Plan requirement
| Subscription | Suite Team or higher, Support Team or higher |
| Access | Admin |
Switching routing on changes how every agent receives work, so the order you do things in matters more than usual.
The order to do it in
- Set up unified agent statuses and make sure agents understand them. Routing only assigns to people who are available.
- Create capacity rules so there is a limit on what each agent holds.
-
Build your routing trigger, if you are using the standard queue. This is the step people miss, and without it email never routes. Messaging and calls arrive routable on their own. Email, which here includes web form, side conversations and the API, needs a trigger that adds both a group and the auto-routing tag. The tag is called
auto-routingunless you chose your own name for it. - Turn on omnichannel routing. In Admin Center, Objects and rules > Omnichannel routing > Routing configuration, click Turn on omnichannel routing.
- Start with one channel and one group.
- Watch it for a week before extending.
Custom queues do not need any of that. The group and the tag are what the standard queue looks for. A ticket matched by a custom queue is routed on that queue's own conditions, so if you build custom queues you can skip the routing trigger entirely. The exception is a ticket that matches no custom queue and that you still want the standard queue to pick up: that one needs the tag after all.
One thing can block you before you start
Routing cannot be activated on an account running live chat alone. Messaging has to be on as well. Worth checking first, because it is a migration rather than a setting and it turns a morning's work into a project.
The same section of Zendesk's own limitations lists a second blocker: omnichannel routing does not support broadcast or hybrid messaging modes. An account handing conversations out that way has to move to assigned routing first, and that changes how agents receive work before you have switched anything else on.
Do not turn it on for everything at once
This is the mistake that produces a bad week. Agents who have chosen their own work for years suddenly receive it, statuses that nobody was maintaining now decide whether they get anything, and every gap in your configuration shows up at once.
One group, one channel, a week of watching. Then the next.
Statuses become important overnight
Literally overnight, in one respect: the moment routing is activated every agent is set to offline, and they are prompted to choose a status themselves. Until they do, nothing is routed to them. Tell people before you flip the switch rather than after the first quiet hour.
Before routing, an agent status is a courtesy. After, it decides whether work reaches them. And the statuses are finer-grained than online and offline: Away does not stop everything, because it can still receive email tickets. Transfers only takes work from other agents alone, and Offline takes nothing. So an agent sitting on Away is still quietly being given email, and an agent who never goes Offline gets tickets while at lunch.
That is a conversation with the team, not a setting, and it is the part most likely to go wrong.
Views still work
Routing does not remove views. Agents can still look at queues, and supervisors still need to. What changes is where the next piece of work comes from.
Comments
0 comments
Article is closed for comments.