Internal Reference · CS Websites & AI Team

🤖 Kodee AI Performance Metrics

Can read and act on deflection rate, CSAT delta, resolution time, and hallucination rate to evaluate AI impact.

90%

Customer support interactions
handled by Kodee (Q2 2026)

500+

Admin-level tasks Kodee
can perform as an AI agent

~€14M

Estimated annual savings
from Kodee's CS automation

50%→90%

CS interaction coverage growth
from start of 2025 to Q2 2026

📊 AI Impact — Core Metrics

The 4 primary metrics for evaluating Kodee's AI impact on customer support quality and efficiency.

🛡️
Metric 01

Deflection Rate

% of conversations fully resolved by AI (Kodee) without being handed off to a human specialist.

First Contact Rate (Chatbot) — % of chatbot conversations not handed off to a specialist.

Tableau — CS KPIs
Tableau — CS Chatbot Results

FCR > 80%

Chatbot SLA: ≤ 30 sec first reply

💬 Lumos Prompt
Object=conversations, Granularity=weekly, Time range=[last 4 weeks], Filters=chatbot, Metric=% of conversations not handed off to specialist (First Contact Rate). Show trend over time.
📈 Open CS Chatbot Dashboard
⭐
Metric 02

CSAT Delta (AI vs Human)

Difference in CSAT scores between AI-resolved conversations and human-resolved conversations — measures AI satisfaction gap.

Happiness Score split by Handler Type (Chatbot vs Specialist). Rated 4 or 5 out of 5 per conversation.

Tableau — CS KPIs (Total-Split view)
Tableau — Customer Satisfaction

CSAT ≥ 93%

Delta goal: AI CSAT within 5 pts of specialist CSAT

💬 Lumos Prompt
Object=conversations, Granularity=monthly, Time range=[last 3 months], Filters=chatbot AND specialist, Metric=CSAT (happiness score ≥4). Compare chatbot CSAT vs specialist CSAT side by side.
📈 Open CS KPIs (Split View)
⚡
Metric 03

Resolution Time

Time from conversation assignment to last reply. Measures how fast AI resolves issues vs human agents.

Average Handling Time (HT) — tracked separately for chatbot and specialist in KodeeDesk / Intercom.

Tableau — CS KPIs
Tableau — CS Chatbot Results (Response Time page)
New Relic — AI Chatbot APM

HT < 25 min (specialist)

Chatbot first reply SLA: ≤ 30 sec
Specialist first reply: < 2 min

💬 Lumos Prompt
Object=conversations, Granularity=weekly, Time range=[last 4 weeks], Metric=median handling time. Split by handler type (chatbot vs specialist). Show trend.
📈 Open Chatbot Response Time
🔬
Metric 04 · Proxy

Hallucination Rate

Rate at which AI generates incorrect, fabricated, or misleading responses. No direct measure exists — tracked via proxy signals.

Crash Count + Function Error Count + Negative CSAT (≤3) + Bot Sentiment Score drops. Combined signal for AI answer quality.

Tableau — CS Chatbot Results
New Relic — AI Chatbot Dashboard
BigQuery — ai-chatbots DB

Crash Count → minimize

Function Error Count → trend ↓
Negative CSAT (≤3) → trend ↓
Bot Sentiment Score → trend ↑

💬 Lumos Prompt
Object=chatbot conversations, Granularity=weekly, Time range=[last 4 weeks], Metric=crash count AND function error count AND negative rating count (≤3). Show combined trend. Flag weeks with spikes.
📈 Open AI Monitoring Dashboard
⚠️ Proxy Metrics Note
⚠️

Hallucination Rate is a Proxy — Not a Direct Measure

Hostinger does not currently have a direct hallucination detection pipeline for Kodee. There is no automated system that flags incorrect AI answers in real time. Instead, hallucination risk is inferred from a combination of observable failure signals:

  • Crash Count — chatbot fails to answer at all and forces a handoff (tracked in CS Chatbot Results dashboard)
  • Function Error Count — conversations where a function call failed or returned an error (tracked in CS Chatbot Results)
  • Negative CSAT (≤3/5) — customer dissatisfaction after a bot-only conversation (Tableau + BigQuery)
  • Bot Sentiment Score drop — customer sentiment worsens mid-conversation, suggesting a bad or confusing AI response (Tableau — CS Chatbot Results)
  • Handoff after short conversation — customer escalates quickly, possibly due to an unhelpful or wrong answer (Handoffed Customer Message Distribution)

⚡ Action: Use these proxies together as a composite signal. A spike in any two simultaneously is a strong indicator of a hallucination-related quality issue worth investigating in New Relic or BigQuery.

💬 CS Chatbot (Kodee) — Core Metrics

Tracked via the CS Chatbot Dashboard. These metrics give a high-level view of Kodee's performance in customer support conversations.

🔢

Conversation Count

Number or percentage of conversations by type: chatbot (started & concluded by bot), not_empty (both specialist & client replied), reopened, and empty.

📊 CS Chatbot Dashboard
🔄

Handoff Rate

Number of conversations handed off to human specialists. A handoff is detected when both a specialist and a bot participate in the same conversation.

📊 CS Chatbot Dashboard
✅

First Contact Resolution (FCR)

For Kodee: % of conversations not handed off to specialists. For specialists: % handled by one specialist without reopening. SLA for chatbot = 30 seconds.

📊 CS KPIs Dashboard
⭐

CSAT (Happiness Score)

Percentage of conversations rated 4 or 5 out of 5 by customers. Split by handler type: Chatbot vs. Specialist.

📊 CS Chatbot Dashboard
😊

Bot Sentiment Score

Customer sentiment during chats with Kodee, scored 1–5. Includes Sentiment Score Distributions (start, mid, end) and Changes in Sentiment throughout the conversation.

📊 CS Chatbot Dashboard
💬

Bot Message Average Count

Average number of messages sent by Kodee per conversation. Helps assess conversation depth and efficiency.

📊 CS Chatbot Dashboard
⚠️

Function Error Count

Number of conversations where Kodee encountered a function error — e.g., a failed MCP tool call or API error during task execution.

📊 CS Chatbot Dashboard
💥

Crash Count

Conversations where Kodee could not answer at all and automatically handed off to a specialist. Distinct from function errors — this is a full failure to respond.

📊 CS Chatbot Dashboard
🛍️ Product Chatbots — Metrics by Bot

Kodee powers multiple specialized chatbots across Hostinger's products. Each has its own metrics focus.

🛒 Sales Assistant (hWebsites)

Deployed on hostinger.com, /pricing, and other landing pages. Helps new clients choose the right plan.

  • Conversion Rate
  • Generated Cart Link Click Rate
  • Total Sales
  • Engagement Rate

Tracked via Amplitude

🌐 Website Builder Assistant

Deployed in Website Builder Edit Mode. Helps users navigate features and find functionalities.

  • User Engagement
  • Feature Utilization (Blog, Heatmap, Image, Logo, Page Generator, Writer)
  • CSAT

Tracked via CS Chatbot Dashboard

🖥️ VPS / Linux Assistant

Kodee as a Virtual SysAdmin — assists with server health, firewall rules, SSH keys, malware scans, and more via MCP.

  • Task Completion Rate (MCP actions)
  • Function Error Count
  • Handoff Rate
  • CSAT

Tracked via CS Chatbot Dashboard + New Relic

📝 WordPress Assistant

Available in /wp-admin pages. Helps users maintain and manage WordPress sites using metadata (plugins, themes, environment).

  • CSAT (only tracked metric)

Data stored in AI-Chatbots DB · Retrievable via BigQuery

📚 hTutorials Assistant

Designed to answer questions about Hostinger tutorials. Syncs conversation history via user_id in browser cookies.

  • Evaluation metrics TBD (not yet released)
  • Amplitude links to be prepared post-launch

Under development by AI team

🤖 CS Chatbot (Support)

Main customer support bot. Handles billing, domains, hosting, email, and general inquiries. Full metrics tracked.

  • All CS Core Metrics (see section above)
  • Conversation type splits
  • Sentiment distributions

CS Chatbot Dashboard · Tableau · New Relic

🛠️ Monitoring & Tracking Tools
Tool What It Tracks Used For
Tableau Chatbot load and performance overview CS chatbot performance, conversation trends, load scheduling
Amplitude Sales chatbot events: Conversion Rate, Cart Link Clicks, Engagement; hPanel AI tool events Product chatbot performance, AI tool engagement tracking
New Relic AI chatbot load data, APM overview (latency, uptime, errors) Real-time infrastructure monitoring for AI chatbot services
BigQuery AI-Chatbots database (WordPress Assistant CSAT, conversation data) Deep data analysis and historical reporting
KodeeDesk (Intercom) Conversation counts, handling time, wait time, FCR, SLA CS team KPIs and specialist performance
🚀 Tool Quick Access
Open in → 📊 Tableau — CS Folder 🤖 Lumos — New Chat 🗄️ BigQuery — hostinger-systems 🔴 New Relic — AI Chatbot Load 📈 Amplitude — Analytics 💬 KodeeDesk (Intercom)
🔬 AI Team — Innovation & Future Capabilities

Since 2024, the AI team focuses on advancing Kodee's core capabilities rather than day-to-day operations (owned by Hostinger Chatbots team).

🧠

NLU Improvement

Enhancing Kodee's Natural Language Understanding to process complex, multi-intent user queries more accurately. Measured by reduction in crash count and handoff rate.

🎯

MCP Tool Accuracy

Tracking accuracy of MCP tool calls (green ✅ vs. red ❌ outcomes). Each MCP server supports up to 128 tools; accuracy is monitored as tool count scales.

🔧 MCP Dashboard
🔄

POC → Production Rate

% of Proof-of-Concepts that successfully transition to production. Target: at least 30% of POCs should progress to production or significantly influence product decisions.

🌐

Multimodal Capability Rollout

Tracking adoption of new input types: text, voice, images. Measured by feature availability and user engagement with non-text inputs.

🤝

Agent Delegation Success

When Kodee delegates to a specialist agent, tracking whether the delegation completes successfully vs. times out or errors. Includes background processing time monitoring.

📉

Handoff Reduction (OKR)

Core OKR goal: reduce total conversations handed off to specialists by improving Kodee's competence, metadata access, and self-critique capabilities.

📋 Conversation Types Reference
Type Definition Significance
chatbot Started and concluded entirely by Kodee — no specialist involved ✅ Ideal outcome — full self-service resolution
not_empty Both specialist and client sent at least one message Partial handoff — Kodee started but specialist finished
reopened Closed/snoozed conversation reopened with multiple specialist replies Indicates unresolved issues on first contact
empty No message from either party Deducted from load stats (spam, duplicates)
👥 Team Ownership

Who Owns What