shorter date range if the query times out. Add campaign details only when they are needed for the analysis.
Connect your sources to Catchr, add the MCP server to your AI assistant, and ask about the work that matters now.
