How to Find Product Qualified Leads in Your Database

A product qualified lead is not someone who downloaded an ebook and disappeared. It is a user or account that has experienced value inside your product and is showing measurable signs that it may buy, upgrade, or expand.

Those signs already live in your application database: completed onboarding steps, repeated use of a valuable feature, invited teammates, rising usage, plan-limit events, pricing-page visits, and billing questions. The practical job is to turn that scattered activity into one reliable view and act while the signal is fresh.

This guide shows you how to define product qualified leads, find them from live database data, build a score your team can inspect, and route high-intent accounts automatically. You can do it without buying a full product analytics stack or waiting for an analyst.

What is a product qualified lead?

A product qualified lead, or PQL, is a user or account that matches your target customer and has demonstrated buying intent through product usage. OpenView describes PQLs as users who signal intent based on product usage rather than only on marketing or sales qualification.

That makes a PQL different from a marketing qualified lead. An MQL may have opened emails, attended a webinar, or submitted a form. A PQL has used the product, reached a meaningful outcome, and behaved like successful paying customers behave.

Do not treat every active user as qualified. Logging in is weak evidence. A useful PQL definition combines three kinds of signal: customer fit, value received, and intent to buy. ProductLed uses the same fit-value-intent framing because any one category by itself creates false positives.

Why your database is the best PQL source

Your database contains the events that matter after signup, even when those events are split across users, workspaces, subscriptions, and feature tables. It can answer account-level questions that a page-view tool often cannot: How many teammates joined? Did the account connect real data? Did it run the core workflow three times? Is it near a plan limit?

Using the database also keeps the definition tied to the product's actual state. You are not waiting for a nightly spreadsheet export or relying on a tracking event that stopped firing after a frontend release. You can refresh the PQL list from live records and inspect why each account qualified.

Product analytics tools are useful for journey exploration. A database query is better when qualification needs to join behavioral activity with account, billing, entitlement, support, and team data. In practice, many SaaS teams use product analytics to discover candidate signals and the database to run the operational score.

Step 1: Define the value moment before scoring

Start with the action that proves a user received the product's core value. Not signup. Not login. The value moment is the smallest completed outcome that makes the user understand why the product is worth keeping.

For a reporting product, it might be creating a dashboard from real data. For an email tool, it might be sending the first campaign to a real audience. For collaboration software, it might be inviting two teammates and completing a shared task. Pick one primary value moment before adding secondary signals.

Then compare converted accounts with non-converted accounts. Ask which actions occur more often before an upgrade, how quickly they happen, and whether the pattern holds across several cohorts. A signal is useful because it predicts revenue, not because it sounds engaged.

Step 2: Choose fit, value, and intent signals

Fit signals tell you whether the account resembles customers you can serve well. Examples include company size, use case, region, work email, team size, or an integration that indicates a relevant stack. Keep fit criteria broad enough that they do not erase genuine usage evidence.

Value signals show that the account has reached useful outcomes. Examples include completing onboarding, using the core feature on three separate days, importing production data, creating a dashboard, inviting teammates, or returning after the first week.

Intent signals show commercial readiness. Examples include visiting pricing, approaching a usage limit, adding a billing contact, attempting a paid feature, asking about security, or increasing seats. Intent without value can be curiosity. Value without intent may be a happy free user. The combination is what makes outreach timely.

Also define negative signals. Test accounts, internal users, deleted workspaces, repeated errors, bounced emails, and inactive accounts should reduce the score or be excluded. A score that only adds points eventually labels everyone a PQL.

Step 3: Map each signal to database fields

Write each business signal in plain language first. For example: 'The workspace connected production data, created two dashboards, invited at least three teammates, and viewed pricing in the last seven days.' Only then map it to tables and fields.

A typical model joins an accounts table to users, product events, subscriptions, feature usage, and support conversations. Use the account or workspace as the scoring unit for B2B products. One enthusiastic user may be important, but the buying decision usually happens at account level.

Document the source and time window for every rule. 'Created a dashboard' is ambiguous unless you specify whether deleted dashboards count and whether the action must be recent. Clear definitions make the score debuggable and stop sales, product, and growth from arguing over different numbers.

Step 4: Build a simple PQL score

Start with rules your team can explain. Predictive models are tempting, but they are a poor first move when you have limited conversion history. A transparent score gets you operational quickly and creates the labeled data needed for a stronger model later.

Here is a practical starting score: add 20 points for completing the value moment, 15 for using the core feature on three days, 10 for inviting two teammates, 15 for reaching 80% of a plan limit, 20 for visiting pricing twice, and 10 for matching your ideal company profile. Subtract 30 for fourteen days of inactivity and exclude staff or test accounts.

Set an initial threshold, such as 50 points, and show the component scores beside the total. A sales rep should see 'pricing viewed twice, three active users, plan limit at 86%' rather than a mysterious score of 72. Reasons create trust and make the outreach relevant.

Review false positives every week. If qualified accounts repeatedly ignore outreach or never convert, inspect which rule promoted them. If recently upgraded accounts were never flagged, find the missing behavior. Adjust one rule at a time so you know what changed the result.

Step 5: Create a live PQL dashboard

The dashboard should answer four questions: Which accounts qualify now? Why did each qualify? What changed since yesterday? What happened after outreach? Keep the default view short enough that a founder or sales lead can work it every morning.

Useful columns include account, score, qualifying reasons, last active date, active users, current plan, owner, outreach status, and first-qualified timestamp. Add filters for new PQLs, unassigned PQLs, high-value accounts, and accounts that crossed the threshold again after going quiet.

Track the funnel from qualified to contacted, meeting booked, opportunity created, and paid. PQL count alone is not success. The score is useful only if it improves conversion, sales efficiency, or expansion revenue.

Step 6: Trigger action while intent is fresh

A dashboard that nobody checks is decorative. When an account first crosses the threshold, route the event to the person or system that can act. Send a Slack alert to the account owner, create a webhook for your CRM, or email the customer with help relevant to the behavior that qualified them.

Use cooldowns and state changes. Trigger once when an account moves from below the threshold to above it, not every time the dashboard refreshes. Record the trigger time, destination, and outcome so the same account does not receive duplicate outreach.

Match the message to the signal. An account nearing a usage limit needs a capacity conversation. A team that invited five colleagues may need workspace setup help. Generic 'saw you using the product' emails waste the context your database already gave you.

A no-SQL workflow with AI for Database

AI for Database lets you connect PostgreSQL, MySQL, Supabase, MongoDB, BigQuery, and other databases, then describe the PQL definition in plain English. You can ask: 'Show workspaces that completed onboarding, used the core feature on three days, invited at least two users, and viewed pricing this week. Exclude internal and cancelled accounts.'

Save the result as a self-refreshing dashboard so product, growth, and sales see the same qualification logic. Then create an action workflow that sends an email, Slack message, or webhook when a new account crosses the threshold. The query, dashboard, and action stay in one workflow instead of being split across a BI tool, a spreadsheet, and an automation service.

Start with read-only database credentials and validate the first result against a handful of known accounts. Once the logic is correct, schedule the dashboard and enable the alert. You can try the workflow at https://aifordatabase.com.

Common PQL mistakes

Scoring activity instead of value is the most common error. Ten logins may mean the user is confused. Prefer completed outcomes, repeated successful use, collaboration, and expansion pressure.

Another mistake is copying another company's magic number. Your value moment, packaging, customer size, and sales motion are different. Borrow the method, then validate the threshold against your own converted accounts.

Finally, do not hide the score inside one team's tool. Product needs to improve the behaviors that create qualified accounts. Sales needs reasons for outreach. Customer success needs to distinguish a new buying signal from a churn risk signal. One shared definition prevents three competing funnels.

Questions teams ask about product qualified leads

What is the fastest way to identify product qualified leads?

Define one value moment, add two or three buying-intent signals, query those conditions at account level, and review the result manually for two weeks. Do not wait for a predictive model. A small explainable rule set is faster to launch and easier to correct.

Can I find PQLs without a product analytics tool?

Yes. If product usage, accounts, and subscription state are stored in your database, you can identify PQLs directly. Product analytics helps with exploration, but it is not required for a working qualification list, dashboard, or alert.

Should a PQL be a user or an account?

For most B2B SaaS products, score the account and retain user-level reasons. Multiple users, shared usage, billing, and plan limits usually determine purchase intent. For individual self-serve products, user-level scoring may be enough.

How often should a PQL score refresh?

Daily is enough for many low-volume SaaS teams. Refresh hourly or near real time when intent decays quickly, such as a trial ending or a hard usage limit. Match frequency to how fast your team can act.

Measure whether the score creates revenue

Launch with a baseline: free-to-paid conversion, time from qualification to contact, meetings per qualified account, and expansion revenue. Compare qualified accounts with similar non-qualified accounts and review results by cohort, not just as one lifetime average.

The first score will be wrong. That is normal. What matters is that the reasons are visible, the actions are recorded, and the rules can be changed without rebuilding the pipeline. Your database already has the evidence. Turn it into a decision your team can use today.

Sources

OpenView, 'Your Guide to Product Qualified Leads (PQLs)': https://openviewpartners.com/blog/your-guide-to-product-qualified-leads-pqls/

ProductLed, 'How to Define PQLs': https://productled.com/blog/how-to-define-pqls