Connect Criteo to ChatGPT

Multi-Advertiser Performance Command Center

Using the Catchr MCP, pull Criteo data for [List of Client Advertiser Names/IDs] over [Period] and [Comparison Period]. For each advertiser, summarize Cost, Display, Clicks, Click Through Rate, Visits, Cost Per Visit, ROAS PC 30d, and Cost Per Order PC 30d, then identify the Campaign Name or AdSet Name responsible for the largest change. Compare every account with [Client KPI Targets or Historical Baseline], label it Healthy, Watch, or Critical, rank the portfolio by urgency and spend exposure, and give the consultant one evidence-based action for each Watch or Critical account.

Client-Ready Retail Media Performance Brief

Using the Catchr MCP, analyze Criteo performance for [Client Advertiser Name/ID] over [Period] versus [Comparison Period], broken down by Campaign Name, AdSet Name, Channel Name, and Device. Report Cost, Display, Clicks, Click Through Rate, Visits, Qualified Visits, Conversion Rate PC 30d, Cost Per Order PC 30d, Advertiser Value, and ROAS PC 30d. Explain which campaigns, ad sets, channels, and devices drove the change, separate reach and traffic gains from commercial efficiency, and produce a concise client-ready brief with key wins, risks, and three prioritized actions.

Delivery and Audience Risk Watchlist

Using the Catchr MCP, review Criteo delivery for [List of Client Advertiser Names/IDs] over [Period], segmented by Advertiser Name, Campaign Name, AdSet Name, AdSet Status, and Audience Name. Use Potential Displays, Display, Win Rate, Reach, Exposed Users, Viewable Displays, Non-Viewable Displays, Untrackable Displays, Cost, and Click Through Rate to find weak auction access, low viewability, limited reach, or spend with poor engagement. Compare with [Comparison Period or Thresholds], return a prioritized client-by-client watchlist with the exact evidence and business risk, and recommend a specific next step such as reviewing audience scope, ad-set status, bid strategy, or campaign allocation without claiming a cause the data cannot prove.

Morning Client Priority Queue

Using the Catchr MCP, pull Criteo performance for [List of Client Advertiser Names/IDs] for [Recent Period, e.g., yesterday or month to date] and [Comparison Period]. For every client, report Cost, Display, Clicks, Click Through Rate, Visits, Cost Per Order PC 30d, and ROAS PC 30d, then identify the Campaign Name or AdSet Name behind the largest negative movement. Compare results with [Client KPI Targets or Historical Baseline], rank clients by urgency and cost at risk, and give me a morning priority queue with one clear task to complete today for each of the top three accounts.

Monthly Client Results Narrative

Using the Catchr MCP, pull Criteo data for [Client Advertiser Name/ID] for [Reporting Period] and [Previous Period]. Break down Cost, Display, Clicks, Click Through Rate, Visits, Qualified Visits, Advertiser Value, Conversion Rate PC 30d, Cost Per Order PC 30d, and ROAS PC 30d by Campaign Name, AdSet Name, and Channel Name. Write a concise client-ready report in plain English explaining what improved, what declined, which segments drove the result, what the attribution window means for interpretation, and the three actions planned for [Next Period], with every conclusion tied to the reported data.

Campaign Optimization Backlog

Using the Catchr MCP, analyze Criteo campaigns for [Client Advertiser Name/ID] over [Period]. For each Campaign Name and AdSet Name, review AdSet Status, Cost, Display, Clicks, Click Through Rate, CPC, Visits, Cost Per Visit, Conversion Rate PC 30d, Cost Per Order PC 30d, and ROAS PC 30d against [Minimum Volume] and [Client KPI Targets]. Diagnose whether the main constraint is delivery, engagement, traffic quality, conversion, or return, then create a freelancer-friendly backlog ranked by expected impact and effort, with one specific action, success metric, and review date for every item.

Profitable Growth and Attribution Scorecard

Using the Catchr MCP, pull Criteo results for [Advertiser Name/ID] over [Period] and [Comparison Period], broken down by Campaign Name and AdSet Name. Report Cost, Advertiser Value, Cost of Advertiser Value, ROAS PC 1d, ROAS PC 7d, ROAS PC 30d, ROAS PV 24h, Conversion Rate PC 30d, and Cost Per Order PC 30d. Show how profitability and efficiency changed across attribution windows, rank campaigns by commercial contribution after applying [Minimum Cost or Display Threshold], flag campaigns that consume cost without meeting [Target ROAS or eCOS], and recommend where to protect, reduce, or investigate spend.

Omnichannel Revenue Contribution Review

Using the Catchr MCP, analyze Criteo omnichannel performance for [Advertiser Name/ID] over [Period] versus [Comparison Period]. By Campaign Name and Channel Name, compare Omnichannel Revenue PC 30d, Omnichannel Sales PC 30d, Omnichannel ROAS PC 30d, Revenue Generated Offline PC 30d, Sales Offline PC 30d, ROAS Offline PC 30d, Cost, and Advertiser Value. Quantify which campaigns and channels drive reported omnichannel and offline outcomes, highlight cases where the performance conclusion changes materially between the views, and provide three actions for budget review, measurement validation, or channel coordination while treating attributed revenue as attribution rather than proven incrementality.

Visit-to-Sale Efficiency Diagnostic

Using the Catchr MCP, evaluate the Criteo path from ad exposure to commercial outcome for [Advertiser Name/ID] over [Period], broken down by Campaign Name, AdSet Name, Device, and Os. Use Display, Clicks, Click Through Rate, Visits, Qualified Visits, Bounce Rate, Cost Per Visit, Conversion Rate PC 30d, Average Cart PC 30d, Cost Per Order PC 30d, and ROAS PC 30d. Compare with [Comparison Period], identify where traffic quality or conversion efficiency deteriorates, rank the affected segments by wasted-cost risk, and return a prioritized optimization plan with the metric to improve, the recommended action, and the success criterion for each test.

Attribution Window Consistency Audit

Using the Catchr MCP, audit Criteo attribution for [Advertiser Name/ID or List of Advertisers] over [Period], grouped by Date and Campaign Name. Compare Conversion Rate, Cost Per Order, ROAS, and eCOS across PC 1d, PC 7d, PC 30d, PV 24h, PC 30d PV 24h, and Client Attribution variants where available. Flag missing window values, abrupt changes in the gap between windows, impossible negative results, and campaigns whose ranking changes materially by attribution view. Return a structured exception table with advertiser, date, campaign, affected fields, observed values, severity, likely measurement impact, and the next validation step.

Delivery Data Integrity Monitor

Using the Catchr MCP, check Criteo data quality for [Advertiser Name/ID or All Connected Advertisers] over [Period] using Date, Extracted Date, Advertiser ID, Campaign ID, AdSet ID, AdSet Status, Cost, Potential Displays, Display, Viewable Displays, Non-Viewable Displays, Untrackable Displays, Clicks, Visits, and Reach. Detect stale extractions, missing dates or identifiers, duplicate dimension rows, negative metrics, active ad sets with no records, clicks above displays, visits above clicks, or viewability components that do not reconcile with Display. Quantify each issue and return a remediation-ready table with the failed rule, affected scope, severity, evidence, and recommended pipeline or source-system check.

Daily Performance Outlier Detection

Using the Catchr MCP, pull at least [Lookback Period, e.g., 90 days] of daily Criteo data for [Advertiser Name/ID or List of Advertisers], segmented by Advertiser Name, Campaign Name, Channel Name, and Device. For Cost, Display, Clicks, Click Through Rate, CPC, Win Rate, Visits, Conversion Rate PC 30d, Cost Per Order PC 30d, and ROAS PC 30d, compare each day with its trailing [Baseline Window, e.g., 28-day] mean and standard deviation and flag deviations beyond [Outlier Threshold, e.g., 2 standard deviations]. Classify each anomaly as likely extraction, tracking, attribution-mix, delivery, or genuine performance change using cross-metric evidence, and provide the exact date, segment, affected metrics, severity, and investigation step.

Replace [variables] Facts ≠ hypotheses Action + follow-up metric

Connect Criteo to ChatGPT

Trusted by marketing teams that talk to data everyday

How to connect Criteo to ChatGPT ?

Authorize the Criteo account in Catchr

Add the Catchr MCP from official ChatGPT store

Choose, adapt, and run a prompt.

A simple client question should not trigger an hour of data prep.

Pull another CSV

Rebuild the context

Summarize it by hand

Ask the account. Get to the explanation.

Prepare data before every question.

Connect once. Keep asking.

Twelve Criteo prompts for analyzing real account data.

Criteo to ChatGPT FAQs

What Criteo data can ChatGPT analyze through Catchr?

How granular can Criteo analysis be in ChatGPT?

Can ChatGPT find efficiency and budget opportunities in Criteo?

Can ChatGPT diagnose targeting, creative, or conversion issues in Criteo?

What types of prompts work best for Criteo analysis?

Can I combine Criteo with other data sources in ChatGPT?

Can I analyze multiple accounts or campaigns together?

What do I need to connect Criteo to ChatGPT with Catchr?

How recent is the Criteo data, and how much history can I analyze?

Can ChatGPT change anything in Criteo through Catchr?

How much does it cost to connect marketing data to ChatGPT with Catchr?

‍Can I try the ChatGPT integration before choosing a plan?

Still have a question ?

Ask the account question before building another export.