Customer support · Production

YAZIO deflects 80% of support tickets without lowering CSAT in 6 months with Forethought Solve

The problem

YAZIO's team of 5-8 agents could not keep up with rapidly increasing ticket volume while also managing help center content, templates, and internal knowledge bases; tickets went unanswered for too long and the team could not hire fast enough.

First attempt

YAZIO had minimal success with an internal AI search tool, and other vendors they evaluated offered only rule-based chatbots built on large decision trees requiring regular maintenance—not true agentic AI.

Workflow diagram · grounded in source
1
Customer submits support request
Trigger
Customers submit support requests through Solve's chatbot interface.
▾ source quote
“YAZIO's users got more comfortable with the chatbot, which was easier to navigate than keywords in a help center search bar”
2
Sensitive data redacted
Validation
Automation redacts sensitive data elements from customer interactions before processing.
▾ source quote
“Forethought met their security requirements by using automation to redact sensitive data elements”
3
Agentic AI processes ticket
Ai action
Agentic AI processes tickets without decision trees, mimicking the experience of a human agent.
▾ source quote
“It was important for YAZIO's team to find a solution that used agentic AI. This would remove the need for decision trees and mimic the type of experience a customer would get with a human agent”
4
Complex tickets routed to agent
Routing
Complex tickets involving third-party connectivity issues or contract cancellations are passed directly to a human agent.
▾ source quote
“passes on complex tickets involving third-party connectivity issues or contract cancellations directly to an agent while solving more simple problems on its own”
5
Resolution delivered in multiple languages
Output
Solve addresses customers using sensitive language that meets YAZIO's expectations across supported languages.
▾ source quote
“Most users are based in Germany but Solve also needs to work with users in English, Portuguese, Italian, French, and Spanish. Today, YAZIO is confident that Solve can address their customers using sensitive language that meets their expectations in these …”
Reported outcome

Solve deflected 80% of YAZIO's tickets in six months while maintaining a 4.0/5.0 CSAT, despite a 40% growth in ticket volume; the in-house team still manages the same workload as before without adding headcount.

Reported metrics
Ticket deflection rate at 6 months80%
CSAT score at 6 months4.0/5.0
Ticket deflection rate at 1 month60%
CSAT score at 1 month3.1/5.0
Show all 10 reported metrics
ticket deflection rate at 6 months80%
CSAT score at 6 months4.0/5.0
ticket deflection rate at 1 month60%
CSAT score at 1 month3.1/5.0
ticket volume growth40%
full-time agents needed to replace Solvethree full-time agents
in-house CX team size5-8
target deflection rate (next 6 months)90%
user base growth without team expansiongrown its user base massively without expanding its team
team workload reliefmassive weight lift from their to-do list
Reported stack
ForethoughtSolveDiscover
◆ 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/yazio-deflects-80-percent-forethought
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Solve deflected 80% of YAZIO's tickets in six months while maintaining a 4.0/5.0 CSAT, despite a 40% growth in ticket volume; the in-house team still manages the same workload as before without adding headcount.

What tools did this team use?

Forethought, Solve, Discover.

What results were reported?

Ticket deflection rate at 6 months: 80%; CSAT score at 6 months: 4.0/5.0; Ticket deflection rate at 1 month: 60%; CSAT score at 1 month: 3.1/5.0 (source-reported, not independently verified).

What failed first in this deployment?

YAZIO had minimal success with an internal AI search tool, and other vendors they evaluated offered only rule-based chatbots built on large decision trees requiring regular maintenance—not true agentic AI.

How is this customer support AI workflow structured?

Customer submits support request → Sensitive data redacted → Agentic AI processes ticket → Complex tickets routed to agent → Resolution delivered in multiple languages.

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 →