Most pipeline teams treat every visit as a new lead, so their alerts describe strangers. Return visitor alerts B2B teams can act on describe something else: a company you have already seen, coming back, and what it did differently the second time. That difference decides whether a rep replies in minutes or never.
A first visit is curiosity. A second visit to a high-intent page — pricing, a comparison, security or integration docs — is evaluation. Between the two, something happened your site cannot see: an internal conversation, a budget check, a shortlist. The return is the visible trace of that conversation, and it is the closest thing to a buying signal your website produces without a form.
Below: what belongs in the alert, how to time the follow-up, and where the privacy line sits. The product doing the work is Prynt Reveal, which names the company behind a visit when that visit comes from the company’s own network.
Why a returning account on pricing is a different signal
A first visit usually means discovery: a search result, a link, a mention, no context beyond your headline. A return means context exists — they read something, compared something, or talked to someone. When the return lands on pricing, the account is past the “what is this” stage and into “what does it cost us, and against what”.
Pricing is where deals are decided and where buyers go to check a number after an internal conversation. One return to pricing is a signal. Two returns with new pages in between — security, docs, an integration page — is shortlist behaviour.
Most alerting fails in one of two ways. It fires once per company and then suppresses everything, so the return never surfaces. Or it fires on every visit with identical content, so reps learn to ignore it. Alert on the change, not the visit.
What a useful return visitor alert contains
An alert is only as good as the evidence behind it. Each return should carry: the company name with a confidence level and the evidence that supports it; what changed since the last visit (new pages, a higher intent score, a new segment); ICP fit, so a low-fit return does not get the same treatment; and the identified person, but only when that person exists in your first-party data from a form, login or CRM import, with a consent record.

Keep the anonymous ones anonymous. Home broadband, mobile carriers, VPNs, proxies and hosting stay unnamed by design — you may see page traffic without a company. Do not let a tool fill that gap with a guess; a wrong name costs a rep more time than no name. Our accuracy page explains how identification confidence is reported.
How to time the follow-up
First return on pricing: same business day
Respond while the context is warm. The message should reference what changed — pricing, plus the security page they opened afterwards — not a generic “saw you stopped by”. If your team lives in Slack, route the alert there with a link to the account timeline.
Return inside your history window: escalate
If the same account returns twice inside your Reveal history window — 30 days on Starter, 90 on Growth, 365 on Scale — treat it as a shortlist entry, not a lead. Change the owner, change the playbook, or send the pricing comparison before they ask for it.
When to wait
Skip the alert when the account opted out or sits on your suppression list, when ICP fit is clearly wrong, when the visitor is an existing customer reading docs, or when bot detection flags the traffic. A skipped alert is cheaper than a bad one.
Wire the alert into the tools your reps already use
Reveal sends the account and the delta to Slack, HubSpot, Salesforce, n8n, Zapier or a signed webhook, and playbooks fire on rules you define — a return to pricing from a company in a target segment, for example. Outreach is AI-drafted but reviewed and sent by a person; nothing goes out automatically. For longer chains — create a task, wait, notify a channel, update a CRM field — Prynt Automate runs the workflow and keeps a run log.

Set the privacy boundaries before you switch it on
Reveal works from first-party identification of companies on their own networks, with no data brokers and no guessing anonymous people. Two things belong in your own documentation before launch: the public opt-out and suppression list, and the data processing terms you rely on. Publish the privacy notice where buyers can find it — ours is at pryntid.com/legal/reveal-privacy-notice — and treat the setup as a compliance question your counsel confirms, not something a marketing tool settles.
FAQ

What counts as a returning account?
A company Reveal identifies again from its own network inside your history window: 7 days on Free, 30 on Starter, 90 on Growth, 365 on Scale. Several people from the same company count as one account.
Do these alerts work if the buyer is on a VPN or home wifi?
No. Those visits stay anonymous by design, so you see the traffic but not the company. Plan for that gap instead of closing it with guesses.
Can I see which person came back?
Only if that person already identified through your first-party data — a form, a login, a CRM import — with a consent record. Reveal does not attach names to anonymous visits.
Start with Reveal. Free names 100 companies a month on one site with 7-day history; Starter and Growth include a 14-day trial when you need a longer window. Create your account and let your first returning account show you the delta.
Try it free
Prynt is device intelligence with a free plan: 5,000 identifications a month, visitor IDs with confidence and bot detection — no credit card. Upgrade to Pro for VPN, proxy, tampering and behavioral signals.