Customer support · Production

Forethought Solve helps Spordle deflect 21,000 chat inquiries at an 86% self-serve rate

The problem

Spordle's small support team could not handle cyclical peak-season ticket volumes reaching nearly 7,000 per month, forcing seven or eight additional employees to work as full-time support agents on top of their normal responsibilities, often for 14+ hour days. Daily volumes ranged from 350 to 600+ tickets, with customers waiting through a two-week backlog.

Workflow diagram · grounded in source
1
Customer initiates chat
Trigger
Customer support inquiries arrive through the Forethought Solve chat widget.
▾ source quote
“Spordle uses Forethought Solve as its chat widget”
2
AI generates response
Ai action
Generative AI automatically serves up accurate, human-like responses to chat inquiries.
▾ source quote
“Solve uses generative AI to automatically serve up accurate, human-like responses to chat inquiries. Solve's generative AI models are automatically trained on Spordle's data, allowing it to comprehend sentence structure, meaning, and tone to extract the perfect answer from many …”
3
Bilingual language routing
Routing
The chat widget switches seamlessly between providing responses in French and English.
▾ source quote
“the chat widget switches seamlessly between providing responses in French and English”
4
Intent detection via Workflow Builder
Ai action
Workflow Builder uses built-in generative AI to detect customer intent and enable seamless self-service.
▾ source quote
“With Workflow Builder, built-in generative AI enables Spordle's support team to build automated workflows that detect customer intent to enable seamless self-service”
5
Self-service resolution delivered
Output
Customer inquiries are deflected from the ticket queue through automated self-service.
▾ source quote
“Spordle has deflected 21,000 chat inquiries, with an 86% self-serve rate”
Reported outcome

Since March 1, 2023, Spordle deflected 21,000 chat inquiries at an 86% self-serve rate.
Over 600 tickets were instantly resolved within the first week after implementation, and three months post-implementation the ROI reached 142%. Agents are no longer exhausted from dealing with heavy ticket volumes.

Reported metrics
chat deflections since March 202321,000
Self-serve rate86%
ROI at 3 months post-implementation142%
Tickets instantly resolved in first week600+
Show all 9 reported metrics
chat deflections since March 202321,000
self-serve rate86%
ROI at 3 months post-implementation142%
tickets instantly resolved in first week600+
peak monthly ticket volume (pre-implementation)nearly 7000
daily ticket volume range (pre-implementation)between 350 and 600+ tickets per day
overflow employees working as agents (pre-implementation)seven or eight
daily hours worked during peak overflow14+ hour days
ticket backlog duration (pre-implementation)two-week backlog
Reported stack
Forethought SolveWorkflow BuilderForethought
◆ Does this fit your context?

Compare to your context

Tell us your scale, team, and constraints. We'll show what changes at your size, what fails at your scale, and whether this case is a fit, needs adaptation, or won't scale to you. Free demo, no signup.

Compare to your context →
~30 seconds · free
Source
https://forethought.ai/case-studies/spordle
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Since March 1, 2023, Spordle deflected 21,000 chat inquiries at an 86% self-serve rate.

What tools did this team use?

Forethought Solve, Workflow Builder, Forethought.

What results were reported?

chat deflections since March 2023: 21,000; Self-serve rate: 86%; ROI at 3 months post-implementation: 142%; Tickets instantly resolved in first week: 600+ (source-reported, not independently verified).

How is this customer support AI workflow structured?

Customer initiates chat → AI generates response → Bilingual language routing → Intent detection via Workflow Builder → Self-service resolution delivered.

Related customer support cases

WHAT TO DO WITH THIS

Now compare it to your context

This case is one data point. Whether its pattern fits you depends on your volumes, your stack, and your exception load — that comparison is the step no case study can do for you.

Browse the Customer support cases →