Customer support · Production

Babbel's Bab the Bot delivers instant answers to 50% of customer queries via AI-powered chatbot

The problem

As Babbel's user base grew, both the volume and complexity of customer service queries increased, threatening response times and quality standards. Earlier chatbot explorations had failed to yield a satisfactory solution.

First attempt

Prior chatbot solutions explored by Babbel fell short because they merely regurgitated Help Centre articles rather than providing genuinely functional and pleasant assistance.

Workflow diagram · grounded in source
1
Learner query received
trigger
“if you're studying late in the evening and encounter an issue or have a question, you can get it resolved on the spot”
2
Intent understanding
ai_action
“The process starts with understanding what the learner is asking for”
3
Information sufficiency check
validation
“determining if we need more information”
4
Automated response or action
output
“providing either the correct information or an automated solution”
5
Human agent handoff
routing
“knowing when to hand-off one of our CS agents”
6
Continuous improvement loop
feedback_loop
“Each conversation creates a feedback loop that makes Bab more and more helpful, resulting in an ever-improving experience for our users”
Reported outcome

Since its 2024 launch, Bab has facilitated 250,000 conversations, delivers instant answers to 50% of all queries, and the rollout to iOS and Android increased monthly conversations by over 50%.

Reported metrics
Total conversations facilitated250,000
Monthly conversations increase after app integrationover 50%
Queries receiving instant answers50%
Reported stack
Bab
Source
https://www.babbel.com/en/magazine/meet-bab-your-customer-service-chatbot
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Since its 2024 launch, Bab has facilitated 250,000 conversations, delivers instant answers to 50% of all queries, and the rollout to iOS and Android increased monthly conversations by over 50%.

What tools did this team use?

Bab.

What results were reported?

Total conversations facilitated: 250,000; Monthly conversations increase after app integration: over 50%; Queries receiving instant answers: 50% (source-reported, not independently verified).

What failed first in this deployment?

Prior chatbot solutions explored by Babbel fell short because they merely regurgitated Help Centre articles rather than providing genuinely functional and pleasant assistance.

How is this customer support AI workflow structured?

Learner query received → Intent understanding → Information sufficiency check → Automated response or action → Human agent handoff → Continuous improvement loop.

Related customer support cases