Chatbot Containment Savings and Break Even Calculator
A chatbot on WhatsApp or Teams is worth building only if enough conversations are resolved without an agent. This calculator shows the saving at your containment rate and, more usefully, the containment you must reach just to break even.
Your numbers
Share resolved by the bot with no human handoff.
Loaded agent cost per conversation, not per hour.
Message charges plus model or hosting cost per conversation.
Platform, hosting, maintenance and review time.
Your results
Assumes every conversation incurs the bot cost, including those handed to humans, and that handoffs cost the same as a fully human conversation.
Monthly saving vs containment rate
Sweeps Containment rate from 0 to 90, holding your other inputs fixed.
Want your containment target sanity checked?
An optional review of intents, data access and handoff design for an ERP support bot.
Optional. The tool and every export work without this. No spam.
How to use it
Count real conversations, not messages, from your helpdesk or shared inbox. Use a loaded cost per handled conversation: agent salary, overhead and tooling divided by conversations handled. For the bot, add per conversation variable costs and a fixed monthly figure that includes the time someone spends reviewing failures. Containment should be measured as resolved with no handoff and no repeat contact within a few days, not merely ended by the user.
- Count repeat contacts as not contained
- Include review and tuning time in the fixed cost
- Start with the three most common, simplest intents
Formulas and reading the result
Cost without a bot equals conversations times human cost. Cost with a bot equals human handled conversations times human cost, plus all conversations times the bot variable cost, plus fixed cost. Break even containment equals bot variable and fixed costs divided by the total human cost, as a percent. With 6,000 conversations, a 4.50 dollar human cost, 0.35 bot cost and 1,800 fixed, break even is about 14 percent. At 45 percent containment the saving is roughly 10,000 dollars a month.
Common mistakes
Containment figures from vendors often count abandoned chats as resolved. Test with sampled transcripts. Also, a bot rarely removes agent headcount in a small team; the saving shows up as capacity, not a smaller payroll, so say that honestly. For ERP related questions such as order status, a simple lookup flow with exact answers beats a general chat model, and private hosting may be needed for sensitive data.
- Sample transcripts to verify real resolution
- Do not claim headcount savings you will not realize
- Offer a human handoff at every step
Frequently Asked Questions
What is a realistic containment rate?
It depends on intent mix. Narrow, repetitive questions such as order status can reach a high share, while open ended support is much lower. Measure on a pilot instead of using a vendor figure.
Should handed off conversations cost more?
Often yes, since the agent must read the bot transcript first. If so, raise the human cost slightly or lower containment to reflect it.
Does this include development cost?
Only as part of the fixed monthly figure. Spread a one time build over 24 to 36 months and add it to fixed to see a true payback.
How is it different from the chatbot vs agent calculator?
This one focuses on containment and break even for a single bot. The related calculator compares a chatbot with an agent style system.
If you want a support or internal helpdesk bot that answers from live ERP data and runs on your own infrastructure, Netray can scope it with you.
Related Tools
Support Automation Deflection Calculator
Model both sides of support automation - tickets fully deflected and tickets handled faster with AI assist - to get net annual savings and FTE capacity freed.
AI Agents & AutomationERP Chatbot Deployment Cost Calculator
Model the total first-year cost of an ERP chatbot, from per-integration build cost through model hosting and ongoing maintenance.
ERP OperationsApproval Notification Speed Savings Calculator
Quantify the cost of approval wait time and the saving from pushing purchase, expense or ECO approvals to chat with actionable buttons.
AI Agents & AutomationWhatsApp Template Delivery to Reply Funnel Calculator
Turn sent, delivered, read and reply rates into a cost per reply for WhatsApp template campaigns such as order updates and supplier reminders.
More AI Agents & Automation tools
See all- Business Process Automation ROI Calculator
- AI Agent Use Case Prioritizer
- Document Processing Automation Savings Calculator
- AP Invoice Automation ROI Calculator
- Order Entry Automation Savings Calculator
- Quote-to-Cash Cycle Time Calculator
- Manufacturing AI Readiness Assessment
- RPA vs AI Agent Selector
- ERP AI Maturity Assessment
Keep exploring
Free tools
- AI Email Triage Savings CalculatorTurn daily email volume, manual handling time, and automation rate into monthly and annual savings from an AI email triage and routing agent.
- Containment Action Plan TemplateA fill-in containment plan that lists every suspect location, the sort or inspection action, the owner and the criteria for releasing stock.
- Natural Language ERP Reporting Savings CalculatorTurn report volume, build time, and automation rate into the monthly hours and dollars saved by letting users ask ERP questions in plain English instead of building reports.
ERP fixes and how-tos
- How to Link a CAPA to ERP Lots and Serial Numbers for Containment and EffectivenessUse ERP lot and serial genealogy to scope containment for a CAPA, record affected jobs and shipments, and verify effectiveness with real data instead of opinion.
- Price Book Session in Infor LN: Setup and Order PricingHow price books work in Infor LN sales pricing: where they are maintained, how they link to price lists and order lines, and how to troubleshoot a price not found.
Guides
- Compression vs Transfer vs Injection Rubber Molding: Which Costs Less?Compression vs transfer vs injection rubber molding cost: how tooling, cycle time, flash, runner waste and labor shift the break-even volume for each process.
- Machine Shop Scheduling for CNC Job Shops: Why Schedules Break and How to Fix ThemMachine shop scheduling for CNC job shops: why schedules break, finite versus infinite capacity, the data it needs, and where AI helps daily.
- ERP Chatbot Architecture: Grounding, Permissions, and AuditDesign an ERP chatbot architecture that grounds answers in live data, enforces per-user permissions, and produces an audit trail your team can defend.
AI on your ERP
- AI for Plex in Automotive Supply: Answers From Your Own Releases, Control Plans and Containment RecordsPrivate AI on Plex for Tier 1 and Tier 2 auto suppliers: ask about PPAP, control plans, customer releases and containment, with read-only API access.
- AI on Your ERP for Automotive Suppliers Working Under IATF 16949Private AI on your ERP for automotive suppliers: EDI releases, PPAP, SCARs, containment and OEM scorecards under IATF 16949, kept on infrastructure you control.
Integrations
- How to add an AI chatbot to Infor SyteLineHow to add an AI chatbot to Infor SyteLine for staff or customers: grounded answers from ERP data, role-based access, portal embedding, and on-prem open models.
- How to add an AI chatbot to Infor VISUALHow to build an AI chatbot on Infor VISUAL data for planners, buyers and customers, grounded in the SQL Server database and able to run fully on-prem.
Troubleshooting
- Why CUDA runs out of memory during fine-tuning even when inference works finetorch.cuda.OutOfMemoryError during fine-tuning means AdamW optimizer state (8 bytes per parameter) exceeds VRAM. Use LoRA, QLoRA, or an 8-bit optimizer.
- Why your RAG system hallucinates even when the retrieved context has the right answerRAG hallucinates despite correct context when prompts skip grounding instructions, passages sit mid-context, or retrieved chunks conflict. Here is the fix.