The PQL Audit: Sorting Signal From Noise
Objective: Given a mock 15-account usage export for free-trial users, apply the lesson's PQL criteria to decide which accounts are genuine product-qualified leads ready for a sales-assist conversation, and which are vanity activity that would waste a rep's time.
You're the PMM at Squarespace validating a proposed PQL definition before it gets wired into the sales team's queue. Sales is complaining that half of the 'qualified' leads they're getting are single-page hobby sites that will never upgrade.
Import the export, score each account against the lesson's PQL signals (feature depth, usage ceiling, team invites, active days), and split the list into route-to-sales, nurture-in-product, and not-yet.
Before you start
What you'll need
Free path (everything below is enough to finish)
No account friction, filters and formulas are enough for a 15-row audit
The process
1 step
Step 01 of 01
The lesson defines a PQL as a free user whose in-product behavior already proves they need the paid tier, repeated feature use, an invited team, or a hit usage ceiling, not a lead score built from campaign clicks.
Of 15 free-trial accounts, which ones have actually proven they need Squarespace's paid tier, versus accounts that are merely active but not commercially ready?
Procedure
- Import the export: account, custom_domain_attempts, team_invites_sent, pages_published, days_active_14d, hit_storage_cap.
- Flag any account with team_invites_sent >= 2 OR hit_storage_cap = TRUE as a strong PQL signal.
- Flag accounts with pages_published = 1 and days_active_14d <= 2 as low-intent, regardless of custom_domain_attempts.
- Cross-check custom_domain_attempts >= 1 as a commercial-intent signal only when paired with at least one other signal.
- Sort into three buckets: route-to-sales, nurture-in-product, not-yet.
ROUTE TO SALES (4 accounts) acct_1044 - 3 team invites, hit storage cap, 9 active days acct_1091 - 2 team invites, custom domain attempt, 11 active days ...2 more NURTURE IN PRODUCT (6 accounts) acct_1052 - 1 page, custom domain attempt, but 1 active day only ...5 more NOT YET (5 accounts) acct_1077 - 1 page, 0 invites, 1 active day, no domain attempt
Healthy
Route-to-sales bucket only contains accounts with a team-invite or storage-cap signal, the two hardest signals to fake.
Unhealthy
Treating a single custom-domain click as sufficient to route to sales, that's curiosity, not proof of need.
What this means
A PQL definition has to survive the question 'could this signal happen by accident,' team invites and storage caps can't, a domain click alone can.
So what do I do about it?
| Symptom | Action | Effort |
|---|---|---|
| Sales says half their 'PQLs' are hobby accounts | Require two independent signals, not one, before routing to sales | 30 min |
Final deliverable
A scored 15-account list split into route-to-sales, nurture-in-product, and not-yet, with the specific signal combination that justified each verdict.
See a reference example
Coinbase free-trial PQL audit (excerpt) ROUTE TO SALES acct_2031 - invited 4 teammates, hit API rate cap, 12/14 active days -> strong PQL, route within 24h NURTURE IN PRODUCT acct_2048 - tried custom webhook setup once, only 2 active days -> send in-product tips, re-check in 7 days NOT YET acct_2059 - 0 invites, 1 active day, no advanced feature touched -> leave in trial nurture
Success criteria
You're done when you can:
- Every route-to-sales account has at least two independent PQL signals, not one
- Verdicts distinguish curiosity clicks (single signal) from proven intent (combined signals)