Most people who will eventually buy from your product-led growth company never touch your product first. They read your pricing page, open two competitor tabs, skim your integrations doc, and leave without starting a trial or filling out a form. Product-qualified leads and marketing-qualified leads are both built to miss this exact person, because both require an action she hasn't taken yet. Website visitor identification catches her before she acts at all, and for most PLG teams, that gap is bigger than they realize.

What Is the PLG Blind Spot in Website Visitor Identification?

PLG companies build two well-instrumented funnels. Product usage data feeds PQL scoring: logins, feature adoption, seat expansion, all tied to an account that already exists. Gated content feeds MQL scoring: ebooks, webinars, demo requests, all tied to a form someone already filled out. Both systems are good at what they do. Both are also downstream of a commitment the visitor hasn't made yet.

The researcher who reads your pricing page three times in a week, checks whether you integrate with her stack, and compares you against two competitors by name, but never signs up and never fills out a form, is invisible to both systems. She has real buying intent and zero record in your CRM. Website visitor identification is the only layer built to sit upstream of the signup, which is exactly where this population lives.

Why PQL and MQL Are Both Structurally Blind to Pre-Signup Intent

It helps to be precise about why, not just that, these two systems miss her.

A PQL requires a product account. Someone has to create a login before any usage event can fire, so a PQL is definitionally a person who has already converted at least once.

An MQL requires a completed form. Someone has to trade an email address for a whitepaper, webinar seat, or demo slot, so an MQL is definitionally a person who has already taken a deliberate action to be tracked.

Anonymous browsing on your marketing site happens entirely before either of those events. No matter how well you tune PQL scoring or MQL lead scoring, you cannot score a person your systems have never recorded. That's not a tuning problem. It's a coverage gap, and it's the specific gap website visitor identification is designed to close.

What This Segment Actually Looks Like in Your Traffic

Once you're looking for it, the pre-signup researcher pattern is recognizable. The strongest versions combine several of these signals from the same account inside a short window:

• Repeat visits to your pricing page with no trial ever started
• Multiple people from the same company visiting in the same week, a buying committee forming before anyone has an account
• Visits to competitor comparison or "vs." pages
• Visits to docs or integration pages from a company with no product account
• A visitor who returns days later without ever converting on a form or trial

Any one of these on its own is weak evidence. Two or three from the same account in the same window is a real signal worth acting on.

How to Prioritize These Signals Without Drowning Sales in Noise

Not every anonymous pricing-page visit deserves a sales touch. Treat this like any other lead-scoring problem and tier it before anyone acts on it:

• 🔴 Very High - Buying committee on the pricing page, no trial started - two or more named contacts from one account hit pricing in the same week and nobody has an account yet. This is the strongest signal in the whole framework.
• 🟠 High - Repeat pricing visits from one contact, no trial started - the same person returns to pricing three or more times without converting. Individual intent that's clearly building toward a decision.
• 🟡 Medium - Competitor or "vs." page visit - evaluation is underway, but you don't yet know how far along.
• 🟢 Low-Medium - Docs or integration page visit, no account - often an engineer scoping technical feasibility before a champion ever signs up. Worth logging, not yet worth a call.
• ⚪ Low - A single pricing-page visit with no repeat behavior - could be a casual browser, a competitor doing research, or genuine early interest. Not enough pattern to act on yet.

The Playbook: What to Do When You Catch One

1. Set the acting bar above your MQL bar, not below it. There's no form fill confirming intent here, so only Very High and High tier accounts should generate a sales touch to start. Widen the funnel once you've validated conversion at the top two tiers, not before.
2. Route to a human with real context, not a generic sequence. The outreach that works references the specific page pattern (the comparison she read, the integration she checked), not a vague "noticed you were on our site" line. Generic outreach on this segment converts worse than no outreach at all, because it signals you're guessing.
3. Offer a guided path, not a push into standard self-serve. Someone who read your pricing page three times and still didn't start a trial has already shown some hesitation about self-serve. A short, specific offer, like a scoped walkthrough of the exact feature she was evaluating, usually converts better than nudging her back into the same self-serve flow she already passed on.
4. Keep self-serve conversion protected. Limit sales-assisted outreach to the top two tiers only. Intercepting every anonymous pricing-page visit interrupts people who would have converted through self-serve on their own, and that's a real cost, not a free action.

Where Website Visitor Identification Fits, and Where It Doesn't

This entire framework depends on being able to identify who's behind an anonymous session, at the company level and, where possible, the person level. Knock2 typically identifies 93%* of engaged sessions at the company level, and 62%* at the person level (name, email, title) for US traffic. Person-level identification works by matching an engaged visitor session against an identity graph built from a consent-based publisher network. It's not a replacement for a signup and it won't catch everyone, but it surfaces real intent that would otherwise never reach a CRM at all.

*Identification rates measured against engaged sessions. Results may vary by traffic profile, geography, and industry.

One practical caveat for PLG teams specifically: a lot of PLG buyers are remote-first or work at small, distributed companies, and person-level match rates run lower for that traffic profile than the blended average. Company-level identification alone is still often enough to trigger the Very High tier play, since a buying committee showing up from one company domain doesn't require knowing every individual's name to be actionable.

How to Know If the Program Is Working

Track pipeline conversion from Very High and High tier accounts against your existing self-serve trial conversion rate as the baseline, not against your overall MQL conversion rate, since those two populations start from a different place. If identified pre-signup accounts aren't converting to pipeline within a reasonable window, the three-branch diagnostic for why website visitor ID isn't converting (data, routing, or rep execution) applies here too. For attributing this segment's contribution once it starts working, use the same sourced, influenced, and accelerated tiers laid out for visitor ID ROI generally.

For teams building the broader play library this segment sits inside, six plays for turning anonymous traffic into pipeline is the fuller tactical reference; this post is the PLG-specific version of that same idea.

Across Knock2's own customer base, the companies running this kind of pre-signup outreach successfully tend to be small, fast-moving software teams, often in the 20 to 90 employee range, running exactly the kind of self-serve trial and pricing-led signup flow this framework is built for. That's not a coincidence. It's the same profile that's most likely to have a real gap between product usage data and marketing form data, and the least likely to have anyone already watching it.

FAQ

What's the difference between a PQL and an identified anonymous visitor?

A PQL comes from product usage after someone creates an account. An identified visitor comes from browsing behavior before any signup happens. They're sequential, not competing: a pre-signup researcher who eventually starts a trial simply becomes a PQL once she's in the product.

Will reaching out to pre-signup researchers hurt self-serve conversion?

Only if the bar for outreach is set too low. Limit sales-assisted touches to the Very High and High tiers described above, and self-serve stays untouched for everyone else.

Does this play work for PLG companies with small deal sizes?

It works best above a deal size where a sales touch is worth the cost of a rep's time. Below that, company-level identification alone, without a 1:1 outreach layer, is usually enough to justify the effort.

How is person-level identification different from tracking logged-in product usage?

Product usage tracking requires an existing account and only sees what a logged-in user does inside the product. Person-level website visitor identification matches an anonymous, logged-out session against an identity graph, so it can surface a real person before any account exists.

What if we don't have a documented buying committee for these accounts yet?

Multiple named contacts visiting the pricing page from the same company in the same week is the buying committee forming. That pattern is the signal, not a prerequisite you need to already have on file.

See which accounts are already reading your pricing page before they've ever started a trial with Knock2's website visitor identification software.