Case Study
Knowledge Base for Consumer Lending Customer Service
Softwise.AI built a comprehensive knowledge management solution for Bocian Pożyczki - one of Poland's leading non-bank consumer lenders - designed to empower the customer service team with efficient knowledge access and organization. The platform combined a search-centric user interface with an intelligent hybrid search engine, giving consultants a single, fast way to find the right answer across the company's full knowledge base.
CLIENT
Bocian Pożyczki
YEAR
2022
INDUSTIRES
Financial Services - Consumer Lending (Customer Service)
DELIVERABLES
End-to-End Knowledge Management Platform, Search-First Interface, Intelligent Hybrid Search Engine
The problem
Bocian Pożyczki's customer service team handled a broad range of questions about loan products, terms, fees, repayment options, and policy details - much of it under regulatory and procedural rules that updated frequently. Consultants needed fast, reliable access to current answers during live customer interactions, but the existing setup forced them to search across multiple sources with tools that did not match how they actually asked questions.
{Key Challenges}
Why traditional keyword search wasn't enough for consumer-lending consultants?
Wide product and policy surface.
A single consultant had to be ready to answer questions across the entire loan range and the rules governing each product.
Keyword search broke on natural-language queries.
Consultants did not ask in keywords; they asked in full questions, often in customers' own wording.
Live-call time pressure.
Every second spent navigating documents was a second the customer waited - a customer-experience cost the team felt directly.
Scattered, frequently updated sources.
Product specs, policy documents, and procedural notes lived in different formats with different refresh cycles.
{Process}
How we built a search-first knowledge platform in 4 stages?
1. Workflow discovery with the customer service team
Softwise.AI mapped the real consultant workflow - which questions came up most often, where time was actually being lost, and what an acceptable answer looked like during a live call.
2. Source aggregation and normalization
The team consolidated knowledge sources into a single, structured estate - making everything searchable from one place without forcing manual content rework.
3. Hybrid search engine design
A search engine was built combining multiple retrieval techniques - so both natural-language questions and exact-term lookups returned relevant, ranked results.
4. Search-first interface rollout
The platform was wrapped in a search-centric UI built around the consultant's actual moment of need - one search bar, fast results, links back to the source document on every answer.
Solution
End-to-end knowledge management platform
A single environment for organizing, updating, and retrieving customer-service knowledge - replacing scattered tools with one coherent surface.
Search-first interface
A search-centric UI designed for the speed of a live customer interaction - minimal navigation, immediate results, frictionless source verification.
Intelligent hybrid search engine
An advanced search engine combining multiple techniques to handle both natural-language questions and exact-term lookups - matching how consultants and customers actually phrase things.
{Results}
A single search interface replacing manual navigation across multiple sources.
Hybrid retrieval returning relevant results for both natural-language questions and exact-term lookups.
Source-linked answers so consultants could verify every piece of information they passed to a customer.
An organized knowledge base that the team could maintain and update without engineering involvement.
{Business impact}
Faster customer answers. Less time hunting through documents meant shorter live interactions and lower customer wait times.
More confident consultants. Source-linked results reduced the uncertainty that drives escalation to senior staff.
A scalable customer-service operation. More queries handled per consultant, on the same headcount.
An always-current knowledge base. Centralized maintenance kept consultants from advising on stale information.
{Key recommendations}
Match the search to how users actually ask. Hybrid retrieval beats keyword-only search wherever questions are phrased in natural language.
Design the UI around the moment of need. A search-first interface earns its keep on every live call.
Aggregate sources before adding intelligence. A unified knowledge estate is the foundation; search and AI sit on top of it.
Always return the source. In regulated lending, verifiability is what makes a knowledge tool actually usable.
SUMMARY
Softwise.AI delivered an end-to-end knowledge management platform for Bocian Pożyczki's customer service team, combining a search-first interface with an intelligent hybrid search engine - giving consultants fast, reliable, source-linked answers during live customer interactions and turning a fragmented knowledge estate into a single, scalable operational asset.