Freshdesk Tags
Tags are free-text labels with no validation, which makes them flexible and makes them rot. Here's how to keep a tag list worth reporting on.
What tags are, and what they're not
A Freshdesk tag is a free-text label attached to a ticket. Agents add them, automations add them, and you can filter views and reports by them.
What they are not is a validated field. There is no dropdown enforcing the options, no required format, and nothing stopping an agent typing a new tag that differs from an existing one by a hyphen. Within a year, a busy account has "billing", "Billing", "billing-issue" and "billing_query" all in circulation, each describing the same thing and none of them countable together.
That's the whole problem with tags, and it's why the rest of this page is mostly about discipline rather than features.
The strength is the flip side of the same coin. Tags cost nothing to add, need no admin change, and can carry several dimensions at once. A ticket can be tagged with a reason, a source and a campaign simultaneously, which no single dropdown can do.
A Freshdesk tags naming convention that survives a team
Four rules. Write them down somewhere agents will see them, and enforce them for the first month.
payment-failed, not Payment Failed. Consistency here alone eliminates most duplication.reason-billing, source-social, product-mobile. The prefix makes tags sort into meaningful groups in every picker and makes reporting filters obvious.Then clean up what exists. Pull a list of every tag in use with its ticket count. Anything used fewer than five times in a year is either a typo or a dead idea, and both should go. Freshdesk lets you manage the tag list from the admin area, and bulk actions on filtered views let you retag before you delete.
Do this once a quarter. It takes half an hour and it is the difference between a tag report you trust and one you argue about.
Let automation do the tagging
The most reliable tag is the one no human typed.
Freshdesk automation rules can add tags on ticket creation and on update, based on conditions you control. Use that everywhere the tag is derivable from something the system already knows.
reason-billing without anyone thinking about it.reason-refund. Imperfect, and still far more consistent than manual tagging.Reserve manual tagging for genuinely human judgement, the sort of thing a rule cannot infer, such as whether a ticket revealed a product bug or a documentation gap. Two or three manual tags that agents actually apply beat thirty they ignore. Patterns for building the rules are in the automations guide.
Reporting on tags without kidding yourself
Tags feed filters, saved views and analytics, so you can count tickets by tag, trend them over time and cross them with resolution time or satisfaction.
One caveat, and it is a big one. Tag counts are only as complete as your tagging. If agents tag sixty percent of tickets, your report describes sixty percent of reality, and the missing forty percent is not random. Rushed agents on busy days skip tagging, and busy days are exactly the days you want to understand.
So before you present any tag-based number, check what proportion of tickets in the period carried at least one tag. If it's below about eighty percent, either fix the tagging with automation or present the figure as a proportion of tagged tickets and say so out loud.
The most useful tag report isn't the volume ranking. It is the tag crossed with a quality measure. Tickets tagged reason-billing have double the average resolution time, or a worse satisfaction score, tells you where to spend engineering effort. Raw counts just tell you what people ask about most, which you probably already knew.
When a custom field is the better tool
This is the decision people get wrong, and the rule is simpler than it looks.
Use a custom field when the answer is one value from a known list, and you need it to be reliable. Ticket category, affected product, root cause. A dropdown validates, can be made required before resolution, supports conditional logic, and gives you clean reporting dimensions with no typo risk.
Use a tag when the labelling is many-valued, ad hoc, or temporary. Flagging every ticket related to one outage. Marking cases for a training review. Anything you'll care about for six weeks and then never again.
The test I use: if you would be annoyed to find the value missing on a ticket, it is a field. If you would merely be mildly disappointed, it is a tag.
A common healthy setup uses both. One required dropdown for ticket reason so every ticket has exactly one clean category, plus a small tag vocabulary for everything situational. Field structure is covered in custom fields, and either way your reporting is only as good as the data agents actually enter.
Frequently asked questions
Should you use tags or custom fields?
Freshdesk tags vs custom fields comes down to reporting. Freshdesk ticket tags are free-form and drift; a field with an option list stays reportable. Freshdesk tag management is easier if you automate tags from rules rather than trusting agents to type them.
How do I automate tags in Freshdesk?
Automation rules can add tags on ticket creation or update, based on conditions such as the receiving email address, keywords in the subject, or the product a ticket arrived through.
Should I use tags or custom fields?
Custom fields when you need one validated value you can rely on for reporting, such as ticket reason. Tags when the labelling is many-valued, ad hoc or temporary, such as flagging tickets from one outage.
How many tags should we have?
Around twenty to thirty active tags. Past fifty, agents stop scanning the list and start typing, and typos multiply from there.
Can I report on tags in Freshdesk?
Yes, through filters, saved views and analytics. Check what proportion of tickets in the period were tagged at all before you trust the numbers, because incomplete tagging is not random.
How do I clean up messy tags?
Export every tag with its ticket count. Retag anything under about five uses a year onto the correct tag using bulk actions, then remove the strays. Repeat quarterly.
Tagged twice, asked once
Duplicate tickets inflate every tag count you report on. Ticket Merger finds and merges them automatically on Freshdesk and Zendesk.
Start free trial14-day free trial. No credit card required.