One Touch Tickets in Zendesk
One touch tickets in Zendesk are solved on the first reply, no back and forth. The cheapest ticket you'll ever handle and the happiest customer you'll ever have.
What counts as one touch tickets in Zendesk
A one touch ticket is solved with a single public reply from an agent. The customer wrote in, somebody answered, that was the end of it.
Internal notes usually don't count against it, and neither does an automated acknowledgement, though you should decide that explicitly and write it down. The definition matters more than which definition you pick, because the number only means something if it means the same thing every month.
It overlaps with first contact resolution without being identical. FCR asks whether the customer had to come back at all, including in a new ticket. One touch is narrower and much easier to measure, which is exactly why it gets used.
Why it is worth watching
Three reasons, in descending order of how convincing they are to a finance director.
Cost. Every additional exchange adds handle time, and handle time is the whole cost model of a support team. A queue that moves from a third one-touch to half one-touch has found capacity without hiring.
Satisfaction. Customers rate resolved-in-one far higher than resolved-eventually, and the gap isn't subtle. Nobody enjoys a four-message thread about a password.
Diagnosis. A low rate tells you something specific. Either you are not collecting enough information at intake, or agents lack the authority to just fix it, or your knowledge is scattered. Each of those has a different fix.
Measuring it in Explore
Explore exposes a count of agent replies per ticket. Build a report over solved tickets, filter to those where agent replies equals one, and divide by all solved tickets in the same period. That ratio is your one touch rate. It isn't complicated, and it's worth saving as a report so nobody quietly recalculates it a different way next quarter.
Slice it before you draw conclusions. By channel, since chat and email behave completely differently. By ticket form or intent, which is where the actionable detail lives. By group, but carefully, because a tier two group should have a low rate and that is a sign of correct escalation rather than a problem.
One filter worth applying to the whole report: exclude tickets tagged closed_by_merge. A merged duplicate closes with zero agent replies and quietly inflates whatever you calculate. That is a small number in most queues and duplicates typically run between 8% and 20% of volume, so it's not always small.
Raising it honestly
One thing not to do: set it as a personal target for individual agents. Publish it by ticket type instead, where it's a diagnosis rather than a scoreboard, and you'll get the improvement without paying for it in gamed numbers.
How the number gets gamed
Any metric attached to an agent gets optimised, and this one has some ugly failure modes.
Solving early and letting the customer reopen. Answering vaguely rather than asking the clarifying question that would have got it right. Moving the conversation to a side channel so the touches happen somewhere the report can't see. Splitting one problem into several tickets so each looks clean.
The defence is to never show one touch rate alone. Pair it with reopen rate and CSAT on the same dashboard. A rising one-touch rate with a rising reopen rate is not an improvement, it's a queue learning to hide work from you.
None of which makes the metric bad. It makes it a ratio, and ratios move for two different reasons: something changed in the numerator, or something changed in the denominator. Work out which one moved before you congratulate anybody.
Frequently asked questions
Is one touch the same as first contact resolution?
Nearly. Zendesk first contact resolution and the one touch resolution rate both mean solved without a second reply, and the Zendesk agent replies metric in Explore is how you measure either of them.
What is a one touch ticket?
A ticket solved with a single public agent reply, with no back and forth. Internal notes and automated acknowledgements are normally excluded, though you should define that once and keep it consistent.
What's a good one touch rate?
It depends entirely on what you support. Consumer ecommerce runs high, technical B2B runs much lower. Compare against your own trend rather than against a benchmark from a different industry.
How do I measure one touch tickets in Zendesk?
In Explore, report over solved tickets using the agent replies metric, count those with exactly one reply, and express it as a share of all solved tickets in the period.
Is one touch the same as first contact resolution?
Close, not identical. FCR asks whether the customer needed to come back at all, including in a brand new ticket. One touch only looks inside a single ticket.
The tickets that need zero touches
A duplicate needs no reply at all, just a merge. Ticket Merger closes them automatically, which also stops them distorting the rate.
Start free trial14-day free trial. No credit card required.