PQLs convert, MQLs leak: building the right signal
Activation is also where you stop trusting the wrong kind of lead signal. A marketing-qualified lead is a guess based on who someone is and what they downloaded; a product qualified lead is evidence based on what they actually did inside the product. The conversion gap between the two is the difference between a sales motion that works and one that burns time on the wrong conversations.
The field data is consistent. PQL-led programmes convert at roughly 25 to 30 percent, versus a median free-trial conversion of 18.5 percent across undifferentiated programmes (OpenView, 2024). The mechanism is straightforward: a PQL has already shown intent through behaviour, so you are not persuading a stranger — you are removing friction for someone who is already three-quarters of the way there.
How to define a PQL without a data science team. You need two components: a usage threshold (the specific actions that signal genuine engagement) and a fit signal (firmographic or role data that confirms the lead can buy). The usage threshold comes from your aha moment definition — it is the activity pattern of users who subsequently converted, read backwards. The fit signal comes from your sign-up form, enrichment data, or the AI qualifier conversation. A PQL is a lead who has crossed the usage threshold AND matches the fit profile. Start simple: one threshold, one fit signal, one routing rule. You can add dimensions after you have seen the first cohort convert.
Trial structure changes the picture. Opt-in free trials — no card required — convert at lower headline rates, often in the high single digits, but activate a larger raw volume of leads that your PQL system then qualifies. Credit-card-required trials convert a higher percentage of a smaller, pre-filtered pool. Which you choose depends on your price point, your first-run confidence, and your motion. A product that delivers visible value in under five minutes can usually afford to go card-required. One with a longer time-to-value curve is better served by a no-card trial with a strong PQL routing layer on top. Decide it against your growth model framework, not by copying whoever you admire.
The AI-first PQL operation. A solo founder cannot manually review product usage logs to find PQLs — but an AI agent can. Set up a scheduled agent that queries your usage table daily, surfaces leads who have crossed the PQL threshold, enriches each one with company-size and role data from Clearbit or Apollo, and writes a personalised outreach draft for each. The founder reviews and sends, or the agent sends automatically for the cleanest matches. This is the whole SDR function collapsed into a nightly script, which is how a lean operation beats a thirty-person sales team on precision even when it cannot beat them on volume.