Containment Rate
Also written: Bot containment
The percentage of customer inquiries a chatbot resolves without human help.
Containment rate measures how often a chatbot can fully answer a customer's question or solve their problem on its own. A higher containment rate means the bot is more effective and needs to pass fewer issues to a human agent.
For example, if a customer asks about their order status, a chatbot with a good containment rate could look up the information and provide it directly. If the bot cannot find the information or the customer has a more complex issue, it would escalate to a human, lowering the containment rate for that interaction.
Containment rate is a key performance indicator for customer service chatbots, representing the percentage of customer interactions fully resolved by the AI without human intervention. It is calculated as (Total Interactions - Interactions Escalated to Humans) / Total Interactions.
High containment rates suggest an efficient and capable AI system, reducing operational costs. However, an overly aggressive pursuit of high containment can lead to customer frustration if complex or sensitive issues are not properly escalated, potentially harming customer satisfaction.
This term is relevant as companies like OpenAI are enhancing their AI models, such as GPT-4, to handle more complex customer service tasks. Improved AI capabilities aim to increase containment rates, setting new standards for automated customer support.
- Using GPT-4 to deliver a new customer service standardOpenAI · OpenAI News ·
- OpenAI ChatGPT
- OpenAI's GPT-4 is being used to improve customer service bots, which directly impacts their ability to achieve higher containment rates.
- Resolution Rate
- While related, resolution rate focuses on whether the customer'
- First Contact Resolution
- This metric specifically measures if an issue was solv
- Automation Rate
- This is a broader term that can include any automated process,
The concept of containment rate emerged from the customer service industry to measure the efficiency of automated support systems, particularly call center IVR (Interactive Voice Response) systems and early chatbots.
Ask a customer service chatbot from any of the major providers (like ChatGPT, Copilot, or Gemini) a common question, such as 'What are your business hours?' or 'How do I reset my password?'. Observe if the bot provides a direct answer or if it offers to connect you to a human agent.
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.
- PromptRole-play a student struggling with math for AI tutor simulationSimulate a student's math struggles and uncertainty for AI tutor evaluation. · 5 min
- TemplateReusable workflow templates for healthcare teamsTeams can create and share templates for common workflows like documentation or prior authorizations. · 15 min
- TemplateTemplate for building an AI tool tailored to your workA structured approach to defining and building a custom AI tool for specific business needs. · 15 min
- TemplateEvals CSV prompt list for agent skillsA CSV file template for listing prompts used in evaluating agent skills. · 20 min
- HeuristicDouble coding agent task completion with docs linksProviding coding agents with a documentation link significantly increases their task completion rate. · 2 min
Questions people ask
- What is a good containment rate?
- A 'good' containment rate varies significantly by industry and complexity of queries, but rates between 70% and 90% are often considered strong for well-defined tasks.
- Can a high containment rate be bad?
- Yes, if the AI fails to escalate complex or sensitive issues, leading to customer frustration and a negative experience, even if the bot technically 'contained' the interaction.
First seen on Subvolts Sep 5, 2024. Updated Sep 15, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.