If you sell to a broad market, say “B2B SaaS companies with 50+ employees,” lead sourcing is mostly a filtering exercise. You open Apollo or ZoomInfo, set your parameters, and you’ve got thousands of results. The hard part is further down the funnel.

But if you sell to a niche, like privacy consultancies or fintech compliance teams, the sourcing itself is the bottleneck. The standard databases don’t have great coverage. The filters don’t capture what makes your prospect different. And you end up doing a lot of manual research to fill a list that, in a broader market, would take ten minutes.

This post is about how we approach that problem with BDR², and some patterns that work no matter what tools you use.

Why niches are hard

The core issue is that niches are defined by attributes that don’t show up in standard firmographic data. “Has an outbound sales team” is not a field in any database. “Recently hired a DPO” is not a filter you can set. “Sells to regulated industries and cares about compliance” requires actually reading the company’s website.

So you can’t just filter your way to a good list. You need to evaluate.

The seed-and-expand approach

Here’s the pattern that works well for niche targeting:

1. Start with a seed list. You probably know 5–10 companies that are exactly what you’re looking for. Maybe they’re existing customers, or companies you’ve spoken to at events, or ones you’ve bookmarked. Write them down. This is your seed.

2. Define what makes them good. Look at your seed list and ask: what do these companies have in common? Not at the firmographic level (“they’re all in London”) but at the qualification level:

• Do they all have a specific function or team?
• Are they all in a specific regulatory environment?
• Do they all sell to a specific type of buyer?
• Is there something on their website that signals relevance?

Turn these into explicit qualifier questions. You want 3-5 yes/no questions where passing all of them is a strong signal that a company is a good fit.

3. Test your qualifiers against the seed. Run your seed companies through the qualifier questions. If your qualifiers correctly identify your seed companies as qualified, you’re on the right track. If a known-good company fails a qualifier, that qualifier might be too strict, or you might be missing a nuance.

4. Expand from the seed. Now use your qualified seed to find similar companies. This can happen through:

• Data providers that support similarity or lookalike searches
• Industry directories and association member lists
• Competitor customer lists
• Job board searches (companies hiring for specific roles)
• Conference attendee lists

5. Qualify the expanded list. Run new candidates through the same qualifier questions. This is where having explicit qualifiers pays off. Instead of gut-checking “does this feel right?” you’re answering specific questions.

6. Iterate. As you accept and reject leads from the expanded list, you’ll notice patterns. Maybe a qualifier you thought was essential doesn’t actually matter. Maybe you need to add one you didn’t think of. Adjust and re-evaluate.

A real example

We recently ran a targeting exercise for a campaign aimed at companies that had hired a new DPO (Data Protection Officer) in the last 90 days. The thinking was that newly hired DPOs are actively reviewing their company’s data practices and more open to tooling that supports compliance.

The qualifiers:

• New DPO hire in the last 90 days
• EU/UK/IE company
• B2B company
• Has outbound sales

We sourced 10 initial leads using a prompt-based search and then triaged them against the qualifiers. Leads that passed all four were accepted; the rest were rejected.

A few things we learned:

• The “new DPO hire” qualifier was the hardest to evaluate and had the most uncertainty. Some leads came back as “undetermined” because the information wasn’t publicly available.
• For accepted leads, enrichment sometimes picked the wrong contact — a reachable exec or sales lead rather than the DPO. We found that role-based mailboxes (dpo@, privacy@) were common and sometimes more appropriate than personal contacts.
• The feedback loop worked: after the first batch, we refined how we phrased the DPO qualifier and added a step to check for role-based privacy contacts.

What makes this different from just… researching?

Manual research works. The question is whether it scales and whether it’s repeatable.

When you do this manually, the knowledge of what makes a good lead lives in the researcher’s head. When they leave, or when you want to hand off list-building to someone else, the quality drops. Writing down your qualifiers gets that knowledge out of someone’s head and into a shared definition that anyone (or any system) can apply consistently.

It also creates an evidence trail. For each lead, you know which questions were asked, what the answers were, and why the lead was accepted or rejected. That matters for compliance (see our post on Legitimate Interest Assessments) and for learning. You can look back at your qualification logic and see where it was right and where it was wrong.

Tooling

BDR² was built around this workflow. Qualifiers are a core concept: you define them, the system evaluates them against each lead, and the results (with evidence) are saved. The seed-and-expand loop, the accept/reject decisions, and the feedback cycle are all built into how it works.

But the pattern itself doesn’t require BDR². If you’re working in a niche, the approach is the same no matter what you use:

1. Start with what you know (seed).
2. Make your qualification criteria explicit (qualifiers).
3. Test, expand, and iterate.
4. Keep a record of your decisions and reasoning.

The tooling just determines how much of this you can do before you run out of hours in the day.