What Leadscraper Is — and What It Isn't

Before you start working with Leadscraper, it's worth taking a clear look at what's actually going on here. Leadscraper works differently from most tools you know in the B2B world. Once you understand why, you'll get significantly more out of it — and you'll also understand why early results get better and better over time.

In short: Leadscraper is not a database subscription and not a ready-made list. It's a specialized research agent that thinks fresh for every search, researches live, and evaluates every single result for you.

What Leadscraper Is Not

To get expectations right from the start, here's the honest distinction.

Not a database scraper. Tools like Apollo, ZoomInfo, or Cognism work with a massive pre-collected database. You filter through millions of static records that were crawled at some point — often months ago. Leadscraper has no such stock. Every search is a fresh investigation in the current web.

Not a purchased lead list. You don't get a bought address list that hundreds of other people are also receiving in parallel. Every run is tailored to your specific request and runs fresh.

Not a generic "AI search tool". This isn't about "an AI doing a quick internet search". Behind every run is a structured sequence: first a plan, then the search, then an individual evaluation of every result.

Not a rigid filter tool. You don't click through dropdowns and checkboxes. You describe in plain words who you're looking for — and the system translates that into a search strategy.

What Leadscraper Is: Three Pillars

1. A Researcher That Thinks Before It Searches

When you send a request, Leadscraper doesn't immediately start a random web search. Instead, the system first plans a strategy.

Concretely, every run goes through several phases:

1. Planning — The system reads your request (company type, region, contact) and develops a plan: what's the goal of the search? Which strategy fits? Which industry directories, chambers, or association registers are the right sources? Which search terms lead to the goal?
2. Search — Only then does the actual research begin: open web, maps, and the planned directories.
3. Enrichment — Found companies are visited: website, legal notice, contact pages, team pages. Structured fields are extracted — company, legal form, location, contact person, contact routes.
4. Evaluation — Every single candidate is checked against your request.

What's special: you watch this thought process live. While a run is happening, steps appear in the activity stream like:

• "Planning search strategy…"
• "Goal: [specific goal of the search]"
• "Strategy: [chosen approach]"
• "Reasoning: [why the system is searching this way]"
• "⭐ Curated Source: [e.g., an industry association]"
• "💡 Own Suggestion: [independently found source]"

This makes it traceable why the search runs the way it does — no black-box feeling.

2. It Learns With Your Feedback

Leadscraper gets better with every rating you give. After every run you see your results and rate them:

• Accepting is implicit — moving a lead further through the pipeline (finding contacts, opening the acquisition plan) signals it fits.
• "Not a fit" — with a structured reason (e.g., wrong industry, wrong role, wrong region, wrong size, not an operating company, no buying signal, already known, poor data) or a short comment. Ratings can be reset at any time.

This feedback is saved and flows into the next searches. Patterns emerge: what types of companies do you consistently move forward, which do you reject? These patterns then guide the reasoning in the next run.

The learning effect is staged:

What this means for you: the first runs are the training phase. Those who rate consistently at the beginning will have noticeably sharper results afterwards. More on this in "How Leadscraper Learns From Your Feedback".

3. Exactly the Right Leads — Not As Many As Possible

Leadscraper delivers up to 10 leads per search. That's intentional. The idea isn't to flood you with hundreds of addresses, most of which don't fit. Instead, every one of these results is individually checked before it reaches you.

The evaluation runs per lead through its own reasoning. The result is one of several classifications:

• Match — clearly fitting, is delivered.
• Possible match — probably fits, but not 100% confirmed. Such leads get a brief note (caveat) explaining what's still uncertain — so you can assess yourself.
• No match — filtered out and not shown at all.

Every delivered lead also comes with a brief justification of why it fits, plus details like industry, location, estimated headcount, and contacts. Quality over quantity.

Where the Data Comes From

Leadscraper accesses three types of sources — live, fresh with every run:

• The open web. A Google-based search finds current candidates for your request.
• German industry directories and association registers. This is where the real depth lies: curated sources like industry-specific associations, chambers, guilds, and directories. Exactly what a generic database doesn't have.
• The companies' own websites. Legal notices, about, contact, and team pages are analyzed to find contacts and contact details. The commercial register is also consulted for company validation.

Because every run is a genuine live investigation, the data is as fresh as the web at the moment of the search — nothing pre-stored, nothing outdated.

How to Get the Best Out of It

The success principle is: start broad, train, then specialize.

1. Start broad. Begin with a request that isn't too narrow. This gives the system enough material to recognize your preferences.
2. Rate. Give consistent feedback on the first runs — reject non-fits with "Not a fit" + a reason, and move matching leads forward. That's exactly what Leadscraper learns from.
3. Specialize. With each run you can refine your request (e.g., "folding carton manufacturers in NRW, purchaser"). This refinement is interpreted at runtime and flows directly into the next run — without permanently changing your base profile.

Optionally, you can also add your own website in the first step. Leadscraper then analyzes it and better understands what you offer and who it's for — that helps with assessing why a lead fits you.

Frequently Asked Questions

Do I get the same companies if I repeat the same search?
No. Already-delivered companies are remembered and skipped in subsequent runs, so you don't get duplicates. Every run brings new candidates.

How many leads does a search deliver?
Up to 10 per search — each individually checked.

Why are the first results sometimes not perfect yet?
Because the system is still getting to know you at the beginning. In the cold start phase, accuracy is typically around 60–70%. With your feedback it rises significantly over the next runs.

Which countries are covered?
The focus is clearly on the DACH region: Germany, Austria, and Switzerland. That's also where the strength of the specialized association and industry sources lies.

Can I export my results?
Yes — leads you've accepted can be downloaded as CSV or Excel. Details are in "Exporting Leads".

What happens with my feedback and my data?
Your preferences and ratings are assigned to your account and can be deleted on request. How that's regulated in detail is explained in the data protection article.

Summary: Leadscraper is a specialized lead researcher that thinks fresh for every search, researches live in the DACH web and industry sources, evaluates every result individually, and gets more precise with your feedback. No database, no ready-made list — research that fits you and gets better with you.

Updated on: 23/06/2026