Product-led growth means the product itself does the work a sales team used to do: it gets a prospect to a working result, shows them the value, and lets them pay before a human ever gets on a call. Done well, it's the cheapest acquisition motion available to a SaaS company. Done badly, it's a free trial with no plan for what happens after signup — which is most PLG implementations I've seen, including ones on teams I've worked with directly.
Free trial and freemium are the acquisition motion, not a feature
The core bet of PLG is that removing the sales conversation from the front of the funnel gets you more qualified prospects than gating access behind a "book a demo" form does. A free trial (time-boxed, full or near-full feature access) and freemium (permanent free tier, capped on usage or features) are the two standard shapes. Trials create urgency but risk losing people who needed more than 14 days to see value; freemium removes the clock but risks supporting a large population of users who will never pay. Neither is universally correct — it depends on how long it genuinely takes a new user to reach a moment of value in your product. A tool that shows its worth in the first session fits a trial. A tool that needs a team to accumulate data over weeks (a CRM, an analytics platform) often does better with freemium, because a 14-day trial ends before the value shows up.
Activation, not signups, is the metric that matters
Signup counts are the easiest number to make go up and the least useful one to optimize for. A signup is someone who typed an email address; it tells you nothing about whether they got value. The metric that actually predicts revenue is activation: the percentage of new signups who reach a specific, product-defined moment that correlates with them sticking around — connecting a data source, inviting a teammate, sending the first message, whatever the equivalent of Slack's "2,000 messages" milestone is for your product. Time-to-value is the companion metric: how long activation takes. A product where activation takes five minutes converts differently than one where it takes five days, and shortening that gap is usually higher-leverage than adding features.
Teams often track five "activation-adjacent" events and average them into a number nobody trusts. Pick the single action that, historically, best separates users who convert and retain from users who churn, and make that the number the whole company watches.
Self-serve checkout removes the friction sales adds
PLG only works end to end if a user can go from trial to paid without talking to anyone. That means pricing that's visible on the website, a checkout flow that takes a credit card without a "contact sales" gate, and an upgrade path that happens inside the product — not a support ticket. Every added step between "I want this" and "I paid for this" is a point where a motivated user gives up or has time to reconsider. The products that do this well put the upgrade prompt at the exact moment a user hits a real limit (a seat cap, a usage cap, a locked feature they just tried to click), not on a generic pricing page they have to go find.
PQLs: scoring usage instead of guessing intent
A product-qualified lead (PQL) is a usage-based alternative to the traditional marketing-qualified lead: instead of scoring someone on form fills and email opens, you score them on what they actually did inside the product — features used, seats invited, usage volume, how close they are to a plan limit. A free-tier account that just invited eight teammates and hit its API rate limit is a far better sales target than one that logged in once and never returned, even if both technically "signed up." The practical use of PQL scoring is routing: high-usage free or trial accounts get a proactive nudge (in-app upgrade prompt, or a real outreach email) instead of waiting for them to convert or churn silently.
It's easy to build a PQL score that sits in a report nobody acts on. The score only earns its keep if it triggers something — an in-app message, a sales handoff, a targeted email — at a defined threshold. If nothing happens when the score crosses the line, you've built an analytics exercise, not a growth motion.
Where PLG stops working: the enterprise handoff
Pure self-serve breaks down once you're selling into accounts that need procurement sign-off, security review, custom contracts, or a champion who has to justify the purchase internally to people who never touched the product. This is where "PLG-only" companies quietly bolt on a sales team anyway — not to replace the product-led motion, but to catch the accounts it can't close by itself. The healthy version of this is sales-assist: the product still drives discovery and the free trial, but once an account crosses a size or usage threshold, a human steps in to handle the parts self-serve genuinely can't — a security questionnaire, a multi-stakeholder rollout, a negotiated contract. Trying to force a 2,000-seat enterprise deal through the same self-serve checkout flow as a five-person startup usually just loses the deal.
The two failure modes that show up most often
The first is optimizing signups instead of activation — running growth experiments that make the top of the funnel bigger without checking whether those new signups ever reach the value moment. Signups go up, activation rate goes down, and the business is no healthier than before, just noisier. The second is gating the wrong features behind the paywall. Gate too aggressively and trial users never get far enough to feel the value that would make them pay; gate too loosely (or gate the wrong thing — say, a cosmetic feature instead of the thing that scales with team size) and there's no reason for anyone to ever upgrade. The fix for both is the same discipline: know what your activation moment is, protect the path to it fiercely, and put the paywall on the axis that grows with the customer's usage, not on an arbitrary feature checklist.
Wrapping up
PLG isn't a marketing trick for skipping sales entirely — it's a bet that a well-instrumented product can do the qualifying and converting that a sales rep used to do, faster and cheaper, for the segment of customers small enough to buy without a procurement process. It lives or dies on activation, not signups, and on knowing exactly which usage signals mean an account is worth a human's attention. The moment an account outgrows self-serve, the honest move is to hand it to a person, not to keep pretending the checkout flow scales to every deal size.
Independent software engineer in Nairobi specialising in Acumatica customisations, Laravel backends, and tax fiscalisation integrations across East and Southern Africa.