Stop fixing your trial. Fix who's in it.
The ultimate AI marketing tech stack for 2026
The best PLG teams don’t win based on how many tools they use, they win when they use tools that help them turn signals into action.
Recently, Scott Strand, Knock’s agent-pilled growth lead, shared the AI marketing tech stack he uses to run marketing as a one-person team, spanning analytics, lifecycle, search, content, creative, and more.
For product-led growth specifically, the foundation looks like this:
Claude and Cursor to move from idea to execution faster.
PostHog and Hightouch to understand product behavior, run experiments, and activate warehouse and product data across your stack.
Knock to orchestrate lifecycle messaging across email, in-app, push, chat, and more.
Check out the full post to see all 15 tools in Scott’s AI marketing stack and learn how he uses them together to automate the marketing lifecycle.
Founders show me their signup graphs a lot. They’re going up and to the right, and they want to talk about the trial experience. Nine times out of ten, that’s the wrong conversation. The trial isn’t the problem. The problem happened upstream, before the person ever clicked sign up, and no amount of onboarding polish is going to fix it.
A few hundred signups a month. Somewhere between 2 and 5% converting to paid. The founder assumes the fix is activation: better onboarding, smarter nudges, a cleaner first-run experience. So that’s where the effort goes. Meanwhile the real story is that most of those signups were curious olo developers having a poke around, students, competitors doing research, or people who genuinely didn’t understand what the product does. You could build the best trial experience in the world and it wouldn’t convert them, because they were never going to buy anything from anyone.
I get why this is easy to miss. Signups are simple to measure and simple to move. A better landing page, a sharper ad, some SEO work, and the number goes up. It feels like progress because it is progress, on the metric you’re looking at. Fixing the trial also feels like a task you can start right away. Figuring out who your actual buyer is doesn’t feel like a task at all. It feels like a much bigger question, so it gets buried under the carpet and deferred.
There’s also a quieter reason this gets missed: almost everything written about trial conversion assumes your signups are already qualified. Read enough PLG playbooks and trial-optimisation posts and you’ll notice they all skip the step where someone checks whether the person converting was ever a real prospect. If your signups aren’t qualified, that advice isn’t wrong, it’s just answering a different question than the one you have. You can A/B test your onboarding flow for a year and never touch the actual constraint.
So how do you actually find out what’s going on? Don’t run another dashboard query. Dashboards will happily tell you conversion is 3% without ever telling you why. Go and look at the people instead. And, dare I say it!?... talk with some of them.
Pull two weeks of signups and go through them by hand, one at a time. Reach out to them and try to talk with a bunch. I know this doesn’t scale, and that’s fine, because it isn’t meant to. If you’re still counting signups in the hundreds rather than the thousands, you can afford two weeks of manual review, and you can’t afford not to. For each one, ask a simple question: does this person look like they could ever, in any world, become a paying customer?
Sort them into three buckets. Real potential buyer: right role, right size of company, right problem. Adjacent: right role at the wrong kind of company, or the reverse. And not a buyer at all: student, competitor, someone browsing. That ratio tells you which game you’re actually playing. If it turns out only 15% of your signups are real potential buyers, and a fifth of those convert, your overall conversion rate is going to look terrible. But the trial might be working exactly as it should. The leak isn’t in the pipes. It’s in the roof.
Once you know that, here’s where I’d put the effort.
Fix targeting before you touch the trial. It’s cheaper, it’s faster, and it moves the number that actually matters. Tighten the landing page copy so it repels the wrong people instead of just attracting more of everyone. Cut the channels bringing in noise. If you have a free tier, use the gating to filter for intent rather than just to hold back features.
Add a qualifying question at signup, but keep it light. Not a form, forms kill conversion. Something in-product: “what are you trying to do?” with a handful of options. The answer tells you the persona immediately, and you don’t need a lead-scoring model to use it.
Stop designing the trial around the average signup. The average person in most trials was never going to pay. Design it for the qualified slice instead, because their experience is the one that actually determines your revenue.
Change what the team looks at. Report qualified signups next to raw signups, not instead of them. If your CEO can’t ask “how many real buyers signed up this week” and get a straight answer, everyone downstream is optimising a number that doesn’t mean anything.
None of this is a knock on activation work. Good onboarding matters, eventually. But it only matters once you know the people going through it were ever going to buy. Fix that first, or you’re just getting better at converting the wrong crowd, faster. Most PLG advice treats the funnel as sacred and goes hunting for the leak somewhere in the middle. Don’t assume that. Half the time the leak is at the top, before anyone’s even in the funnel proper, and no amount of tinkering downstream will find it.
Signups going up and to the right feels like momentum, but sometimes it’s just more of the wrong people finding you sooner.
Book a free 1:1 consultation call with me - I keep a handful of slots open each week for founders and product growth leaders to explore working together and get some free advice along the way. Book a call.
View your free public Dev Tool AI Market Presence Report - 500+ dev tools across 47+ verticals and growing, all in the Dev Tool AI Search Landscape.
Sponsor this newsletter - Reach over 11,000 founders, leaders and operators working in product and growth at some of the world’s best tech companies including Paypal, Adobe, Canva, Miro, Amplitude, Google, Meta, Tailscale, Twilio and Salesforce.
Show some love to today’s sponsor - Knock