Email activity data: know which
addresses are already dead.

Six signals per address: last-open timestamp, click recency, reply patterns, device diversity, domain-level engagement trends, and a composite 0 to 100 recency score that rolls the first five into a single freshness number. No mailbox content is ever read. The signals describe whether a mailbox is in active use, not what is inside it, which is exactly the question a list owner needs answered before the next send.

We aggregate engagement events across the sending ecosystem: open pings, click-throughs, reply activity, and the device fingerprints tied to an address over time. The most recent confirmed event becomes the last-open date, and the gap between open, click, and reply events feeds the recency score. Preview-pane auto-opens are discounted, so a mailbox that technically registers opens but never clicks or replies still drifts toward the dormant segment.

A recency score is a 0 to 100 freshness rating for one address. Above 70, the mailbox engaged recently and is safe to email. Between 30 and 70 it is fading and belongs in a re-engagement segment. Below 30 it is stale, and below 10 it is effectively dead. Scores recompute every 24 hours, so a subscriber who comes back climbs out of the dormant bucket on their own.

A sunset policy is a written rulebook for inactivity milestones: re-engage at 30 days, throttle at 60, send a final "still want these?" at 90, suppress at 120, remove at 180. Without one, dead addresses accumulate silently, bounce and complaint rates creep up, and Gmail and Outlook start filtering you for everyone, including the readers who still open every send. The full five-stage policy is laid out above on this page.

Yes, used correctly. Activity data measures engagement events, not message content or personal attributes, and Verifox processes addresses in memory without storing, logging, or selling them. We are SOC 2 Type II audited. You remain the controller of your own list, and suppressing subscribers who stopped responding is widely treated as good practice under GDPR. Details live in our privacy policy.

Recency scores refresh on a 24-hour cycle, so the same list uploaded a month apart shows addresses migrating between segments as people return or go quiet. Lookups are fast too: roughly 380 ms median for a single address, sub-50 ms when cached, and bulk jobs run at about 10,000 emails per minute on the standard tier. A 50,000-row list finishes in minutes.

Verification answers "does this mailbox exist and accept mail?" through our 9-point engine: syntax, MX records, SMTP handshake, catch-all detection, and more. Activity data answers the next question: does anyone still read it? A mailbox can verify perfectly and sit abandoned for two years. Run email verification first to strip invalid addresses, then use engagement data to decide who actually deserves your sends.

Yes. The REST API returns the recency score, last-activity window, and segment label alongside the standard verification verdict, so one call covers both checks. Request and response shapes are documented in the API reference. If you build with AI agents, the same fields are exposed through our native MCP server, so Claude, Cursor, or a custom agent can query your list without glue code.

It runs on the same pay-as-you-go credits as verification. Sign up free and you get 1,000 credits, or 2,500 when you register with a work address. No card required, and credits never expire. Volume rates are shown for your region on the pricing page, with discounts that deepen as usage grows.

Start with anything silent for 180 days or more after a re-engagement attempt. Those addresses are abandoned, recycled, or filtering you to trash, and some turn into spam traps. Before deleting, run them through the email bounce checker to confirm hard-bounce risk, and watch your domain with deliverability monitoring after the cut. Most teams see placement improve within two or three sends.