The Keyword Gate: A $737 Lesson in Not Wasting Ad Spend

Two weeks ago I wrote about running 20 agents that shipped 10 landing pages in a week. The output was impressive. The signal was not.

This week I want to talk about the thing that actually started working: a filter so simple it feels almost embarrassing to write about. We call it the keyword gate.

The Problem We Were Solving

LeanAI Studio runs 16 micro-SaaS validation bets in parallel. Each bet gets a landing page, a Google Ads campaign, and a waitlist form. The thesis is straightforward: if nobody searches for what you are building, nobody will sign up. If people do search for it, you will know within days.

The problem was that we launched all of them at once. Fourteen campaigns went live. Total ad spend hit $737. Some bets got hundreds of visitors. Others got single-digit impressions. And until last week, we had no systematic way to predict which would be which before spending the money.

What the Keyword Gate Actually Is

The keyword gate is a pre-launch demand check. Before we build a landing page or spend a dollar on ads, we run a simple test.

Step one: draft the 5 to 7 keywords you would actually bid on for this bet. Not aspirational keywords. The exact terms you would put into a Google Ads campaign.

Step two: run each keyword through 12 months of Google Trends data for the US market.

Step three: count how many keywords show a stable or rising trend. If at least 3 out of 6 pass, the bet gets a green light. If fewer than 3 pass, the bet is killed. No landing page. No campaign. No budget.

That is it. No complex scoring model. No multi-week research phase. A binary pass/fail gate that takes about ten minutes to run.

The Data That Made Us Believe It

We ran the keyword gate retroactively on our Wave 3 bets, six products that had already been launched with live ads. The results lined up almost perfectly with real-world performance.

PropReport scored 4 out of 6 keywords passing. "Commercial real estate reporting" showed a rising trend with strong average interest. Within days of launching, PropReport became our first post-fix conversion at a customer acquisition cost of $19.24. It is now our lead horse.

COIBot scored 3 out of 6. "Insurance certificate compliance" showed rising interest. The campaign ramped steadily and started converting.

On the other side: PatchBook scored 2 out of 6. It generated 1 ad impression in two days. MSPDash scored 2 out of 6. It struggled to gain any traction. SubComply scored 2 out of 6, with most keywords showing zero search volume. HCPSync also scored 2 out of 6 and could not gain meaningful traffic.

Every bet that passed the gate ramped and showed conversion signals. Every bet that failed the gate struggled or flatlined. The correlation was not subtle.

Why This Is Not Obvious

Most validation advice tells you to talk to potential customers, survey the market, analyze competitors, or build an MVP and see what happens. All of that is useful but slow.

The keyword gate tests something different: is anyone actively searching for a solution to this problem right now? Not "would they buy it if you showed it to them." Not "is this market growing." The specific question: are real humans typing these words into Google today?

The conventional wisdom in SaaS marketing says you need $3,000 to $5,000 per month in ad spend to get meaningful data from Google Ads. That is true if you are optimizing a campaign. It is not true if you are just asking whether demand exists. Google Trends data is free. You do not need to spend anything to check if anyone is searching for what you want to build.

We spent $737 learning this the expensive way. Now we check before we spend.

What Changed After We Implemented It

The keyword gate is now a hard pre-launch requirement at LeanAI Studio. No bet moves past validation into LP creation without passing it. Our Marketing Campaigns Manager agent runs the gate automatically as the last step before requesting a landing page build.

The practical impact: we stop building landing pages for bets where nobody is searching for the solution. We stop spending ad budget on campaigns that will never generate impressions. The agents that used to build everything fast now build only the things that have a chance.

In Week 1, we optimized for throughput. Ship everything, measure later. In Week 2, we started measuring. In Week 3, we are finally filtering. The keyword gate is the difference between running fast and running in the right direction.

The Bigger Lesson

Every startup has a version of this problem. You build something nobody wants, and you find out after months of work and thousands of dollars. The standard advice is to validate early. But "validate early" is vague. It does not tell you what to test or when to test it.

The keyword gate is specific. Test search demand before you build. Test it with the actual keywords you would bid on. Set a clear pass/fail threshold. And have the discipline to kill bets that fail, even when you are excited about the idea.

I killed four bets this week based on keyword gate data. Every one of them had a compelling competitive analysis, a clear ICP, and a well-built landing page. None of them had search demand. The exciting part of my brain wanted to give them more time. The data said no.

The data was right.

What Comes Next

PropReport is our lead horse. It passed the keyword gate with the highest score in our portfolio, it converted first, and it has the lowest customer acquisition cost. This week we are increasing its ad budget, prioritizing LP optimization, and watching whether the early signal holds.

The keyword gate will keep running on every new bet that enters the pipeline. If a bet cannot pass a ten-minute search demand check, it does not deserve a landing page. That is the rule now. And so far, the rule is working.