The lead filters, and when to use each
Every filter, what it actually tests, which export column it corresponds to, and the order to apply them in so you spend credits on rows worth contacting.
Filters are free and they are the main way to stop paying for rows you would never call. Every one of them is available on every plan, including the trial.
The filter panel, group by group
The panel is built from the filter list the API publishes, so it can never offer you a filter the server would reject. Each one lines up with a column in the 64-column export:
| Filter | What it tests | Export column |
|---|---|---|
| Main activity only | The type you searched for is the listing’s primary type, not one of its secondary ones | Main type |
| Permanently closed | Include, exclude, or return only closed businesses. Excluded by default. | Is closed permanently |
| Has a website | A website URL is present | Website |
| Has a phone number | Any phone number is present | Phone |
| Phone type | Mobile, Fixed line or Other | Phone type |
| Has an e-mail on the website | At least one address was found on the business’s own site | Email |
| Has a contact page | A contact page was found on that site | Contact page 1 |
| Reviews at least / at most | The review-count band falls inside a range | Reviews count |
| Rating at least / at most | Star rating falls inside a min–max range | Reviews rating |
| Facebook, Instagram, LinkedIn, Twitter, Youtube, Tiktok link | A profile URL for that network is present — one filter each | the matching … link column |
| Runs advertising pixels | At least one ad pixel was detected on the site | Website ad pixels |
| Technology contains / Ad pixel contains | Free text matched against what was detected on the site | Website technologies, Website ad pixels |
| First seen on or after / before | When the business entered our index | First seen on |
The website filters depend on enrichment that is still rolling out
Everything read from the map listing — types, phone, reviews, closed status, first seen — filters against data we already hold for every row. The filters that read the business’s own website, particularly Has a contact page, Runs advertising pixels and the two free-text technology filters, are only as complete as the crawl behind them, and that crawl is still being rolled out. Check the match count before you build a list on one of them: if it collapses to near zero, that is coverage, not your market.
Apply them in this order
Filters are free, so the cost of getting the order wrong is only your time. But this sequence gets to a useful list fastest:
- Permanently closed — already excluded. Leave it that way unless you are studying market churn.
- Has an e-mail on the website or Has a phone number, depending on your channel. This is the biggest single cut, and it is the one that most changes what you pay for.
- Reviews at least — a proxy for "this business is actually trading". The
1-50band is a sensible floor in most markets. - Rating — only if your offer depends on it.
- Socials — narrow. Use them when the channel matters, not by default.
Watch the count, not the rows
Each filter updates the match count immediately. The interesting moment is when a filter halves the count — that tells you something real about the market before you have spent anything. See searching is free.
Filters worth understanding properly
Phone type
Stricter than Has a phone number. It tests the classification written into the
Phone type column, so setting it to Mobile returns only businesses
whose listed number is a mobile line. For SMS or WhatsApp outreach that is the setting you
want; for cold calling, Has a phone number is broader and usually better.
Rating
A min–max range, not just a floor. A maximum is genuinely useful: if you sell reputation management, businesses rated 3.0–4.0 are a better list than 4.5+. Businesses with no rating at all are excluded once you set any range — if you want them, leave the rating filter off.
Reviews at least / at most
These are bands, not numbers: 0, 1-50,
51-200, 201-500, 501+. They use exactly the same
vocabulary as the Reviews count column, so the filter and the file can never
disagree — but you cannot ask for "at least 37 reviews", because that is not a distinction the
data makes. Read why Reviews count is a band.
First seen on
A date range over when a business entered our index. It is the closest thing to a "new businesses" filter: set First seen on or after to last month and you get listings that have only recently appeared. It is not a proxy for how recently the row was refreshed.
What filters do not do
They do not unmask anything. Contact details stay masked in the preview whatever you filter on — why previews are masked. And they do not change the credit cost per row: filtering just means fewer rows.
Over the API
The same set arrives as the filters object on POST /api/v1/search.
A null means "do not filter on this"; false genuinely means "only
rows without it". See the search endpoint.
Updated on: