Most website visitor tracking for agencies breaks on the second client, not the first. The first client gets a clean dashboard, a Slack channel and a short list of companies to call. The second client gets a shared view where nobody can tell whose visitor is whose, a report you rebuild in a spreadsheet every Friday, and an opt-out request you apply in three places because it lives in someone’s memory.
An agency does not need more data. It needs separation: one login, one bill, and a hard wall between each client’s visitors, reports and suppression lists. Prynt Reveal is built around that split with agency workspaces, and this is how to run it from your first client to your twentieth.
Why a shared dashboard fails at the second client
The failure is procedural, not technical, and it shows up in three places.
- Mixed data. When every client shares one site list, one report and one export, a visitor from one client’s own account can surface in another client’s list.
- Manual reports. If the tool knows one dashboard and one CSV, your account manager becomes the reporting layer. The retainer starts paying for Friday spreadsheets instead of strategy.
- Opt-out drift. A prospect asks not to be tracked. Someone notes it, and the next report still includes them because the suppression sat in a person’s head rather than in the system.
Set up one workspace per client
A workspace is the unit of separation: its own sites, ICP segments, playbooks and delivery destination. Create it that way on day one, and keep three conventions.
![]()
Name by client, not by campaign
Campaign names change quarterly; clients do not. A workspace called “client — domain” (never “Q3 retargeting test”) is the difference between a five-minute audit and an afternoon of guessing which site belongs to whom.
Keep delivery inside the client’s stack
Point each workspace’s playbooks and webhooks at the client’s own Slack channel, HubSpot pipeline or Salesforce queue. The client then sees accounts arriving with source attached, and you keep the analysis layer instead of becoming the router.
Decide ownership at offboarding
Agree in the contract who keeps the history if the engagement ends; an export is not a handover.
Explain confidence before you send a list
Reveal names the company behind a visit when the visit comes from that company’s own network, with confidence and evidence attached. Home broadband, mobile carriers, hosting and VPN visits stay anonymous. The tradeoff is the point: a short list you can defend beats a long list your client stops trusting.
Two habits make that work in a report. First, show the evidence next to the name — the matched network, the pages visited and the explainable intent score — so a client can see why an account is at the top. Second, say what a miss looks like: an anonymous visit is a correct decline, not a failed match. Our accuracy page explains how confidence and evidence are produced.
Named people come only from first-party data — forms, logins, CRM imports — with a consent record; no data brokers, no guessing anonymous visitors into names.
Handle opt-out and consent once, for every client
Opt-out gets messy because the request arrives at the agency while the record lives with the client. Write the process down before the first report ships.
![]()
- Use the public Reveal opt-out page so a person has one obvious place to ask, and log every request with date, workspace and who handled it.
- Apply suppression the same business day, and check the list before each export, not after.
- Keep consent evidence next to the person record: which form, which site, which timestamp.
- Hand legal review something real: our Reveal DPA covers the processing terms, while the controller question is a conversation with each client.
Documented opt-out handling is often the first thing a client’s legal team asks about.
Deliver into the client’s stack, not your inbox
Rule-based playbooks turn a visit into a next step: an account crosses your ICP threshold, visits a pricing page twice, and lands in the client’s CRM with the evidence attached. Delivery runs to Slack, HubSpot, Salesforce, Pipedrive, n8n, Zapier or a signed webhook, so the record lives where the client works.
Outreach drafts are AI-assisted and reviewed by a person before anything sends; Reveal does not send email on its own. For an agency, that is the right default.
Plan capacity by client sites, not seats
Reveal plans are counted in client companies per month and sites, not in agency seats. Free covers 100 companies on one site with 7-day history; Growth is $129/month for 600 companies, 2 sites and 90-day history; Scale is $299/month for 2,000 companies, 5 sites and 365 days; Enterprise starts at $799/month. Starter and Growth include a 14-day trial.
![]()
Sites are the constraint. A ten-client roster does not fit a five-site plan, so count client domains before you pitch, and count history length too: 90 days answers a quarterly review, while “who visited before the campaign” needs a year.
FAQ
Can each client see only their own workspace?
Yes if you scope it that way: one workspace per client, with its own sites, segments and reporting. Share reports and exports per workspace instead of handing over the agency login.
Do we need a separate subscription for every client?
No. Workspaces sit under one agency account and one bill. What grows with clients is company volume and site count, which is what the plans meter.
How do we stop tracking someone who opted out?
Log the request, add the person to the workspace suppression list the same day, and confirm before the next export; the public opt-out page keeps requests in one place.
If you want to see the split working, create a workspace per client at pryntid.com. Free covers 100 companies a month on one site; paid plans add sites, volume and history. Set up two workspaces before you move the whole roster.
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.