How to Build Reports in Zendesk Explore (Step by Step)

We'll stick to the actual menu labels and the real click path, every step verified against Zendesk's own Explore documentation. Zendesk renames buttons from time to time, so confirm labels in your own account — but the structure below is current as of mid-2026. If you want the conceptual tour of Explore first, our companion piece Zendesk Explore explained covers the what-and-why; this post is the hands-on how.

Explore basics: what you're actually working with

Explore is a full business-intelligence tool bolted onto Zendesk. It ships two kinds of reporting, and the distinction matters before you build anything:

• Prebuilt dashboards — ready-made, mostly read-only dashboards Zendesk maintains for you (ticket volume, agent activity, satisfaction, and so on). Great for a quick glance, but you can't fundamentally restructure them.
• Custom reports (queries) — the ones you build in the report builder by asking a specific business question and rendering the answer as a chart or table. This is where the real power is, and it's what this guide is about.

A quick note on terminology: Zendesk has used "report," "query," and "report builder" somewhat interchangeably over the years. In today's UI you create a report; the workspace you build it in is the report builder. Same thing the older docs called a query.

Does your plan include report building?

This trips people up, so check before you start. Per Zendesk's plan-types documentation, Explore comes in three editions tied to your Suite tier:

The headline: if you're on Suite Team or Growth (Explore Lite), you can look but not build. Creating custom reports requires Explore Professional or Enterprise. You'll also need the Admin or Editor role in Explore — viewers can't save new reports.

How fresh is the data?

Explore is not real-time by default, and assuming it is causes a lot of "the numbers are wrong" panic. Per Zendesk's data refresh intervals doc:

• Explore Lite refreshes daily (data updates around midnight).
• Explore Professional and Enterprise refresh hourly — the next sync starts at a randomized point within the hour after the previous one finishes, and on busy accounts it can occasionally take two hours or more.
• Enterprise also gets the live dashboard, which updates in near-real time (typically 5–10 seconds) for a fixed set of live metrics.

One gotcha worth knowing: if nobody opens a report or dashboard for more than 30 days, Zendesk quietly drops that account's refresh to weekly until something is accessed again. So a dashboard that "stopped updating" may just need a visit.

The building blocks: datasets, metrics, and attributes

Three concepts do all the work in Explore. Get these and the rest is mechanical.

Datasets are the foundation. A dataset is a curated collection of metrics and attributes drawn from one Zendesk product — for example Support: Tickets, which holds everything about your tickets. There are separate datasets for Talk, Guide/Knowledge, messaging, and so on. You pick one dataset per report, and it determines what you're allowed to measure and slice. The full field list for the most common one lives in Zendesk's metrics and attributes for Support reference.

Metrics are the numbers — the what you're counting. They're quantifiable: Tickets (a count of tickets), number of comments, number of updates, reopens, resolution time, and so on. Every metric carries an aggregator (COUNT, D_COUNT for distinct, AVG, MEDIAN, etc.) that you can change.

Attributes are the slices — the how you want it broken down. They're non-quantifiable values like Ticket status, Assignee name, Ticket group, Ticket channel, or a date like Ticket created. Attributes go into the Columns, Rows, Filters, or Explosions of your report to reshape it.

The mental model: a metric is the answer, attributes are the question's "by what?" "Number of tickets (metric) by group (attribute), filtered to last month (attribute as filter)." Hold that sentence and the builder makes sense. (For more on where these fields come from — statuses, groups, assignees — our Zendesk ticketing system explained breaks down the anatomy of a ticket.)

Step 1 — Open the report builder

From anywhere in Zendesk, open the Zendesk products menu (the grid icon) and choose Analytics to launch Explore. Then:

1. Click the Reports icon in the left sidebar.
2. Click New report (top right).
3. The dataset picker opens. Choose your product and dataset — for this walkthrough, Support → Support: Tickets.
4. Click Start report.

You're now in the report builder: a chart canvas in the middle, and panels around it for Metrics, Columns, Rows, Filters, and Explosions. Everything that follows happens here.

Step 2 — Add a metric

Nothing renders until you tell Explore what to count.

1. In the Metrics panel, click Add.
2. Expand the relevant group (for tickets, expand Tickets) and select Tickets.
3. Click Apply.

A single number — your total ticket count for the dataset's default time range — appears on the canvas. Explore auto-picks a sensible aggregator (here, a count). To change how a metric is calculated, click the metric in the panel and pick a different aggregator — for example switch a count to an average or median when you're measuring something like resolution time.

Step 3 — Slice it with an attribute

A lone total is rarely the report you want. Now break it down.

1. In the Columns panel, click Add and select an attribute — say Ticket group.
2. Click Apply.

Your single number becomes a chart of tickets per group. Where you drop an attribute changes the shape of the result, and this is the part most tutorials gloss over:

• Columns — slices your results within one chart (e.g., a bar per group).
• Rows — produces a separate chart for each value (one mini-chart per group).
• Explosions — splits results into multiple charts shown side by side.
• Filters — narrows the data without displaying the attribute on the chart (covered next).

For a classic two-dimensional view, add a second attribute. Following Zendesk's hands-on tutorial, try Assignee name in Columns and Ticket status in Rows — you get tickets broken down by both assignee and status at once. Add only what answers your question; piling on attributes turns a clean chart into noise.

Step 4 — Filter and set a date range

Almost every real report is scoped to a time window. There are two ways to do it.

In the report: add a time attribute — commonly Ticket created (often available as a Date sub-attribute) — to the Filters panel, then open it and choose a range: a relative range like last 30 days or this month, or a fixed start/end. Because it's in Filters, it constrains the data but doesn't clutter the chart.

You can filter on non-date attributes the same way — drop Ticket channel into Filters and select Email to limit the report to email tickets, for instance. Filters are also where you exclude noise like spam or a test group.

On the dashboard: you can instead add a date filter widget to the dashboard (Step 6) so viewers re-scope every report at once. Use report-level filters for fixed reports, dashboard-level filters for interactive ones.

Need a calculation Explore doesn't ship — like "tickets resolved on first contact" or a custom bucket? That's a standard calculated metric or attribute, written with a small formula in the calculations menu. Zendesk's guide to creating standard calculated metrics and attributes covers the syntax; it's the natural next step once the basics click.

Step 5 — Choose a visualization and save

By default Explore renders a sensible chart, but you control the type.

1. Open the visualization type menu (the chart-shape icon in the toolbar).
2. Pick the format that fits the question — Column or Bar for comparisons across groups, Line for trends over time, Table for exact numbers, KPI for a single headline figure.

Then save — and this is the step people forget, because Explore does not auto-save:

1. Click the report name to give it something descriptive ("Tickets by group — last 30 days").
2. Click Save (top right).

If you started from an existing report and want a variant, use the dropdown next to Save and choose Save as new so you don't overwrite the original. Navigate away without saving and your work is gone — Explore won't warn you the way a document editor would.

Step 6 — Add the report to a dashboard

A saved report lives in your Reports library, but the way you actually use and share reporting in Explore is through dashboards — canvases that hold several reports plus shared filters. Per Zendesk's dashboards documentation:

1. In Explore, click the Dashboards icon, then New dashboard (or open an existing one).
2. Click Add → Add report.
3. Select the report you just saved, then click Add reports.
4. Drag to position and resize the widget on the grid.
5. Add a date filter (and any data filters) so the whole dashboard re-scopes in one click.
6. Double-click the title to rename the dashboard (e.g., "Team performance").

Repeat to stack several related reports — volume, backlog, resolution time, CSAT — onto one screen your team actually checks.

Step 7 — Share and schedule

A dashboard nobody sees is wasted effort. To distribute it:

• Share for access: click Share, select the Zendesk groups (or specific people) who should see it, and click Invite. Recipients get an email and can open the live dashboard themselves.
• Schedule deliveries: set up a recurring delivery so the dashboard is emailed automatically — Monday-morning numbers in the right inboxes without anyone logging in. This is the single highest-leverage habit in Explore: reporting that arrives beats reporting people have to remember to pull.

Remember the refresh cadence from earlier — on Professional/Enterprise the data behind a scheduled email is at most an hour old, which is fine for daily and weekly rhythms but not for minute-by-minute monitoring (that's what the Enterprise live dashboard is for).

Best practices and common mistakes

A few habits separate clean Explore reporting from a graveyard of half-built charts:

• One question per report. "Tickets by group" and "resolution time by group" are two reports, not one overloaded chart. Compose them on a dashboard instead.
• Name everything as you go. "Copy of Copy of report" is how libraries rot. Name the report and the dashboard the moment you save.
• Mind the aggregator. A count of resolution-time values is meaningless; you almost always want an average or median. Click the metric and check.
• Filter out test and spam data. Add a Filters-panel exclusion for your test group or spam tag so it doesn't skew totals.
• Don't mistake hourly for live. If a number looks "wrong," it may just be up to an hour stale (or weekly, if the dashboard went untouched for 30+ days). Check the refresh interval before you debug the formula.
• Build on prebuilt where you can. On Professional/Enterprise you can clone a prebuilt dashboard's report and tweak it — faster than starting from a blank canvas.

Where AI fits into your reporting

Explore is only as insightful as the data flowing into it — and a growing share of that data now comes from AI handling tickets before a human ever does. That's worth a moment of honesty about what reporting can and can't tell you.

When you run an AI agent like Macha on top of Zendesk, it resolves and triages conversations directly in the helpdesk — and every one of those actions becomes data Explore can report on. Deflected tickets, AI-resolved conversations, escalations with full context, tags the agent applied: it all lands as standard ticket data you can slice by the same metrics and attributes you used above. So a "tickets by channel" or "resolution rate" report quietly starts reflecting your automation, not just your humans.

The honest framing: Macha is not a reporting tool and it's not a Zendesk replacement — it's an AI agent layer that runs on top of your existing Zendesk. It won't build your Explore dashboards for you. What it does is generate cleaner, richer ticket data (consistent tagging, clear AI-vs-human resolution, escalation reasons) so the reports you build in Explore actually mean something. On cost, Macha bills per AI action — each automated step it takes, like drafting a reply, tagging, or routing — not per resolved ticket, since most automation is work done along the way rather than a tidy "resolution." If a meaningful slice of your volume is repetitive questions your knowledge base could answer, that's the work it takes off your queue — and then reports on. You can try it free — 7-day free trial, no credit card required.

Frequently asked questions

How do I create a report in Zendesk Explore? Open Analytics → Reports → New report, pick a dataset (e.g., Support: Tickets) and click Start report. In the Metrics panel click Add and choose a metric like Tickets, then add an attribute (e.g., Ticket group) to the Columns panel to slice it. Choose a visualization, name the report, and click Save.

What's the difference between a metric and an attribute? A metric is the number you're measuring — it's quantifiable, like a count of tickets or an average resolution time. An attribute is a non-quantifiable value you break that number down by, like ticket status, assignee, group, or created date. Metrics answer "how many?"; attributes answer "by what?"

Why is my Zendesk Explore data not up to date? Explore isn't real-time. Lite refreshes daily; Professional and Enterprise refresh hourly (the sync starts at a random point within the hour). And if no report or dashboard is opened for over 30 days, the interval drops to weekly until you access one again. Only the Enterprise live dashboard updates in near-real time (5–10 seconds).

Can I build custom reports on every Zendesk plan? No. Building custom reports requires Explore Professional (Suite Professional) or Explore Enterprise (Suite Enterprise/Enterprise Plus). Explore Lite (Suite Team/Growth) is view-only for prebuilt dashboards. You also need the Admin or Editor role to save reports.

How do I add a report to a dashboard and share it? Open the Dashboards icon → New dashboard → Add → Add report, select your report, and click Add reports. Position the widget, add a date filter, then click Share, pick the Zendesk groups, and Invite them. You can also schedule the dashboard to be emailed on a recurring basis.

How do I filter a report by date range? Add a time attribute such as Ticket created to the Filters panel and choose a relative range (e.g., last 30 days) or fixed dates. For interactive dashboards, add a date filter widget instead so viewers can re-scope every report at once.

The bottom line

Building a report in Zendesk Explore is one repeatable loop: open the report builder (Analytics → Reports → New report), pick a dataset (Support: Tickets), add a metric (Tickets), slice it with an attribute (Ticket group in Columns), filter by date, choose a visualization, and Save — then drop it on a dashboard and share or schedule it. Internalize the metric-versus-attribute distinction and the four panels (Columns, Rows, Filters, Explosions) and you can answer almost any operational question without exporting a thing. Just remember Explore's two quirks: report building needs Professional or Enterprise, and the data is hourly, not live. For the bigger-picture tour of how Explore is structured, see Zendesk Explore explained.

Steps verified against Zendesk's official Explore documentation, June 2026. Zendesk updates its product periodically — confirm labels in your own account before relying on them.