TenantWin

Demand Letter Generator for Wronged Tenants

Visit Website
April 13, 2026 I spent 13 days getting 10 impressions on Google Ads. The problem wasn't my bid, it was my keywords.

I launched TenantWin a few weeks ago — a rules-based tool that tells tenants whether their landlord violated state security deposit return law and generates a demand letter citing the statute. $17 per letter, live in Texas, more states coming.

Today I finally audited the Google Ads campaign I'd been running for almost two weeks. The numbers were grim:

- 13 days live

- 10 impressions total

- 0 clicks

- $0 spent

- Budget: $15/day (capacity I never touched)

My first instinct was "bid too low." I was right that the bid was wrong ($0.01 max CPC — a story for another post), but fixing the bid alone wouldn't have helped. The real problem was my keyword selection.

I had 4 keywords. All exact match. All variations of "security deposit demand letter texas."

The issue: nobody searches for the product they need. They search for the problem they have. A Texas tenant whose landlord kept their deposit doesn't type "security deposit demand letter texas" — they type "landlord didn't return my deposit" or "how long does landlord have to return deposit texas" or they just rant on Reddit.

I was bidding on solution language. My customers search in problem language.

The fix: I replaced the 4 exact-match solution-language keywords with 15 phrase-match problem-language keywords. Things like:

- "landlord didn't return my deposit"

- "landlord kept my security deposit"

- "how long does landlord have to return deposit"

- "sue landlord for security deposit"

The campaign has been running with the new keywords for about 4 hours. Too early to tell if it worked, but the framing shift feels important enough to share.

The lesson I'm taking away: for early-stage products, audit your keywords by reading your customer's actual complaint threads on Reddit, not by brainstorming what you'd search for if you were looking for your own product. You're the worst possible proxy for your user on day one — you already know what exists.

Anyone else run into the "I was bidding on solution language" trap? Curious if this is a common mistake or if I was just slow to catch it.

3 Comments

  1. 1

    That shift from solution to problem language is the biggest hurdle in search ads. Most people over-optimize for the product name and miss the 90% who are just searching for the law or venting. Since you moved to phrase match are you planning to monitor the search terms report daily for negative keywords to keep the spend efficient?

    1. 1

      Yes, daily for the first two weeks at minimum. The search terms report is the only way to know what 'sue landlord for security deposit' is actually matching to. Already identified a few to exclude (attorney referral queries, commercial lease terms). The goal is to tighten the intent signal before scaling budget.

      1. 1

        Really liked your shift from solution to problem language. That’s usually where most campaigns get unstuck.

        One thing I’ve seen at this exact stage: even when you start getting the right traffic, conversion stays low because the product is still “unknown” in the user’s mind especially for something legal.

        Out of curiosity — once you start seeing impressions/clicks, how are you planning to handle trust on the landing side? Are you thinking testimonials / authority signals or just testing messaging first?

About

Friends kept losing their security deposits because they didn't know state law was on their side. TenantWin tells tenants if they have a case in 30 seconds. $17 for the letter.