Case study: Vico Homes
AI Summarisation
Improving Call Wrap Up with Agentic AI
Case study: Vico Homes
AI Summarisation
Improving Call Wrap Up with Agentic AI
Vico Homes (formerly known as Wakefield and District Housing) was established in 2005 and is one of the UK’s largest social housing providers, with over 32,000 homes and 65,000 tenants across the Wakefield district and the North of England. To support its strategic aim to improve agent efficiency, Vico Homes were looking to roll out Amazon Contact Lens and Agentforce to generate call insight and automatically summarise the call to improve agent efficiencies and reduce call wrap up times.
Challenge
Within the existing contact centre solution, several issues were evident, these included:
Limited insight into call content and outcomes
There was a general lack of visibility into what was actually discussed, agreed or promised during calls, beyond direct access to the call recording.
No consistent call summary or record of decisions
Agents manually entered notes post-call, leading to incomplete, inconsistent, or missing summaries.
Difficulty understanding caller sentiment and vulnerability
There was no automated way to assess tenant frustration, distress, or vulnerability, increasing the risk of unmet safeguarding or support needs.
No objective measurement of agent behaviour
Supervisors struggled to understand communication quality without listening to full recordings.
Manual and error-prone call classification
Agents needed to select call reason codes manually, which were often inaccurate, rushed, or inconsistent.
Inability to flag priority or high-risk calls automatically
Calls involving complaints, safeguarding, arrears escalation, or repeat contact were not reliably flagged to supervisors in real time.
Extended and inefficient wrap-up times
After-call work increased handling time, reducing agent availability and increasing queue lengths.
Inconsistent compliance and audit trails
Manual notes made it difficult to evidence regulatory compliance or respond to complaints and disputes.
What we did
We initially worked with Vico Homes on the refinement of their user stories and provided a solution design which explained how the requirements would be met. Following sign-off of the design we then built out the required functionality using a combination of Amazon Connect Contact Lens, Service Cloud Voice and Agentforce over a series of two-week delivery sprints. After each sprint a Show & Tell session was held that demonstrated the capability and provided a mechanism to incorporate feedback.
Following build, the solution underwent system testing and then training was provided on using the solution. UAT support was then provided to ensure the system fully met internal requirements. Following successful completion of UAT, the solution was promoted to the live environment and Alscient then provided a week of hypercare.
Results
The project was delivered on time and to budget with the following business benefits being realised:
Consistent, auditable call summaries generated automatically
Gen AI summaries were created providing a standardised record of call intent, outcomes, and commitments without reliance on manual agent note taking.
Improved visibility into call efficiency and handling quality
Capture of call duration, agent talk-time, customer talk-time, and non-talk time enabled objective analysis of call flow and agent effectiveness.
Quantifiable insight into caller and agent sentiment
Automatic sentiment scoring allowed Vico Homes to objectively identify caller (and agent) frustration.
Reduced after-call work and faster wrap-up times
Automated summaries significantly reduce manual wrap-up, improving agent availability.
Accurate, automated call categorisation
Contact Lens detects matched categories for call insight, removing subjective and inconsistent manual call classification.
Improved identification of high-risk or priority interactions
Sentiment, matched keywords and category matches enable objective detection of complaints, escalation risk and safeguarding concerns.
Real-time call flagging
Supervisors could live monitor calls and assist agents directly on the call.
Improved quality assurance and coaching outcomes
Supervisors could focus on flagged, high-impact calls rather than random sampling, improving coaching efficiency and effectiveness.
Stronger compliance and dispute resolution capability
Structured summaries, sentiment data and direct recording access provided a clear, defensible audit trail for complaints and regulatory reviews.
Actionable insights directly embedded in Salesforce
All summaries, metrics, flags and recordings were surfaced directly within Salesforce enabling data-driven operational decisions to be made.
Expected Benefits
Metrics suggest that a customer services agent saves 1–3 minutes of wrap-up time per call. For an agent handling 800–1,200 calls per month, this equates to 15–40 hours saved per agent per month. Additional expected improvements include 10–25% lower after-call work, 5–15% higher agent productivity, improved note consistency and accuracy, faster case resolution, and reduced agent fatigue. Secondary benefits often include better compliance, quicker supervisor reviews due to more complete and timely records.
Next case study






