Most B2B teams already have the list: companies that visited the site this week. What they lack is a rule for what happens next. Sales playbooks for website visitors close that gap — they turn a visit into a named owner, a specific step and a date. Without them, the list becomes a Monday ritual where a rep picks the loudest logo and the rest of the accounts wait.
A playbook is a decision table, not an email sequence. For each account you define the condition, the segment, the action and who executes it. The same signals then produce the same step, whether the account is handled by an SDR in Austin or a founder in Lisbon.
This guide builds that table from the first identifiable visit to the pricing-page return.
Start with signals you can defend
A playbook is only as good as the evidence under it. Reveal names the company behind a visit when the traffic comes from that company’s own network, and reports a confidence level with the evidence attached — see how we measure it on the accuracy page. Home broadband, mobile carriers, hosting and VPN visits stay anonymous. People are linked to companies only from first-party data: a form fill, a login, a CRM import, each with a consent record.
Build rules on the parts, not on a score
An intent score nobody can decompose becomes a number reps ignore. Write rules on the components: which pages are hot, how many sessions, how recent, and whether the account fits your ICP. Then the alert reads “three visits to pricing and one to security in four days” instead of a bare score.
Keep anonymous visits out
If a session cannot be tied to a company network or to first-party data, it is a pageview, not a lead. Leave it out of account-level actions.
Segment before you write a single rule

Gate on ICP fit first
The first rule in a healthy playbook is often “do nothing.” Set a minimum ICP fit and let weak-fit accounts accumulate history instead of rep time.
Split by stage, not by size
Four segments cover most B2B motion: first qualified visit, multi-page evaluation (docs, security, pricing), pricing-page return, and existing customer or open opportunity. A first visit rarely justifies a call; a pricing return usually does.
Four playbooks that run from first visit to pricing return
1. First qualified visit
Condition: new account, ICP fit above your threshold, one high-intent page. Action: notify the owner in Slack with company, page and evidence, and create a CRM task with a date you set. Hold outreach unless the person already exists in your CRM.
2. Multi-person interest
Condition: two or more identified people from the same company, or several sessions across evaluation pages. Action: assign the AE, attach the page history to the CRM record, and open with context instead of “just checking in.”
3. Pricing-page return
Condition: the account returns to pricing inside your window and intent rises. Action: escalate — draft the outreach for review, move the CRM stage, post in the deal channel. Teams forget to write this rule, and it is usually the one that pays.
4. Quiet after pricing
Condition: a known account visits pricing, then goes quiet. Action: drop the frequency, route to nurture, keep it in a watch segment. Rep time here costs more than it returns.
Deliver each step where the rep already works
A playbook that lives in a separate dashboard gets checked once a week. Reveal delivers accounts to Slack, HubSpot, Salesforce, n8n, Zapier and signed webhooks, and Automate chains the follow-up steps with retries and a run log — all on one browser agent and one account. Draft messages are written by AI and reviewed by a person before sending; nothing leaves automatically. A Slack step can carry the company, the hot pages and {{trigger.url}} so the rep lands on the page the account read.

Keep the people layer first-party and auditable
Every person-level record in Reveal comes from your own first-party data with a consent record. Public opt-out and suppression lists keep accounts out of playbooks when they ask, and the reveal privacy notice spells out what is collected and why. Your own legal review decides what you can act on in each market.
FAQ

How is a playbook different from an intent score?
An intent score ranks accounts. A playbook decides who acts, what they do and by when. The score alone changes nothing.
Can a playbook contact the prospect automatically?
No. Outreach is drafted and a person reviews and sends it. Automate routes steps to Slack, CRM and webhooks; the send stays human.
What happens when a visit comes from home broadband or a VPN?
It stays anonymous. Reveal names a company when traffic comes from that company’s own network; home, mobile carrier and VPN visits are not attributed to an employer.
Try Prynt Reveal free
The free plan covers 100 companies a month, one site and seven days of history — enough to write your first two rules and test whether the evidence holds. Connect your site at https://pryntid.com and build the playbook this week.
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.