Customer service cost per ticket runs from about $2.70 in retail to $25 to $35 in SaaS and $15 to $30 in banking and fintech, with complex regulated cases reaching $50 or more, according to [LiveChatAI's analysis of published cost studies](https://livechatai.com/blog/customer-support-cost-benchmarks). Work it out as total support operating cost divided by tickets handled, then divide the same cost by unique issues resolved to see how much repeat contacts add. In 2026 the biggest lever is moving eligible tickets to AI that resolves them, and several vendors now publish a price per resolved ticket, Lorikeet among them.

Cost per ticket is one of the most widely tracked metrics in customer support, and one of the most commonly misread. The number most teams see is cost per contact. The number that matters for budget planning is cost per issue resolved, which accounts for how many contacts each problem takes. This guide, updated 22 October 2026, covers the benchmarks, what pushes your number above them, and the levers that move it.

Key takeaways

• Benchmarks by sector: retail and e-commerce $2.70 to $5.60, banking and fintech $15 to $30, SaaS $25 to $35, telecom and utilities $20 to $30, with a cross-industry baseline of about $6 to $7 per contact (LiveChatAI).

• Agent salaries and benefits make up 60% to 80% of total support cost (LiveChatAI), so cost only falls meaningfully when fewer issues need an agent.

• Repeat contacts are the hidden multiplier: if each issue takes 2.3 contacts on average, your real cost per issue is 2.3 times your cost-per-contact figure.

• Published AI prices per resolved ticket in October 2026 range from $0.75 (Help Scout AI Answers) to $2.00 (Zendesk pay-as-you-go automated resolutions), so check what counts as a resolution and what else you pay for.

Which tools help lower cost per ticket?

The tools that lower cost per ticket are the ones that resolve whole issues, not only answer questions, and price by the resolved ticket so you can model the saving. Four published options:

• Lorikeet resolves tickets end to end inside your existing helpdesk at $0.99 per chat, email or SMS resolution on Start and $0.90 on Scale, on top of an annual plan; escalations to a person and unresolved tickets are not charged.

• Fin (now part of Salesforce) charges $0.99 per outcome, with a $49 monthly minimum that includes 50 outcomes, on Intercom or another helpdesk.

• Zendesk includes AI agents in every Suite and Support plan and bills per automated resolution, with pay-as-you-go resolutions at US$2.00.

• Help Scout prices AI Answers at $0.75 per resolution.

What is cost per ticket in customer service?

Cost per ticket is the total cost of handling one support interaction, calculated as total support operating cost divided by ticket volume over the same period. It is the primary unit cost metric for most teams, but on its own it says nothing about whether the ticket was actually resolved.

The calculation includes direct costs (agent salaries, benefits, tooling licences) and indirect costs (management overhead, training, quality review). LiveChatAI's cost breakdown puts agent salaries and benefits at 60% to 80% of total support cost, software and tools at 10% to 25%, and overhead, facilities and training at 10% to 15%. That is why reduction strategies that do not change how much agent time each issue needs, either by resolving eligible tickets with AI or by fixing issues the first time, have limited effect on the headline number.

What are the industry benchmarks for cost per ticket?

Most published benchmarks put cost per ticket between about $3 and $35, with regulated and technical sectors spending three to ten times the cross-industry baseline. The figures below come from the studies collected in LiveChatAI's 2025 analysis, last updated 13 October 2025; treat them as ranges for orientation, not targets.

By industry

By channel

Phone is the most expensive channel per ticket, because an agent can handle only one call at a time and handle times run longer. Live chat agents can work several sessions at once, which lowers the effective cost per ticket. AI resolution changes the unit entirely: instead of paying for agent minutes, you pay a published price per resolved ticket. Lorikeet, for example, charges $0.99 per chat, email or SMS resolution on Start and $0.90 on Scale, and $1.50 or $1.20 per voice resolution based on a three-minute average call, and does not charge for escalations to a human agent.

How much does an AI-resolved ticket cost all-in?

An AI-resolved ticket costs the per-resolution price plus your share of any plan or seat fee, so the all-in figure depends on volume. Using Lorikeet's published plans as a worked example:

Run the same sum for any vendor: add seat, platform or minimum fees, divide by resolved tickets, and compare the result with the human cost per ticket for the same ticket types. At low volume a fixed plan fee dominates; at higher volume the per-resolution rate does.

What drives cost per ticket higher than benchmarks?

When cost per ticket sits above your sector's benchmark, the cause is almost always repeat contacts, misrouting, channel mix, or AI that escalates more than it resolves.

• Repeat contacts. If 30% of your issues need two or more contacts, your effective cost per issue is well above your cost per contact. Tracking the contact-per-issue rate reveals the multiplier, and reducing it is the highest-impact fix available to most teams.

• Channel misrouting. When a simple password reset lands in a queue behind complex cases, you pay for queue time and agent overhead you do not need. Routing that matches ticket type to the right path, whether AI, self-service or a specialist, removes that overhead.

• AI that answers but does not resolve. AI that hands customers an article and leaves them to finish the task moves contacts around rather than removing them. AI that takes actions end to end removes agent time for those ticket types altogether.

• High escalation rates from AI to agents. If AI escalates half of what it touches, you pay for the AI step and the agent step on the same issue. A high escalation rate usually means the AI's scope needs narrowing to the ticket types it can finish.

What results can you expect from reducing cost per ticket?

The size of the saving depends on how far above benchmark you start and on which levers you pull together. Teams near benchmark have less room; teams well above it usually find several overlapping causes.

Moving your highest-volume, rule-based ticket types to AI resolution changes their unit cost from an agent-handled figure, such as the $25 to $35 SaaS benchmark, to a per-resolution price plus plan fees, as in the worked example above. Raising first-contact resolution cuts the number of contacts each issue generates without changing which tickets are automated. Reviewing 100% of conversations with automated QA surfaces the ticket categories where answers are inconsistent and customers come back. Read more on the wider playbook in our guide to reducing customer support costs with AI.

Cost reduction that comes from resolving issues lasts. Cost reduction that comes from deflecting customers erodes, because they find another way to contact you or leave.

Proof from a published customer story: healthcare provider easykind runs 24/7 patient support without adding headcount, and its email response times dropped 92%, from 24 hours to roughly two.

Lorikeet's take on cost per ticket

Track cost per issue resolved as your primary unit, because cost per contact can fall while total support cost stays flat. We regularly see teams celebrate a lower cost per contact while contacts per issue quietly climb. That is also why Lorikeet prices per resolution and does not charge for escalations or for tickets you mark as bad: the incentive stays on resolving issues. One honest limit: a per-resolution plan still carries an annual plan fee, so at very low volumes the all-in cost per ticket is higher than the headline rate. If your cost per ticket is above benchmark or is not falling despite automation, start a 30-day free trial or book a demo.

Frequently asked questions

How do you calculate cost per ticket?

Divide total support operating costs (agent labour, tooling, overhead, management) by total ticket volume for the same period. For a more accurate view, also calculate cost per issue resolved: divide total operating cost by the number of unique issues resolved in the period. The gap between the two reveals your repeat-contact overhead, which is where most hidden cost lives.

What is a good cost per ticket for SaaS companies?

Published benchmarks put SaaS support at about $25 to $35 per ticket for human-handled interactions (LiveChatAI, citing SaaS Capital). A good number is one that is falling while resolution quality holds. Blended cost drops as more eligible tickets are resolved by AI at a published per-resolution price.

Does automating tickets actually reduce cost per ticket?

Yes, when automation finishes the work rather than only triaging it. AI that routes a ticket to an agent saves little. AI that resolves the ticket end to end removes the agent cost for that ticket. The number to watch is resolution rate: automation that hands most tickets to agents after a first reply has a much smaller cost effect than automation that resolves them.