Your lead list is not your campaign

On a recent consult, a team told me their first day of sending had already produced a bunch of invalid-address bounces.

They re-verified the list. A lot of the addresses they thought they could use disappeared.

Then came the obvious question: where do we find enough leads to keep sending?

I would pull a much larger company list first, then make qualification cheap enough to run across the whole thing.

That gives you more of the market. You only pay to enrich the companies that fit a real campaign.

Here is the workflow.

Start with at least ten times the coverage

Start with at least ten times the company coverage you need for the final campaign.

You want enough coverage to find segments you did not think to search for at the start. A small, over-filtered list hides those patterns before you get the chance to see them.

I would start with directories. Then I would use a database like QuickEnrich to fill the gaps.

Directories give you the broad surface area. QuickEnrich helps complete it. Neither source gets to decide who belongs in a campaign.

Write broad inclusion rules first

Take a sample of the raw list and have Claude help you refine the rules.

Start with broad inclusion rules. What would make a company *possibly* relevant?

Then write the exclusions. What clearly makes one a bad fit?

Run those rules against another sample. Look at the false positives and false negatives. Update the rules, then test again.

This is where the agent is useful. It can show you where the rules fail so the next pass gets better.

Use regex before you pay for qualification

You can remove a lot of obvious noise with regex and other simple checks before asking a model to review a company.

Look for the words, page patterns, URLs, and categories that reliably include or exclude a business. Run the cheap rules first. Send the uncertain companies to the more expensive check.

Once the obvious noise is gone, you can qualify companies from their homepages for roughly $6 per 10,000 companies.

That figure covers company qualification from the homepage. Contact enrichment comes later and has its own cost.

Turn the qualified pool into campaigns

Now you can ask better questions about the companies that remain.

• Who is running Meta ads?
• Who is already producing user-generated content (UGC)?
• Who is hiring for a role connected to the problem you solve?
• Which companies share a pattern that deserves its own message?

Each answer can become a segment with a specific reason to write.

A store using UGC in ads but not on its product pages has a specific reason to write.

Expand from the companies that survived

Once you have a group of companies that clearly fit, use Discolike to find more businesses that resemble them.

This gives you one more expansion layer after directories and QuickEnrich. Run the new companies through the same qualification rules before you enrich anyone.

Similar does not automatically mean relevant.

Find people last

Contact discovery comes after the company qualifies for a specific campaign.

Before your next enrichment run, take a sample of the raw list and write down why each company belongs in a specific campaign.

- Mitch

P.S. Something I've been thinking about: have the agent show you where your rules failed, then save those fixes. The next list should be easier to qualify.