What 4,100 Cold Emails Taught Me About Selling Data Products

I spent three months building a property-data platform — 20 Texas counties, 800,000+ records, five signal types cross-referenced against 8.8 million free appraisal parcels. Then I spent three weeks emailing about it.

Here's the honest scorecard.

The setup

Texas Signals is a distressed-property data platform. Pre-foreclosures, tax delinquencies, code violations, cash buyers, permits — layered together, cross-referenced against county appraisal rolls, searchable by address or ZIP or signal type. The thesis is straightforward: public data that's technically free but practically locked up, cleaned and served through a real interface. Investors and agents pay monthly to use it.

The product was built. The data was good (A-grade across most tables, with honest gaps documented). What we needed was customers. So I did what every SaaS playbook says to do first: cold email.

Campaign 1: Construction companies

Target: Texas construction firms — general contractors, specialty trades, commercial builders. List source: Business directories and state license databases. Result: 9.1% bounce rate. Dead on arrival.

The list was bad. Not "poorly targeted" bad — wrong data bad. The email addresses we scraped from business directories were stale, misspelled, or associated with businesses that had moved or closed. A 9% bounce rate means your sender reputation takes a hit before the first real human even sees your email.

We killed this campaign after the first batch. The bounce rate alone would have degraded deliverability for every other campaign sharing the same sending infrastructure.

Lesson: List quality isn't a nice-to-have, it's the load-bearing wall. A brilliant email to a bad list doesn't underperform — it actively damages your ability to send anything else.

Campaign 2: Wholesalers

Target: Real estate wholesalers in Texas metros. List source: Scraped from investor forums, meetup groups, and public filings. Result: 12.7% bounce rate. Worse than construction.

Wholesalers are a transient population. They come and go. A lot of the email addresses we found were for people who'd been active a year ago and had since moved on to something else — or were using throwaway addresses for their wholesale business in the first place. The bounce rate was catastrophic.

This one hurt because the audience should have been perfect. Wholesalers need exactly what Texas Signals provides: distressed properties with addresses, values, and owner information. The product-market fit was there on paper. But you can't sell to an audience you can't reach, and the available email data for Texas wholesalers turned out to be garbage.

Lesson: The right audience with the wrong contact data produces the same result as the wrong audience entirely.

Campaign 3: Real estate agents

Target: Licensed Texas real estate agents — verified emails from state licensing databases (TREC). List source: Texas Real Estate Commission public records. Result: 1,362 emails sent. 23.4% open rate. 0.44% bounce rate. Zero human replies.

This was the clean campaign. The list was excellent — real, verified, current email addresses from a government database. The bounce rate proves it. The open rate was above industry average. People saw the email. Some of them read it.

Nobody wrote back. Not to say yes, not to say no, not to say stop emailing me. Complete silence.

Here's what I think happened: real estate agents are the most cold-emailed profession in America. They hear from lenders, title companies, coaches, CRMs, lead-gen tools, transaction coordinators, and marketing agencies every single day. My email — no matter how well-written — landed in an inbox that was already hostile to pitches.

But the bigger problem was category. I was selling a data product that most agents don't know they need. An investor understands "pre-foreclosure data cross-referenced against appraisal rolls" immediately — that's their workflow. A traditional agent doesn't wake up thinking about foreclosure data. The email asked them to understand a new category and evaluate my product in it, simultaneously. That's too much for an inbox scan.

Lesson: Cold email sells known solutions to aware buyers. It does not introduce new categories to unaware ones.

Campaign 4: Investors

Target: Texas real estate investors — verified through property records (actual cash purchases). List source: County deed records cross-referenced with our own cash-buyer database, emails verified through MillionVerifier. Result: Two batches running. 351 sent so far. 1.98% bounce on Batch 1, 1.0% on Batch 2. Four replies at ~400 sent — 1.0% reply rate. Too early to call, but it's the only campaign that produced a human response.

The investor list is qualitatively different from the other three. These aren't people we found in a directory — they're people who actually bought property for cash in Texas in the last twelve months, which we know because we have the deed records. The email addresses were individually verified before sending. And the product maps directly to their existing workflow: they already look for distressed properties, they already analyze deals, they already pay for data.

Four replies out of 400 isn't a home run. But it's the first signal in 4,100 emails that a human being read what I wrote and cared enough to respond. That signal came from the only list where the data was first-party and the audience was already buying what I sell.

The numbers, all together

Add the RE Launch campaign (a parallel real-estate-agent outreach with slightly different positioning): 1,762 sent, 0.62% bounce, 49% unique open rate, 0 human replies. Grand total across all campaigns: roughly 4,100 emails, four human replies.

The pivot

Cold email didn't fail because of execution. The infrastructure was clean — warmed domains, SPF/DKIM/DMARC, verified lists, bounce-guard that auto-pauses at 3%. The copy was reasonable. The targeting was thoughtful for three of four campaigns.

It failed because cold email is the wrong channel for what I'm selling.

Here's why I believe that: the people who need this product are already looking for it. An investor hunting for distressed properties in Houston is already searching "Houston pre-foreclosure data" or "Texas tax delinquent properties" or "how to find cash buyers." They have existing intent. They're pulling up county records, scrolling through paid data services, asking in forums.

Cold email interrupts people who aren't looking. Content meets people who are.

So we're building Data Arbitrage — this blog, the newsletter, the playbooks — instead of sending more emails. Every post is a search result waiting to happen. Every real number shared is proof of concept. Every reader who subscribes has already self-selected as someone who cares about this exact niche.

The investor campaign is still running because it's the one list where the audience matches. But the growth strategy has shifted permanently from push to pull.

The takeaway

Cold email works when you have three things aligned: a good list, a known category, and an aware buyer. I had one of those (the investor list) and zero of them for the other three campaigns. The 4,100-email experiment cost about three weeks of time and $20 in list verification. The tuition was cheap. The lesson was clear.

Content lets you find the people who already want this. Cold email tries to convince people who don't. When you're selling a new category — not a better version of something they already buy, but something they haven't thought about yet — pulling beats pushing, every time.

The data's still sitting there. The product still works. The channel was wrong.

If you're building a data product and want to learn from someone else's $20 tuition instead of paying your own, the newsletter is where the real numbers land first.