First level resolution vs first contact resolution

To work it out, divide the tickets your first line solved without passing them on by all the tickets it logged, then multiply by 100. If your first line logged 1,000 tickets in March and solved 700 of them itself, your first level resolution rate was 70%.

It isn’t the same as first contact resolution. First contact resolution asks whether a problem was solved in the customer’s first contact, whoever solved it. First level resolution asks whether it was solved without leaving the first level, however many contacts that took. MetricNet, which benchmarks IT service desks, calls the first a quality metric and the second a cost metric.

Service desks call the first line level 1, and many teams say tier 1. In HDI’s definitions, level 1 is the generalists who receive tickets and escalate what they can’t solve, level 2 the specialists with more skills or access rights, and level 3 the engineers who build and maintain the product.

A founder answering support alone is all three levels at once. Deacon is an AI customer support agent that becomes your first level. It answers your customers from your own docs in seconds, hands you the questions that need you, and has a free plan.

How to calculate first level resolution rate

Net leaves out the carve-outs, MetricNet’s name for the kinds of ticket level 1 can never solve. Some are physical, such as replacing a hardware part remotely. Others are policy, such as level 1 having no access to a system. Take March again. If 100 of its 1,000 tickets were refunds only the finance team could approve, net first level resolution was 700 ÷ 900, or 78%.

MetricNet and HDI disagree about which one to use. MetricNet’s Jeff Rumburg calls net FLR “by far the more important of the two metrics”, because it measures level 1 on the work it’s allowed to do. HDI’s Roy Atkinson notes that some organisations prefer net “knowing that they will have high numbers and ‘look good’”, and advises “If you are going to measure only one of these, measure gross FLRR.” Gross tells you how much of the whole load level 1 carries, and net tells you how well it does the work it’s given.

First level resolution vs first contact resolution

MetricNet explains the difference with two tickets. In the first, a level 1 agent can’t solve the problem on the call, looks into it and calls the customer back with the fix. In the second, the agent warm-transfers the caller to a level 3 engineer, who solves it before the caller hangs up.

So a team can raise one and lower the other. Transferring every hard call to an expert while the customer waits lifts first contact resolution and cuts first level resolution. Rumburg links the first to customer satisfaction and the second to the total cost of support.

What is a good first level resolution rate?

These are the published figures we found for IT service desks, read on 23 September 2026. We found none for software companies’ own support.

MetricNet puts the low end down to “log-and-dispatch” desks, which solve simple things such as password resets and pass everything else on. The desks near the top have well-trained agents with knowledge management systems and remote diagnostic tools. The 2017 column gives the same average and range as the 2011 one, and charts net first level resolution climbing as a desk’s knowledge base matures.

MetricNet’s figures are net, so they leave the carve-outs out, while both of HDI’s count every ticket. Compare your rate with your own past months before anyone else’s.

Why first level resolution is worth raising

Every level up costs more. These are MetricNet’s North American averages from HDI’s 2017 column, counting salaries, facilities, technology and training in the cost of each ticket.

The costs add up, too. A ticket logged at level 1 and solved at level 2 costs $22 plus $69, or $91. Hence Rumburg’s rule that “any ticket that can be resolved at Level 1 should be resolved at Level 1”.

HDI calls moving work down the levels shifting left. Its advice for raising first level resolution is to give level 1 more current documentation and more training, so that what once needed an expert becomes something level 1 can do.

Deacon as your first level

When you answer support alone, you’re level 1, level 2 and the engineer who built the product. Deacon takes level 1 off your hands. It answers what your docs cover in seconds, at any hour and in the visitor’s own language, and passes the rest up to you.

Some questions are carve-outs, in MetricNet’s sense. Deacon always declines a question about the visitor’s own account, invoice or data, whatever your docs say, and it takes no actions in other systems, such as a refund. Those need you every time.

The rest reach you only when Deacon can’t help. When your docs don’t cover a question, it says so instead of guessing, and the widget offers the visitor a box for their email. When a question is too vague to answer, it asks one question back, and after three it stops asking and hands over, summing up in one sentence what the visitor told it. And when a visitor asks for a person, you’re told, whether or not they leave an email.

When a visitor leaves their email or asks for a person, the owners on your team get an email three minutes later, unless one of you has opened the conversation or has the dashboard open, where a toast or the unread badge has already told you. Deacon keeps answering their other questions until you reply. Your reply takes the conversation over, and a visitor who still has the chat open sees it there under your first name and your company’s, such as “Maya from Acme”. A button hands the conversation back, and Deacon answers again. Reply while they’re still in the chat and you’ve done what MetricNet’s warm transfer does, solving it in the first contact but not at the first level.

Then shift the question left for next time. Press Answer this on a question your docs didn’t cover and write the reply, and it becomes part of what Deacon knows. Save & check asks Deacon the question again and shows you the reply it now gives. If Deacon now answers it, Save & check also tells you how many other open questions your answer covers. That’s your knowledge base growing one answer at a time, and the more you answer, the less Deacon needs you.

Topics show which questions keep coming up. They group similar questions, count the people who asked and give most groups a type, such as how-to, account and billing, or feature ask. A how-to that keeps reaching you is one you can move down to Deacon by writing an answer. Account and billing is where the carve-outs gather. And a feature ask is customer feedback about what to build next.

Deacon doesn’t give you a first level resolution rate. Filter Conversations by Unanswered or Wants a person to read what it passed up to you. The Answered figure on Analytics › Widget isn’t that rate either, since it counts a greeting or a request for a person as answered. Human handoff walks through one handover, and you can try Deacon on the free plan, which needs no card.