Support desk reporting needs seven metrics, and every mainstream service desk tool holds the data for all of them: ticket volume and open backlog, first response time, resolution time, first-contact resolution rate, reopen rate, SLA attainment, and customer satisfaction, each cut by queue, by agent, by priority and by customer. HubSpot Service Hub, Dynamics 365 Customer Service, Zendesk, Freshdesk and Jira Service Management all report the operational view (how the desk is doing this week) well enough out of the box. What they show less well is the customer view over time: which accounts generate the most tickets, whether that's rising, how much of the desk's effort goes to each customer relative to what they pay, and whether a customer's experience is getting worse before they leave. That view needs the ticket history joined to the customer and contract data, and it's the one the account managers and the MD ask for.
The seven metrics
| Metric | Definition | Formula | Why it matters |
|---|---|---|---|
| Volume and backlog | Tickets created in the period; tickets open at period end | Count created; count where status is open at the date | Demand and whether the desk is keeping up |
| First response time | How long a customer waits for the first human reply | Median (not mean) of first agent reply time − created time, in business hours | The metric customers feel most |
| Resolution time | How long from created to resolved | Median of resolved time − created time, in business hours, by priority | Whether problems get fixed |
| First-contact resolution | Share of tickets resolved without a second customer contact | Tickets resolved with one customer interaction ÷ tickets resolved | Efficiency and customer effort |
| Reopen rate | Share of resolved tickets reopened | Reopened ÷ resolved | Whether "resolved" meant resolved |
| SLA attainment | Share of tickets meeting their response and resolution SLA | Met ÷ tickets with an SLA, by priority | Contractual and reputational |
| Customer satisfaction (CSAT) | Post-ticket rating | Positive ratings ÷ ratings received; show response rate too | The customer's verdict, with the caveat that few respond |
Source: standard service desk definitions. Use medians for the time metrics: one ticket open for three months makes a mean meaningless.
Two rules. Business hours, not calendar hours, for the time metrics, using the desk's actual hours; and by priority, always, because a P1 resolved in four hours and a P4 resolved in four days are both fine and the blended number hides both.
What does each tool report on its own?
| Tool | Operational reporting | Customer-over-time view | Power BI route |
|---|---|---|---|
| HubSpot Service Hub | Ticket dashboards, SLA reports, CSAT, agent performance | Limited; per-company ticket lists | Third-party connector or API (no native Microsoft connector) |
| Dynamics 365 Customer Service | Built-in dashboards, Customer Service Insights / analytics, SLA and CSAT | Some, via analytics | Native Dataverse connector |
| Zendesk | Explore (reporting and dashboards), SLA, CSAT | Some, via Explore | Zendesk API or third-party connector |
| Freshdesk | Analytics, SLA, CSAT, agent reports | Limited | API or third-party connector |
| Jira Service Management | Reports, SLA, CSAT, dashboards | Limited | Native Jira connector in Power BI (Atlassian), or API |
Source: each vendor's documentation, read 21 September 2026. Plan tiers change what's included; check yours.
For the desk manager's weekly view, the built-in reporting is usually enough, and the honest advice is to use it. The gap is the second column.
What does the customer view show?
- Tickets by customer, trend. The top twenty accounts by ticket volume over the last twelve months, with the trend. An account whose tickets doubled in a quarter is either growing or unhappy, and the account manager should know which.
- Effort by customer. Tickets, and if the desk logs time, hours, per customer per month, against the contract value or the revenue from the accounts system. A customer consuming ten percent of the desk on two percent of revenue is a pricing or a product conversation.
- Experience by customer. First response and resolution medians, SLA attainment and CSAT, per customer. This is the early warning for churn and it's invisible in the desk tool's agent-centric reports.
- Recurring problems. Ticket categories or products by volume and by resolution time, over time. What's generating the work.
- Backlog age. Open tickets by age bucket and priority, with the owner. The list the desk manager works from.
That's four pages: the desk this week (operational, from the tool's own data), customers, problems, and the backlog list. Row-level security so each account manager sees their accounts.
Where does the data come from, and what joins?
The ticket data from the desk tool, at ticket level with created, first-response, resolved and reopened timestamps, priority, category, agent, and the customer or company. That alone gives the operational view and the customer volume and experience views, which makes the first version a single-source build.
The join to contract value or revenue needs the accounts or the CRM: a customer key held in both (the accounts customer ID stored on the desk tool's company record, or a mapping table), and invoices or contracts by customer. That's the effort-versus-value page and it's a second source.
Practical notes:
- Business hours have to be defined once in the model (a table of working days and hours) so the time metrics match what the desk tool calculates. If they don't match, nobody will trust either.
- Reopens and merges are handled differently by each tool. Decide how the model treats a merged ticket and a reopened one and write it on the definitions page.
- CSAT response rate goes next to the CSAT score, always. A 95% score from 4% of customers is a different fact from 80% from 40%.
- Reconcile the model's ticket counts by month against the tool's own report. If they differ, the model has a status or date filter wrong.
What does it cost?
Power BI Pro is £10.80 per user per month on Microsoft's UK page as of September 2026, paid yearly, excluding VAT, for everyone who opens it: the desk manager, the account managers, whoever runs the customer reviews.
The single-source version (desk tool only: operational view, customers, problems, backlog) is what the First Report is for: one system, one report answering one written question ("which customers are we serving badly, and is it getting worse?"), the headline figure reconciled back to the desk tool, £1,950 fixed, ten working days (see the First Report). Adding contract value or revenue from the accounts is a second source and a Reporting Sprint.
Common questions
What metrics should a support desk report include?
Volume and backlog, first response time, resolution time, first-contact resolution, reopen rate, SLA attainment and CSAT with its response rate, each by queue, agent, priority and customer. Use medians for time metrics and business hours, not calendar hours.
What is a good first response time?
It depends on the SLA you've promised and the priority. The useful measure is the median in business hours, by priority, against the SLA target, with the share of tickets meeting it. A single blended average hides the P1s.
Can HubSpot or Zendesk report by customer over time?
Both hold the data and show per-company ticket lists, but the trend of volume, experience and effort by customer over twelve months, and the join to what the customer pays, usually needs the tickets in a model. The desk tools are agent- and queue-centric by design.
How do you calculate first-contact resolution?
Tickets resolved with a single customer interaction divided by tickets resolved, in the period. Define "single interaction" (one inbound message before resolution is the usual rule) and print it on the report.
Why should CSAT be shown with its response rate?
Because the score is only from customers who answered, and that's usually a small, unrepresentative share. A 95% score from 4% of tickets and 80% from 40% are different facts; showing both stops the score being over-read.
Can Power BI connect to our service desk tool?
Dynamics 365 Customer Service through the native Dataverse connector; Jira Service Management through the Atlassian connector; HubSpot Service Hub, Zendesk and Freshdesk through their APIs or third-party connectors. All hold the ticket timestamps the metrics need.
How much does a support desk report cost to build?
From the desk tool alone, a First Report at £1,950 fixed, ten working days, reconciled to the tool's own counts. Adding contract or revenue data from the accounts is a Reporting Sprint. Power BI Pro is £10.80 per viewer per month (Microsoft UK, September 2026).