Tool to evaluate response times for a Zendesk ticket.
For every customer comment, we evaluate the time taken for the next agent response.
Additionally, we observe whether the response time involved the weekends (Saturday and Sunday).
The tool (script) generates a CSV file of the following columns:
Column | Type | Description |
---|---|---|
string | Email of customer or agent | |
user type | string | customer or agent |
commented at | string | ISO8601-formatted time when the customer or agent responded |
formula | string | Arithmetic showing how the response time was calculated |
weekends | string | If weekends were involved, this would print the dates of the weekends |
response time | string | Total time taken for this response from agent (empty if customer comment) |
You can see examples under the examples folder.
You can run the script locally or set up on CircleCI for convenience.
You would also need to set up the following secrets via environment variables:
ZENDESK_EMAIL
: Zendesk user emailZENDESK_TOKEN
: Zendesk user API tokenZENDESK_SUBDOMAIN
: Zendesk subdomain (e.g., foobar if you are on foobar.zendesk.com)
Ensure you have Python 3.10.1.
Install dependencies via poetry:
poetry install
To evaluate a ticket, run:
$ export ZENDESK_SUBDOMAIN=foobar
$ export ZENDESK_EMAIL=<email of API token owner>
$ export ZENDESK_TOKEN=<API token value>
# generates a 123.csv (default) for your evaluation
$ poetry run python cli.py https://foobar.zendesk.com/agent/tickets/123
$ open 123.csv
$ poetry run python cli.py https://foobar.zendesk.com/agent/tickets/99 -o foobar-99.csv
$ open foobar-99.csv
Note: Please set up the project on CircleCI first, and add the required secrets as project environment variables.
đź’ˇ This is a template repository on GitHub. Hence, you can create your own from this.
You can then trigger this via the build UI on CircleCI.
- Go to your main branch.
- Click on Trigger Pipeline on the top right.
- Add a pipeline parameter
url
, where the value is the ticket URL (e.g., https://foobar.zendesk.com/agent/tickets/123) - Download the csv from the Artifacts tab when done!