Case Study · AI Implementations

BoxGPT Asset Search

Enterprise content · GPT search · Summaries · Metadata tagging · Scoped access controls

The challenge

Content assets were technically stored, but practically hidden.

Slow file search

Teams spent 10-15 minutes per request navigating folders to find the right document.

Content duplication

Poor version tracking created confusion, repeated work and inconsistent asset reuse.

Scattered assets

Low discoverability slowed campaigns and reduced reuse of high-performing material.

Manual sorting

Teams wasted hours each week in repetitive tagging, folder cleanup and asset triage.

What we built

A governed GPT layer over enterprise files.

Instant file search

Users find documents in seconds using natural-language intent instead of folder paths.

AI summaries

GPT summaries help users understand content, differences and likely fit without opening every file.

Smart tagging

Metadata tagging and deduplication improve discovery and reduce duplicated work.

Scoped access

Search respects existing access controls so AI retrieval does not bypass permission boundaries.

Results

Creative operations moved faster.

98.9%

Reduction in file search time.

90%

Less manual sorting effort.

+40%

Content reuse through better discovery.

3+ days

Removed from recurring campaign cycles.

← All case studies  ·  Talk to our team →