Back office ops · Production

Funding Societies democratizes AI across 8 departments with Activepieces, saving nearly a quarter of a year in collective manual hours

The problem

Funding Societies needed to scale AI and automation adoption across all business departments while meeting stringent compliance and data sovereignty requirements across five countries, but their traditional approach of building custom AI systems with dedicated engineers created bottlenecks that prevented non-engineering teams from participating.

First attempt

The previous external training approach for automation tools failed because outside trainers lacked the internal company context needed to make support sessions effective.

Workflow diagram · grounded in source
1
Payment creation triggers review
trigger
“The workflow triggers when a new payment is created”
2
OCR key data extraction
ai_action
“An intelligent document review system combining OCR and LLM technologies. The workflow triggers when a new payment is created, extracts key data using OCR”
3
Authenticity and fuzzy-match validation
validation
“validates authenticity (official logos, invoice structures), performs fuzzy matching against expected payment data”
4
Manual fallback for edge cases
human_review
“provides real-time feedback with manual fallback for edge cases”
5
RAG support request routing
ai_action
“A Retrieval-Augmented Generation system that automatically routes support requests. All past production support requests are stored in a vector database. When new requests arrive, the system analyzes them, compares to historical patterns…”
6
Automated learning from resolved tickets
feedback_loop
“The system learns: accuracy improves through automated learning from resolved tickets”
7
AI-driven SEO content generation
ai_action
“generate new content aligned with financial content guidelines, and implement strategic FAQ formatting for AI search optimization”
8
Customer insights report output
output
“compiling insights into comprehensive product development reports”
Reported outcome

Funding Societies deployed 100+ workflows in production across eight departments, saving nearly a quarter of a year in collective manual task hours, with content production time reduced by 85% and organic traffic up 20-30%.

Reported metrics
Workflows in production100+
Collective manual task hours savednearly a quarter of a year in collective manual task hours
Content production time reduction85%
Organic traffic increase20-30%
Show all 7 reported metrics
workflows in production100+
collective manual task hours savednearly a quarter of a year in collective manual task hours
content production time reduction85%
organic traffic increase20-30%
content production time after automationunder 30 minutes
document review hours eliminatedthousands of manual review hours yearly
customer insights generation timeinstantaneous
Reported stack
ActivepiecesOCRLLMRAGvector databaseNLPIntercom
Source
https://www.activepieces.com/customers/funding-societies
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Funding Societies deployed 100+ workflows in production across eight departments, saving nearly a quarter of a year in collective manual task hours, with content production time reduced by 85% and organic traffic up 2…

What tools did this team use?

Activepieces, OCR, LLM, RAG, vector database, NLP, Intercom.

What results were reported?

Workflows in production: 100+; Collective manual task hours saved: nearly a quarter of a year in collective manual task hours; Content production time reduction: 85%; Organic traffic increase: 20-30% (source-reported, not independently verified).

What failed first in this deployment?

The previous external training approach for automation tools failed because outside trainers lacked the internal company context needed to make support sessions effective.

How is this back office ops AI workflow structured?

Payment creation triggers review → OCR key data extraction → Authenticity and fuzzy-match validation → Manual fallback for edge cases → RAG support request routing → Automated learning from resolved tickets → AI-driven SEO content generation → Customer insights report output.

Related back office ops cases