What challenges did ImagineArt face before AI implementation?

Creative production was fragmented, slow, and expensive. Traditional planning, human modeling, and physical shoots required weeks of coordination across multiple teams. The company needed a unified stack that could support realistic, high-fidelity creative workflows at scale.

What was the impact of the AI solution?

Compressed campaign launch cycles from weeks to days. Attributed over 50% of API costs to Google models due to user preference and output quality. Eliminated external third-party audio providers through native audio generation. Enabled high-fidelity campaigns without traditional overhead.

Source

GCPScaled productionEvidence: Medium65/100

ImagineArt builds agentic generative creative workflows with Gemini Enterprise Agent Platform on Google Cloud

Use case typeWorkflow automationUpdated Jun 13, 2026

ImagineArt consolidated global ad and analytics data in BigQuery and used Google Cloud models to support a high-fidelity creative production workflow. The company integrated Gemini in its Chatly platform, added specialized generative models such as Veo and Nano Banana, and moved toward agentic workflows through Imagine Assist and Omni Agent. By combining these services with real-time tracking in Data Studio, ImagineArt replaced manual planning, modeling, and shoots with a more automated creative operating model.

Organization: ImagineArt
Industry: Consumer & Food
Location: United States
Published: June 2026

Reported outcomes

−50%

campaign launch cycle durationTime & speed

+50%API cost share attributed to Google models

Strategic outcomes

Speed & agilityCompressed creative production from weeks to daysOther strategic outcomeUnified creative workflows into a single Google Cloud stackOther strategic outcomeEnabled high-fidelity, production-ready creative generation

Why do we believe this?Outcome claims, sources, and evidence checks

Normalized claim

Campaign launch cycle duration: 50% decrease

Google Cloud Customer StoryJun 13, 2026Customer storyInferred claimMedium evidence strength

Reduces campaign launch cycles from weeks to just days

Normalized claim

API cost share attributed to Google models: 50% increase

Google Cloud Customer StoryJun 13, 2026Customer storyExplicit claimMedium evidence strength

over 50% of ImagineArt’s API costs are attributed to Google models

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: ImagineArt
Provider: GCP
Maturity: Scaled Production
Linked source: Google Cloud Customer Story

The company needed a unified stack that could support realistic, high-fidelity creative workflows at scale

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 1 of 1

1Workflow automation

Creative production was fragmented, slow, and expensive.
Traditional planning, human modeling, and physical shoots required weeks of coordination across multiple teams.
The company needed a unified stack that could support realistic, high-fidelity creative workflows at scale.

Integrated Gemini for text-based productivity in Chatly.
Used BigQuery to consolidate disparate global ad and analytics data.
Used Data Studio for user trend and ROI visualization.
Built agentic workflows through Imagine Assist and Omni Agent using Gemini as the reasoning engine.
Leveraged Veo 3 and Nano Banana for video and image generation, including native audio pipelines.

Technologies

Gemini Enterprise Agent Platform BigQuery Data Studio Veo Nano Banana Gemini

Compressed campaign launch cycles from weeks to days.
Attributed over 50% of API costs to Google models due to user preference and output quality.
Eliminated external third-party audio providers through native audio generation.
Enabled high-fidelity campaigns without traditional overhead.

Architecture

ImagineArt built a unified creative AI stack on Google Cloud, combining Gemini for text productivity and reasoning, BigQuery for data consolidation, Data Studio for performance visibility, and specialized generation models such as Veo and Nano Banana. It then organized these capabilities into agentic workflows, including Imagine Assist and Omni Agent, and used model-native audio generation to remove external production steps.

Sources & evidence1

Evidence: Medium65/100Evidence strength