You don’t need a dashboard platform for one alert.

What Grafana alerting actually asks of you

Grafana is a real alerting engine, and for a team already running dashboards on it, using it for a SQL alert too is the obvious move. But if you don't already have Grafana running, standing one up just to get pinged when a number changes is a lot of infrastructure for one question. You need a data source configured, a panel built against a query, an alert rule with an evaluation interval, and a contact point wired to Slack or email, before the first notification ever fires.

What Watch This does instead

Watch This turns a query, or a saved Ask answer, into an alert directly, no dashboard platform underneath it. Run a query, click Watch This, choose what counts as a change, row count, a specific value, or a threshold, and pick where it goes: Slack, Teams, email, webhook, or a native macOS notification. It runs on a schedule you set, checking on your interval and firing only when the condition is actually met.

A worked example

Say you want to know the moment a Snowflake orders table's daily count looks off:

SELECT COUNT(*) AS orders_today
FROM ANALYTICS.PUBLIC.ORDERS
WHERE ORDER_DATE = CURRENT_DATE();

In Grafana, this means a Snowflake data source, a panel showing that count, and an alert rule with a threshold condition and a notification policy pointed at a Slack contact point, four separate pieces of configuration before the first alert can fire. In QueryFlow, you run the query, click Watch This, set the condition to a value change or threshold on orders_today, and pick Slack as the destination. One toggle, one destination, done.

Feature comparison

When Grafana is still the right call

If you're already running Grafana for dashboards across a dozen data sources, or you need visualization, not just a ping, alongside the alert, adding one more alert rule to an existing Grafana instance is genuinely less work than standing up anything new. Grafana is also the better fit when alerting rules need to reason across multiple panels or data sources at once, which is outside what a single watched query is built for. QueryFlow's Watch This is built for the narrower, more common case: you have one query, you want to know when its answer changes, and you don't want to run a server to find out.

Where QueryFlow sits versus other desktop SQL tools

This gap isn't unique to Grafana comparisons. Among desktop SQL IDEs specifically, QueryFlow is the only one we found where you can watch a query and get pinged in Slack or Teams when the result changes. DBeaver, DataGrip, TablePlus, and Postico have nothing like it; DbVisualizer's Monitored Query comes closest, but it only auto-refreshes a view inside the open app, it doesn't send an alert anywhere.

Connecting it to your warehouse

Watch This works against any of QueryFlow's connectors: Snowflake, Redshift, PostgreSQL, MySQL, BigQuery, Databricks, Salesforce, Google Sheets, or a CSV/Excel file. Connect the source under Databases, write or reuse a query, and Watch This is available from the results view the same way regardless of which connector produced them.

The cost that shows up after setup, not during it

The Grafana comparison above is about getting the first alert running, but the ongoing cost matters just as much. A self-hosted Grafana instance needs patching, upgrades, and someone who notices when the alerting engine itself goes down, since an alert that silently stops firing is worse than no alert at all. Grafana Cloud removes the hosting burden but adds a subscription on top of whatever you're already paying for the underlying data warehouse. A Watch This alert has none of that surface: it's a toggle on a query inside an app you already have open, with no separate service to keep healthy.

Migrating an existing Grafana alert

If you already have a handful of simple threshold alerts in Grafana that only exist to ping Slack when a number crosses a line, moving them to Watch This is usually a matter of copying the underlying SQL query out of the panel, running it in QueryFlow against the same connection, and recreating the same threshold as a Watch This condition. Anything more complex, an alert that correlates multiple panels or applies Grafana's own alerting math across several queries, doesn't have a direct equivalent and is a reason to keep it in Grafana rather than force a one-to-one migration.