Prism
Start free
Actions

Find out first, and make it somebody’s job.

Know about the drop before anyone has to ask — rules and anomaly detection watch each day's numbers as they land, and whatever fires joins a queue with an owner, a status and a due date.

Prism's Action Center: three items needing attention, three high severity, four tracked, one overdue and one unassigned, above a monitoring status panel showing five rules armed, last evaluated three hours ago on the Search Console sync, seven alerts ever raised, and where alerts get delivered.
Monitoring · on every sync

Two ways something reaches the queue.

One you write. One you do not have to.

Rules you write
Hear about it the day a number crosses your line — a metric, a condition and a threshold, checked against the previous period as each day's data lands. Rules pause rather than delete.
Anomaly detection
One freak day can't hide the next real drop — anomalies are scored against a typical day, not an average one.
The task queue
What fires becomes an owned task with a status and a due date, and the work lands in whichever tracker your team already uses.

Log what you shipped. Let the chart judge it.

An annotation takes ten seconds on the day you ship, and turns the note into a claim the data either supports or does not.

  • Ten seconds to log

    Name the release, pick the day. The annotation sits on every performance timeline from then on.

  • A verdict, not a memory

    Each one carries a before-and-after click comparison over its own measurement window.

  • On the report too

    The scheduled report's chart carries your annotations, so a spike arrives with its explanation attached.

Prism's annotations log: two entries, a technical migration and a content cluster launch, each with the clicks before and after its measurement window and a percentage impact verdict.

Alerts fire on numbers Google has finalised.

Alerting runs on your Search Console metrics — the figures Google settles every day. Everything else Prism watches is monitored and reported on its own page.

Search Console metrics
The alerting source. Clicks, impressions, CTR and average position, site-wide: your rules plus automatic anomaly detection, evaluated on every daily sync.
Site Health findings
Found on every crawl, classified by effect on traffic in Site Health, and surfaced in the triage queue where they become owned tasks.
AI answer visibility
Measured on every scheduled run and reported across the AI Visibility suite, with every answer stored word for word.
GA4 conversions
Reported in Traffic on the same dates as your search data, and used to price how much a task is worth doing.

Frequently asked questions

What can an alert rule watch?

Clicks, impressions, click-through rate and average position, site-wide, from Search Console. A rule is a metric, a condition (drops by a percentage, rises by a percentage, or crosses a threshold) and a number. Rules are evaluated against the current period and the one before it on every Search Console sync.

What is anomaly detection doing that my rules are not?

It catches the drops nobody wrote a rule for. Prism keeps a rolling snapshot of each metric, takes the median of the last 28 days and the typical spread around it, and raises an incident when today sits far outside that range.

How does an alert reach me?

A fired alert lands as an in-app notification for the rule owner. What fires can become a task with an owner and a due date, and task creation fires a signed HTTPS webhook, so the work can land in whichever tracker your team already uses. The scheduled report goes out by email and in-app on the cadence you set.

How often do alerts run?

Daily, on every Search Console sync, aligned to when Google finalises its numbers. Every alert fires on a finalised figure you can act on.

Set one rule. It runs every day.

Connect Search Console, arm a rule on clicks, and Prism checks it against the previous period on every sync from then on.