I Scored 12,000 B2B Leads This Year — 91% Were Never Going to Buy

P.S. If you want more B2B outbound and lead-generation breakdowns like this one delivered to your inbox every week, join 1,000+ founders getting them free: https://wuki.beehiiv.com/subscribe

Twelve thousand leads. That's how many companies my outbound pipeline touched in the last 12 months. I scored every one of them: firmographics, technographics, intent signals, engagement history, the works.

1,080 of them were ever going to buy. The other 10,920 were a donation of my time.

I didn't figure this out on month one or month three. I figured it out by obsessively tracking what happened to every single lead through the funnel — from first touch to closed deal — and then going back to find out which pre-contact signals actually separated the 9% from the 91%.

Here's what the data said, what I changed, and what I'd tell anyone building a B2B product right now.

The setup

I run a small B2B SaaS. Like most solo founders, I started the way everyone tells you to: build a list, write a personalized email, follow up forever, never give up.

The tooling I used for lead generation was originally my own scraper — a pile of Python scripts pulling company data from public sources, enriched with whatever I could get my hands on. It worked, in the sense that the CSV exports were enormous.

But here's the thing nobody tells you about enormous lists: they're mostly noise, and the noise has a cost. Every email I sent to the 91% cost me reputation, deliverability, and hours I could have spent talking to real prospects.

What I scored

For every lead I captured, I recorded 14 signals before any outreach:

1. Company size and funding stage
2. Industry and sub-industry
3. Tech stack (via public sources)
4. Recent hiring (open roles, new roles filled)
5. Job changes in the target role
6. Recent funding announcements
7. Website/content changes
8. Search and social activity around the problem space
9. Whether they already used a competitor
10. Whether they had tried solving this before (blog posts, job posts, docs mentioning the problem)
11. Time since the last trigger event
12. Decision-maker count at the company
13. Domain authority and company age
14. Whether the lead came inbound or outbound

Then I waited. And tracked. Replies, meetings, trials, and deals — for 12 months.

The 91% never stood a chance

Here's the brutal breakdown of who never bought:

• 78% of the list had no active trigger. They might have matched my ICP perfectly on paper — right industry, right size, right title. But nothing was happening at those companies. No hiring, no funding, no public activity around the problem. They were ticking boxes, not raising hands.
• 9% had a trigger but no pull. They were doing something, but the problem wasn't on their radar. One email, even a great one, doesn't create a burning need.
• 4% were actively solving it with someone else — and they were happy enough that churn wasn't a topic.

That's 91% with essentially zero purchase intent at the moment of contact. It didn't matter how good my copy was. The math was already done before I hit send.

What actually predicted a buyer

The 9% had three things in common. Not one, not two — all three:

1. A trigger in the last 30–45 days. Funding, a new head of [department] hired, a competing product just deployed, a public complaint about the exact workflow my product replaces. Timing is the single strongest predictor I found. A lead contacted within two weeks of a trigger event converted at roughly 6x the rate of the same lead contacted three months later.

2. Evidence they'd tried to solve this before. Job postings for a role that exists only if the problem is painful. Internal docs mentioning the workflow (public GitHub repos, product changelogs). A "how do we do X" question on Reddit, Stack Overflow, or a Slack transcript. Past behavior is the best predictor of future behavior, and these people had already raised their hand once — just not to me.

3. A proximal buyer. This one hurt. Half my high-intent leads had the right title on paper — the person whose problem I solved. But a third of them couldn't approve a purchase without a second signature, and those deals died in approval limbo. The buyers who moved fast were at companies small enough that the person with the pain was the budget holder.

The killer combination: trigger + prior attempt + proximal buyer. When all three were present, my close rate was 23%. When any one was missing, it collapsed below 3%.

What I'd do differently

I'd stop building lists. I'd build a detection system.

If I could go back 12 months, I wouldn't write another scraper that dumps 40,000 companies into a CSV. I'd write a system that watches for the trigger events — hires, funding, job posts, public problem mentions, competitor churn signals — and hands me a short queue of companies that are already moving.

The difference isn't subtle. My 9% leads averaged 2.4 touches to book a meeting. The 91% averaged 9 touches and mostly ghosted. Same product, same copy, same me — different intent.

I also learned to kill leads faster. I kept a "maybe later" bucket of 3,000 leads and re-scored them monthly. When a trigger fired — funding round, new hire, public mention — they'd pop back into the active queue automatically. About 40% of my closed deals came from that re-engagement loop, not from the original outreach. List building got me the first touch; list re-scoring got me the deals.

The honest part

This level of filtering feels counterintuitive when you're starting out. Every lead-generation guru tells you volume is the game, that buying intent is a myth, that a great offer converts anyone.

The data says the opposite: intent is real, it's measurable, and it's concentrated. In my 12,000-lead sample, 62% of all revenue came from leads with a trigger event in the prior 45 days — a group that was about 11% of the list. Quality isn't a soft skill. It's a filter you can build.

What I built

At some point I got tired of maintaining the detection system myself, so I turned it into a product — clienthunter.ai — that watches for trigger events across companies and scores them the same way I describe above. If you're doing B2B outbound and you've never looked at your own data this way, I'd genuinely love to hear what your close rates look like.

Question for you

If you run outbound for a B2B product: have you ever scored your leads before contact? What's your actual reply-to-conversion rate?

I have 12 months of data and I'm happy to share the full breakdown — ask me anything in the comments.