project showcase
SecuRita ChatBot
Advanced conversational interface to optimize business processes in the CX domain, leading to a notable reduction in customer support initiatives.
View Prototype →
Overwiew
TimeLine
12 Months
Disciplines
Conversational interface,
UX Design,
Voiceflow
My Role
UX Research & Analysis
Chatbot Development
User Testing
Team Size
4
A little about the company...
Securitas Technology delivers advanced security solutions for commercial and residential spaces, including access control, remote monitoring, and intrusion systems. Their flagship app, HQ, provides users with real-time monitoring, instant alert responses, and seamless management of security systems, ensuring effective and proactive security oversight.
The story..
During the Covid-19 pandemic, we experienced a drastic 600% increase in customer service calls. This surge led to longer wait times, more abandoned calls, and added stress on our support team. To tackle this challenge, we introduced a conversational user interface bot. Designed with our customers in mind, this bot efficiently handles inquiries, captures user needs, and helps us maintain strong customer relationship, even during peak times.
600%
Increase in number of calls
20 mins
Wait Time
20%
Abandoned calls
Problem Statement
“How might we enable Securitas Technology to envision a conversational user interface for their security management system through the HQ web application”
Key Features
While harmonizing transparency and humanized conversations
Cultivating Trust Through Knowledge-Driven Conversations
Instead of enduring lengthy waits on the 1-800 number
Process
HQ application: HQ is a security management system that enables users to monitor multiple locations and perform tasks such as testing systems, setting up alerts, and assigning access permissions to buildings.
Target end users: Security Analyst, Building Managers
The top competitor for Securitas is ADT which uses a chatbot Olivia
Impact of ADT Olivia
+7.2k
Monthly Conversations
97%
Conversations end in resolution
24/7
Support Available
• Understanding key tasks of users using HQ Application
• Identifying the pain points of users when they call customer care
• Identifying users familiarity and comfort with conversational interfaces
• Preferences for the type of conversational interface: chatbot vs voice assistants.
"
I would rather use a chat on my browser window than keep a call waiting on my phone.
I would rather use a chat on my browser window than keep a call waiting on my phone.
We created a persona based on the gathered information to guide the process moving forward.
• Putting the system on test
• Service Dispatch
Call rails additionally gave us insights on the possibility of automating the responses as they had a fixed sequence of steps to follow to resolve the issues.
Using VoiceFlow, A chatbot was built based on the insights from Call Rails s into some commonly used phrases to train the bot.
Research Goals:
1. How the participants perceive the language of the bot?
2. Is the conversation flow as expected?
Users were tested on 3 task scenarios:
• Putting their system in test
• Checking status of a ticket
• Creating a new ticket
We used the atomic We used the atomic UX Research canvas to consolidate our findings. to consolidate our findings.
Need for context
Users need to be logged in for the chat flow to work. This aids in auto suggesting inputs and traceability of different actions performed in the system.
Natural language & Intent recognition
More humane and natural language needed to make the user more comfortable while using a chatbot while allowing for better input recognition.
Affirmations and Confirmations
Appropriate formatting for confirmations based on the provided input improving the users trust on the AI system.
Feedback Loop
Handling errors by redirecting to the right department of customer care and taking in negative feedback to improve on bot flow.
