Plan requirement
| Subscription | Suite Professional or higher, Support Professional or higher |
| Access | Admin |
Poor labelling has three causes, and all three are fixed by changing the topics rather than waiting for the model.
Measure it first
- Take twenty recent tickets for one topic.
- Read each and decide whether the label is right.
- Note the wrong ones and what they should have been.
- Repeat for your two or three most-used topics.
That gives you a number instead of an impression, and impressions about this are usually formed from the one bad example somebody remembers.
Measure again after you change anything, on tickets that arrived since. Triage classifies a ticket when it is submitted with a public comment, so a rewritten description reaches new tickets and leaves the old labels where they are. Re-reading the same twenty tickets tells you nothing about the change.
One: topics that overlap
Two topics a person could not reliably tell apart from the ticket text. The labels will be split arbitrarily between them.
Merge them, or rewrite both descriptions until the boundary is a sentence you could give a new colleague.
You do not have to hunt for these on your own. Every time you create, change or deactivate a topic, intelligent triage re-examines the set and flags duplicates and overlaps it finds. So make one change, then read what it says about the rest.
Two: a topic that is missing
A real subject with nowhere to go lands in whatever is nearest, contaminating that topic as well. A large "other" group, or one topic that has quietly become a catch-all, both point here.
Topics are not the place for product names
A common cause of a messy topic list is asking topics to carry detail they were never meant to hold: a specific product, a branch, a subscription type. Zendesk is explicit that topics are not for that, and points to entity classification instead, which exists to pull those details out of the text.
If your list has grown a topic per product, that is the thing to undo before tuning anything else.
Three: descriptions that are too thin
A topic named "Technical" with no description matches almost anything. Write what belongs in it and, more importantly, what does not.
Use the agents' corrections
Agents can change the value when the field is displayed and editable for them, which makes the correction itself a signal. Where agents keep changing one topic to another, that pair is your problem, identified for free. Look at the corrections monthly.
If the fields are hidden from agents, you lose that signal entirely, and this is the argument for showing them.
Read the weekly suggestions
Intelligent triage looks for gaps in your coverage and proposes new topics from your own ticket data, week by week. Those proposals are the same evidence you would gather by hand in the exercise above, already collected.
Change one thing, then wait
Adjust one topic and give it two weeks of tickets before judging. Changing four descriptions at once leaves you unable to tell which helped.
Comments
0 comments
Article is closed for comments.