Salesforce's Data 360 team built the data foundation behind the Informatica Help Agent in 24 days, ahead of a 30-day deadline, converting 100,000 unstructured documents into a searchable knowledge base. After launch, the agent achieved an 80% resolution rate with only 5% human escalation.
Irina Malkova, Vice President of Product and Success Data, and Alexander Smith wrote: "The Informatica Help Agent achieved an 80% resolution rate with only 5% human escalation."
The team was challenged to enable the Informatica agent in 30 days after the acquisition completed on November 18, 2025. Informatica's documentation was written for human readers rather than AI: raw HTML files contained headers, footers, and navigation menus that interfere with retrieval quality, and manual cleaning was impossible at this scale. The knowledge base also had complex versioning, with many near-duplicate articles differing only slightly across product versions. The team managed product versioning through prompting and configuration rather than changing system logic, and delivered a production-grade foundation instead of addressing every complex detail in the initial release.
Data 360 is how the team unifies, standardizes, indexes, and activates unstructured knowledge. Content hosted on Informatica's website through a content management system was ingested with the new Data 360 "sitemaps" feature, which crawls the website and creates conforming Data 360 knowledge. For more unique content, Python workflows managed the extraction while Data 360 handled ingestion and storage. The first ingestion of developer documentation finished in about three hours, and later updates ran faster as the pipelines stabilized. Data 360 patterns normalized the content and removed noise while keeping the original meaning, turning HTML into consistent chunks for embedding and retrieval. Without Data 360, preparing this volume of content would have taken weeks; the team finished data preparation in days.
The documents came from different systems with unique structures and metadata. The team managed limitations in filtering and refresh timing through preprocessing and configuration, and Data 360 pipelines supported hundreds of thousands of documents. For retrieval, the team reused chunking strategies proven for Customer Success and added filters and metadata tags during ingestion to narrow results to the most relevant context.
Rather than rebuilding, the team extended established configurations for ingestion, chunking, indexing, and retrieval to Informatica content. Adjustments stayed localized because the pipelines and infrastructure follow a standard design. A rebuild would have required a much larger team and months of extra work. The team supports all agents that augment the Customer Success business motion, including those on help.salesforce.com and slack.com/help, balancing tailored answers for each agent with a durable data foundation that can power future agents.