How to Build a B2B Lead List That Isn't Garbage
The list is the campaign
People buy scraping tools to get 10,000 contacts, send to all of them, and conclude that cold email does not work. What failed was the list, not the channel.
A usable lead list has four properties: the right company, the right person, a real problem, and a deliverable email address. Get all four and a mediocre email gets replies. Miss any one and perfect copy gets nothing.
Filter one: the company can buy what you sell
The most common mistake is targeting companies too small to have the problem or too large to buy from you.
Ask three questions per company:
- Can they afford the price without a procurement process? (Usually under 50 people, or a budget that already exists.)
- Do they have the problem in a form you can fix in weeks, not months?
- Is there a real person whose job it is to care?
If you are selling a $200 product, companies with 500 people will take six weeks to approve a purchase order. That is not a lead, that is a project.
Filter two: the person owns the outcome
Three roles exist in every sale: the person who feels the problem, the person who approves it, and the person who uses the fix. Cold email should go to the first one, and mention the second one.
For operations problems that is often an ops manager or a founder in small companies. For revenue problems it is a sales lead or founder. For a compliance problem there is no feeling — go straight to the person with budget, and lead with the deadline.
Get the title right before you get the email right. A perfect email to the wrong role gets forwarded and dies.
Filter three: there is a trigger
A trigger is anything that makes the problem worse this month: a new hire, a funding round, a product launch, a new regulation, a public complaint, a bad review about the exact thing you fix.
Without a trigger you are asking someone to care about a background problem. With one, you are interrupting a live one.
Signals worth watching:
- Job postings mentioning the process you automate ("reconciling," "manual reporting," "spreadsheet")
- Review sites: one-star reviews about the thing you solve
- LinkedIn: new head of ops / new head of revenue in the last 60 days
- Company blog or changelog announcing something new
Filter four: the address is real
This is the one that determines whether anything else matters, because a 15% bounce rate can get your domain throttled.
How to check, in order of effort:
- Pattern inference. Most companies use
first@,first.last@, orf.last@. Do not guess in the send; infer the pattern from a known address (a public press contact, a support page) and apply it. - Verification. Use a verifier that checks MX records and mailbox existence. Free tiers handle a few hundred; anything is better than nothing.
- Send to the first 20 as a test. Bounces on your first batch tell you if the source is bad. Stop and re-check before scaling.
Never buy a list where the seller cannot show you the source. Scraped lists from 2019 are mostly dead addresses, and dead addresses are worse than no addresses.
100 beats 2,000
If you send 100 emails with a real trigger and a specific ask to companies that fit, you will get more replies than 2,000 emails with a merge field and a generic pitch. This is not a moral claim; it is arithmetic. Cold email reply rates move almost entirely with relevance.
A workable cadence for a one-person business:
- 20 new prospects researched per day, 60 minutes
- 1 send, 3 follow-ups per prospect
- 100 active prospects in the pipeline at all times
That produces roughly 300 emails a week with real research behind each one. It is enough to fill a pipeline and small enough to keep quality.
The sheet structure
Six columns, no more:
| Column | Why |
|---|---|
| Company | dedupe key |
| Contact name + title | role filter |
| verified | |
| Trigger | the one line you will reference |
| Status | new / sent / follow-up / replied / dead |
| Last touch | so follow-ups do not slip |
Everything else you think you need is a distraction. Notes go in the trigger column.
The mistake to avoid in week one
Sending. The first week should be 400 researched prospects and zero sends. If you start sending while the list is half-built, you will find out the filters were wrong at the worst possible time — after your domain has already been used to send a batch of irrelevant mail.
Build the list, verify the addresses, pick 20, send those, read the replies, then scale.
If your reply rate is under 3% on a verified, triggered list, the problem is usually the offer or the ask, not the volume. The reply rate calculator will tell you which.
Free tool: reply rate calculator — no signup.