Connect your AI assistant to your Zendesk metrics
Geckoboard’s MCP lets Claude, ChatGPT, and other AI tools work with your Zendesk metrics. Ask questions, automate reports, and build custom tools using the same trusted and reliable figures as your dashboards.
You: Why did CSAT fall last week?
Geckoboard: CSAT fell from 94% to 87%. Most of the change came from the billing queue. Median first reply time there rose from two hours to nine hours after Tuesday, while your other queues were broadly unchanged.
Sign up for early access and we’ll help you get connected.
Turn your Zendesk metrics into answers, updates, and tools
Geckoboard’s MCP makes accurate Zendesk metrics available in your favorite AI tool. Investigate a change, prepare a regular update, or build a custom tool, without exporting data or rebuilding metric definitions from scratch.
Most of the change came from the EMEA group, where CSAT fell 12 points and median first reply time increased to 9 hours.
Powered by the Geckoboard Metrics Engine
For over 15 years, we’ve helped teams turn messy business data into accurate metrics they can trust on a dashboard. Behind those dashboards is our metrics engine — it connects to your help desk, call center, CRM, and analytics, and turns the raw data into clean, consistent metrics. Geckoboard MCP opens that engine up to your AI assistant.
How it works
Getting trustworthy metrics in front of your AI assistant takes minutes.
Connect your Zendesk account
Create a Geckoboard account and connect Zendesk Support. Geckoboard’s Metrics Engine automatically defines and computes over 50 Zendesk metrics for you.
Connect the MCP
Sign up for early access below. Once we’ve enabled access, you can add Geckoboard as an MCP server in your favorite AI tool.
Ask about your support metrics
Experience the power of simply chatting with your Zendesk metrics instead of wrestling with complex analytics tools.
Starter prompts
Not sure where to start? Copy one of the prompts below into your assistant, then adapt it to suit your team.
Find the source of a change
Compare CSAT this week with the previous four-week average. Break down the change by channel, then show me which related metrics moved over the same period.
Prepare the weekly support review
Summarise tickets created and solved, backlog movement, SLA attainment and CSAT for the past week. Call out the largest changes and draft a concise update for the support channel.
Review SLA performance
Show SLA attainment for urgent tickets by SLA policy. Compare it with last month and identify where the largest decline occurred.
Prepare for agent one-to-ones
Summarise first reply time, ticket touches, and CSAT for each agent over the past month. Present the figures as coaching context, without making performance judgements.
The MCP server is a brilliant added value to my team and my overall analysis, it gives me a great overview, especially on annual reports and trends. I really like it and have started to use it on a daily basis.
Keep your support team on track with live dashboards
The same accurate Zendesk metrics available through the MCP can also power live Geckoboard dashboards. Bring queues, backlog, first reply time, SLA attainment and CSAT into one shared view, where the whole team can see them.
When performance is visible, the team can spot pressure building and respond while there’s still time to make a difference. Managers spend less time chasing updates, and everyone works from the same numbers.
Anyone responsible for understanding or reporting on Zendesk performance can benefit from direct access to their support metrics: support leaders, Support Ops teams, analysts and people building internal tools.
It is particularly useful for investigating changes, preparing regular updates, automating reports and building tools that need consistent Zendesk metrics.
Zendesk doesn’t have an MCP yet, although one is expected mid-2026. When it arrives, it will be built for looking up and acting on tickets, rather than calculating support metrics. That makes it useful for things like help desk automation, but it’s not a reliable way to calculate metrics such as first reply time, SLA attainment, or CSAT.
Geckoboard’s MCP provides pre-calculated metrics with consistent definitions that can’t change from one chat to the next.
No. Connecting an AI tool directly to ticket data would leave it to work from raw records and reconstruct each metric. Geckoboard provides pre-calculated Zendesk metrics with consistent definitions, so the answers match the dashboards your team already uses.
Create a Geckoboard account, connect Zendesk Support and sign up for early access using the form on this page. Once access is enabled, you can connect Geckoboard to any MCP-compatible client. Native connectors for Claude and ChatGPT are coming soon.
Geckoboard’s MCP works with any MCP-compatible client. You can add it as an MCP server today; native connectors for Claude and ChatGPT are coming soon.
As well as Zendesk, Geckoboard’s MCP also supports HubSpot, Pipedrive, Intercom, Aircall, Google Analytics, Google Ads, and Google Search Console, with more being added. If you need another source, tell us what you’d like to see.
Dashboards and the MCP use the same Zendesk metrics in different ways. A dashboard gives the whole support team a shared, live view of performance, with secure sharing and permissions handled for you. The MCP is suited to one-off questions, automations and custom tools. The two work alongside each other.
No. Geckoboard’s MCP provides read-only access to your Zendesk metrics. It can answer questions about your data, but it cannot edit tickets, dashboards or account settings.
The MCP is available to try free for 14 days. It is included on all Geckoboard plans during this pre-release phase. Data limits may apply.
An AI analyst for Zendesk, powered by metrics you can trust
Sign up for early access and we’ll help you get set up.