Customer support · Production

Lime achieves 77% reduction in response time with Forethought AI triage and automated support

The problem

Lime's support center operated entirely manually with no ticket routing or prioritization, forcing all agents to handle every ticket type regardless of urgency. Multilingual tickets required agents to manually toggle between Google Translate and their cases, there was no self-service channel, and critical tickets such as accident reports were handled in the order received rather than by severity.

Workflow diagram · grounded in source

Support ticket arrives

trigger

“Support inquiries were received through email, phone, the Lime app, and a web form to be followed up on by support agents”

Triage classifies language and category

ai_action

“Triage layers the language and category classifications, and works internally to determine different service levels based on severity”

Route to agent queue

routing

“Lime has custom tags and triggers set up so that when Forethought predicts a specific language and category, the case is routed to a specific queue based on those tags. Now, support cases are tagged so the most critical inquiries are han…”

Solve interprets and responds

ai_action

“they're able to interpret the intent of a customer's request, through their chat widget or email, and instantly search their entire database of knowledge articles and previously resolved tickets to provide the most accurate response–all …”

RPA workflow executes

integration

“These RPA-guided workflows are automated from beginning to end through pre-set rules. These workflows integrate with internal tools to complete automation tasks around damaged vehicles, charge and payment issues, and receipt questions”

Complex issues escalate to agent

human_review

“If the customers' question is complex, they're automatically routed to the help center web form, which is then passed on to a support agent”

Reported outcome

Forethought automated 27% of email and web cases and tagged 98% of support tickets automatically out of more than 1.7 million tickets per year, yielding a 77% reduction in time to first response along with significant cost savings and improved customer satisfaction.

Reported metrics

Cases automated via email and web27%

Language and category tags predictedover 2.5 million

Support tickets tagged automatically98%

Time to first response77% reduction

Show all 8 reported metrics

cases automated via email and web27%

language and category tags predictedover 2.5 million

support tickets tagged automatically98%

time to first response77% reduction

annual support ticket volumemore than 1.7 million

cost savingssignificant cost savings

customer satisfactionimproving customer satisfaction

agents' time savedhelps save agents' time in a massive way

Reported stack

Forethought TriageForethought SolveWorkflow BuilderRobotic Process AutomationGoogle Translate

Source

https://forethought.ai/case-studies/lime-case-study

Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Forethought automated 27% of email and web cases and tagged 98% of support tickets automatically out of more than 1.7 million tickets per year, yielding a 77% reduction in time to first response along with significant…

What tools did this team use?

Forethought Triage, Forethought Solve, Workflow Builder, Robotic Process Automation, Google Translate.

What results were reported?

Cases automated via email and web: 27%; Language and category tags predicted: over 2.5 million; Support tickets tagged automatically: 98%; Time to first response: 77% reduction (source-reported, not independently verified).

How is this customer support AI workflow structured?

Support ticket arrives → Triage classifies language and category → Route to agent queue → Solve interprets and responds → RPA workflow executes → Complex issues escalate to agent.