Telecommunications2022Generative AINLPRecommendation SystemsB2B
BT Business

BT Business's 'Albert' AI Knowledge System Lifts First-Contact Resolution 8% and NPS 5% Across 10,000 Contact Centre Agents

After merging four separate agent knowledge bases following the EE, BT, and PlusNet integration, BT Business deployed an eGain-powered AI guided-help system called Albert — achieving an 8% improvement in first-contact resolution, a 5% NPS increase, and a 50% reduction in agent time to competency.

First Contact Resolution Improvement8%
NPS Improvement5%
Agent Time-to-Competency50% faster
Agents Covered10000 agents
5 min read

Background

The integration of EE, BT, and PlusNet created a fragmented knowledge environment. Four incompatible knowledge bases meant agents had inconsistent access to resolution guidance. With 3.5 million calls per month and 19 million annual callbacks attributable to first-contact failures, the operational and customer experience cost was substantial. A unified, AI-guided system was prioritized as critical infrastructure for the combined organization.

What Was Implemented

  • eGain AI knowledge management platform deployed as "Albert" across BT Business contact centres
  • Shift from reactive search to guided-help model: AI walks agents through step-by-step compliance-ready resolution paths
  • 20,000 articles distilled into AI-process guidance flows by eGain
  • CRM integration: customer and device context surfaced within seconds of call start
  • Call pause-and-resume capability for complex resolutions
  • Department-specific pathways (compliance, elderly customers, digital voice migration)
  • Continuous improvement program with eGain: ongoing conversational guidance refinement, faster resolution paths, further automation

Results

Albert achieved the following verified improvements across 10,000 contact centre agents: - 8% increase in first-contact resolution - 5% increase in Net Promoter Score - 50% reduction in agent time-to-competency - 10% growth across key metrics for the top 10% of users All four metrics are confirmed in the BT Business primary source authored by Stacy Young.

Lessons

  • Consolidating fragmented post-merger knowledge bases into a single AI-guided system is a high-value, near-term integration play in telecommunications M&A
  • Guided-help outperforms search-based knowledge systems in agent compliance and consistency because it removes the discretion-and-error step of selecting the right article
  • Recruiting high-performing agents as knowledge management subject matter experts ensures that resolution pathways reflect real-world performance, not just process documentation
  • Investing in CRM integration so context flows into the knowledge system automatically at call start meaningfully improves agent speed and caller experience

Ready to implement AI in your commerce operations?

McFadyen Digital helps teams move from case study to live implementation.

Talk to an expert →