How Zenith Replaced Hours of Manual Analysis with Real-Time Celeste AI Insights for Accor
Overview
Zenith International manages paid search advertising for Accor across two large MCCs, spanning thousands of campaigns on Google and Microsoft Ads in markets around the world. As the volume of data grew, so did the gap between having information and being able to use it. Analysts spent hours pulling reports, reconciling formats, and combining insights from multiple engines to understand performance and identify opportunities. The process was manual, time-consuming, and difficult to scale. By the time insights reached decision-makers, the moment to act had often passed.
The challenge was not access to information. It was transforming information into action. Zenith needed a faster, more intelligent way to uncover insights and respond at the speed of search.
Thousands of Campaigns. Not Enough Hours to Analyze Them.
Comparing performance across both Google and Microsoft Ads manually created a constant reporting burden. Different tools, inconsistent outputs, and multi-step workflows to reconcile different formats meant that by the time insights were ready, the window to act had typically passed. For a global advertiser as large and complex as Accor, that gap between insight and action continued to grow. Zenith needed to close that gap without adding resources or complexity.
Turning Celeste AI into a Search Intelligence Partner
Zenith deployed Celeste AI as the connective layer across both MCCs. Instead of pulling data from multiple systems and manually comparing results, the team built structured prompts that standardized how Celeste retrieved, analyzed, and compared data across the two accounts. This removed the variability of ad hoc querying and produced consistent outputs that could be trusted regardless of who ran the analysis. When Zenith compared Celeste’s analysis with TARS and platform-level Search Query Reports, the findings aligned closely. This gave the team confidence that the insights reflected what was happening across the search program. With Skai providing an independent measurement layer, Zenith could evaluate performance from a unified source of truth instead of relying only on publisher-reported metrics.
Celeste also changed the type of output the team received. Instead of ending with a spreadsheet or report, each query returned optimization recommendations and projected business impact. What once required several steps and hours of work became a decision-ready briefing in a single workflow.
Results
The shift to Celeste AI produced immediate, measurable improvements:
- ~4–5 minute average turnaround per prompt, replacing manual multi-step analysis that previously took hours
- Validated accuracy with figures cross-checked against TARS and platform SQR outputs
- Optimization recommendations with projected impact included in every response, enabling faster decisions
- Scalable without additional resource strain, enabling parallel MCC comparisons that were not previously feasible
The success quickly expanded beyond the initial deployment. Accor teams in multiple regions adopted the same Celeste AI framework, replacing manual MCC analysis with a faster, more consistent approach. What began as a solution for one team became a scalable model for how Zenith’s search operators work, spending less time pulling data and more time acting on insights. In the agentic era of paid search, success belongs to teams that can transform data into action at scale.
About Zenith International
Zenith International is a data-driven media agency within Publicis Media that helps brands plan, buy, and optimize advertising across channels, using data, technology, and analytics to connect businesses with the right audiences.
About Accor
Accor is one of the world’s largest hotel companies, operating thousands of properties across more than 110 countries with brands ranging from luxury to budget.
Solution used
