Skip to main content

Analytics & Logs

The Analytics page is where you confirm your automations are pulling their weight. It also doubles as your audit trail, alongside the Rule Logs.

Two sides of the same coin

Analytics answers "are my rules working?" with charts and totals. The Rule Logs answer "what exactly happened to this email?" line by line — and let you undo a rule's changes.

Time range

Pick a time range at the top of the page:

  • 7 Days
  • 30 Days (the default)
  • 90 Days

Several numbers also show a trend — how the current period compares to the one before it (for example, executions this period vs. the previous period).

Overview

Four headline numbers across the top:

MetricWhat it tells you
Total RulesHow many rules you have, with a note of how many are active.
Rules ExecutedHow many times rules ran in the period, with a trend.
Success RateThe share of executions that matched out of the total evaluated.
Emails ProcessedHow many emails MailPrism handled in the period.

Execution breakdown

A clear split of what happened across every time a rule was evaluated:

  • Matched — the conditions were met (the rule applied).
  • Not matched — the rule was evaluated but didn't apply.

Each is shown as a count, a percentage of the total, and a bar — plus the total executions for the period.

Top performing rules

A table of your busiest rules, sorted by how often they ran:

ColumnMeaning
Rule NameThe rule.
ExecutionsHow many times it ran in the period.
Success RateThe share that matched, shown as a bar (green / amber / red).
ActionsHow many actions it carried out.

Click the arrow on any row to open that rule's own per-rule analytics for a closer look.

Action distribution

A breakdown of which actions your rules took most — labels applied, archived, forwarded, AI drafts created, auto-replies sent, tracking labels, marked read, starred, webhooks, nudges, and so on. Each shows a count and a bar, so you can see where the automation is actually happening.

AI usage

If your rules use AI features, this section keeps spending transparent:

MetricWhat it shows
OperationsTotal AI operations in the period.
Total CostThe estimated cost of those operations.
Tokens UsedTotal tokens consumed.
BYOKThe share of operations that ran on your own API keys.
Provider distributionThe split between providers (for example, OpenAI vs. Anthropic).

The BYOK badge only appears once some operations run on your own keys. See Bring your own key (BYOK) for how to bring your own keys.

Export & refresh

Open the More menu (top right) for two options:

  • Refresh — pull the latest numbers.
  • Export Data — download a CSV that includes the overview, top rules, action distribution, AI usage, and a day-by-day breakdown for the selected range.

Rule Logs

From the header, Rule Logs opens the line-by-line history of every rule execution — what matched, what didn't, the actions taken, and (for AI rules) the reasoning behind the decision. It's also where you undo a rule's changes.

→ See Rule Logs & Undo.

Use logs to debug a rule

If a rule isn't behaving, the Rule Logs show exactly what it matched (or didn't). That's usually faster than guessing — see Troubleshooting.