In this article 6 sections
On 148 real business websites checked by hand, LeadTrapper's email finder found 85.7% of the published emails and 78.2% of the phone numbers, with 2.6% and 4.7% of its results wrong. On 42 Spanish, Portuguese and Russian sites it found 79.5% and 86.6%. No email finder finds everything; here is what it misses and why.
- We recorded by hand what 190 real business websites publish, then ran LeadTrapper's email finder on them and counted what it found, missed and got wrong.
- On 148 websites (Wix, WordPress, Shopify, custom-built, Arabic, German, Turkish and Indian) it found 85.7% of the published emails and 78.2% of the phone numbers; 2.6% of its emails and 4.7% of its phones were wrong.
- On 42 Spanish, Portuguese and Russian websites it found 79.5% of the emails and 86.6% of the phones.
- Before this work, version 2.1.126 found 58.7% of the emails and 43.8% of the phones on the same 148 sites, with 9.3% and 14.5% wrong.
- No finder gets everything: websites that block automatic visitors, sites that were down, and contacts shown only in pictures account for most misses.
- The test runs before every release. A new version may not find less, or get more wrong, by more than 2 points.
Every email finder says it is accurate. Few say how they measured it. We wanted a figure we could defend, so we built a test: real business websites, every contact they publish recorded by hand, and the finder’s results compared line by line. This page publishes the result, including where it falls short.
How the Test Works
The websites. 150 real business websites picked from OpenStreetMap in 8 groups of about 20: sites built on Wix, WordPress and Shopify, custom-built sites, and sites in Arabic, German, Turkish and from India. A second set adds 45 websites from Madrid, Barcelona, Mexico City, Lisbon, Porto, São Paulo and Moscow, 15 each in Spanish, Portuguese and Russian.
The hand check. On 15 and 16 November 2026 we saved a browser copy of each website and its contact, team, about and legal pages, and wrote down every email address and phone number the business itself publishes. Websites that were already gone are left out, which leaves 148 and 42 websites. The two lists hold 349 emails and about 850 phone numbers, and 83 emails and 142 phones.
The run. The email finder in the LeadTrapper Windows app reads the same websites the way a customer runs it: one website at a time, with a pause between them, reading the home page and up to six contact, team, branch and about pages.
The score. For emails and phones separately:
- Found is the share of the published contacts that the finder returned.
- Wrong is the share of the finder’s results that are not the business’s own contact.
The Results by Site Type
Version 2.1.133, run on 17 November 2026:
| Site type | Websites | Emails found | Emails wrong | Phones found | Phones wrong |
|---|---|---|---|---|---|
| Wix | 18 | 87.8% | 0.0% | 87.0% | 16.7% |
| WordPress | 17 | 94.6% | 5.4% | 95.5% | 5.9% |
| Shopify | 16 | 82.2% | 5.1% | 95.9% | 0.7% |
| Custom-built | 17 | 75.0% | 6.2% | 81.2% | 18.8% |
| Arabic | 20 | 88.0% | 0.0% | 87.0% | 11.8% |
| German | 20 | 95.4% | 1.0% | 79.2% | 4.1% |
| Turkish | 20 | 95.2% | 9.1% | 80.0% | 7.0% |
| Indian | 20 | 59.6% | 0.0% | 56.3% | 0.8% |
| All 148 | 85.7% | 2.6% | 78.2% | 4.7% | |
| Portuguese | 14 | 95.8% | 4.2% | 98.0% | 0.0% |
| Russian | 13 | 67.4% | 3.1% | 80.8% | 0.0% |
| Spanish | 15 | 92.3% | 0.0% | 78.6% | 15.4% |
| All 42 | 79.5% | 2.9% | 86.6% | 1.6% |
Small groups move a lot with one website: a single site with twenty branch numbers can shift a group’s phone figure by ten points. The totals are the figures to rely on.
Before and After
The test was built before the finder was improved, so the same websites measure the change:
| Compared | Emails found | Emails wrong | Phones found | Phones wrong |
|---|---|---|---|---|
| 148 websites, version 2.1.126 | 58.7% | 9.3% | 43.8% | 14.5% |
| 148 websites, version 2.1.133 | 85.7% | 2.6% | 78.2% | 4.7% |
| 42 websites, version 2.1.127 | 63.9% | 7.0% | 57.0% | 3.6% |
| 42 websites, version 2.1.133 | 79.5% | 2.9% | 86.6% | 1.6% |
The biggest gains came from finding the right pages (sitemaps, menus and contact pages in more languages), reading store lists that appear only after the page’s scripts run, taking each website’s own country for numbers written without a code, and leaving out contacts that belong to someone else.
The test now runs before every release. A new version may not find less, or get more wrong, by more than 2 points.
What It Misses, and Why
Most misses are contacts the finder could not see:
- Security checks. A few websites show a “Verify you are human” page to automatic visitors. LeadTrapper never gets past one by itself; it marks the website “Security check, open it yourself” and offers to read it in a window where you pass the check. Our guide to websites that block contact finders covers the options.
- Websites that are down. On the day of a run some websites time out, or their address cannot be looked up from the computer running the test.
- Contacts in pictures. A phone number inside a banner or a scanned business card can be misread, or not read at all. See extracting emails and phone numbers from images and PDFs.
The Indian group scores lowest on both emails and phones, at about 60% and 56%. We publish it as it is rather than leave the group out.
Websites also change between the hand check and a run. When a result differs from the list, we open the live page: a new contact the business published is added to the list, and one it removed is not counted against the finder.
When a Result Is Wrong
“Wrong” counts every result that is not the business’s own email or phone, even when it is printed on the page:
- Fax numbers and numbers in a reference or ID.
- Other organisations named in a legal notice: dispute boards, regulators, hosting companies, debt collectors.
- The web designer, credited in the footer.
- Other businesses on a page that lists stockists or partners.
The finder leaves most of these out on purpose. The rest are why every contact in the app also carries how sure it is, High, Medium or Low, with the reasons. The best email and phone for each business are marked, so you can send to those first.
Reading Any Tool's Accuracy Claim
Before trusting a figure, ask four questions:
- Accuracy of what? “95% of emails verified” says the addresses exist, not that the tool found the right person or every address.
- Measured how? A figure from the vendor’s own database is not the same as a hand check against real websites.
- On which websites? Results on large company sites say little about small local businesses, and the reverse.
- Does it claim 100%? No tool finds every contact on every website. A tool that says it does has not measured.
When you want to check a few results yourself, our guides to checking if an email is valid and finding business email addresses show the manual steps. Lists also age: a contact that was right in spring may be gone by autumn, which is why the app can keep a lead list up to date.
Frequently Asked Questions
It depends on the method and the websites. In our hand-checked test of 148 real business websites, LeadTrapper's email finder found 85.7% of the emails the sites publish and 78.2% of the phone numbers, and 2.6% of the emails it returned were not the business's own. Results differ by site type, from 95% of emails on WordPress and German sites to about 60% on Indian sites.
Found is the share of the emails or phone numbers a website publishes that the finder returned. Wrong is the share of what it returned that is not the business's own contact, such as a fax number, a regulator's address or the web designer's email.
Mostly because it could not see them: the website showed a security check or was down, the contact was only inside a picture, or the page was built by scripts the finder could not run. LeadTrapper reads pictures and PDFs and opens script-built pages in a hidden browser, but it never gets past a security check by itself.
No. Websites change, some block automatic visitors and some show contacts in ways no tool reads perfectly. Be wary of any tool that claims 100%, and check how a figure was measured.
We picked 150 real business websites from OpenStreetMap in 8 groups, plus 45 Spanish, Portuguese and Russian ones, saved a browser copy of each, and recorded by hand every email and phone number the business publishes. The finder's results were then compared with that list. Websites that were gone when the list was made are left out, which leaves 148 and 42.
About LeadTrapper
LeadTrapper is a Windows app for finding business leads and contacting them on WhatsApp. It collects business contacts from Google Maps, Instagram, Facebook, TikTok, LinkedIn and business websites, checks which numbers are on WhatsApp with its Number Filter, and sends personalised campaigns with automatic delays from one or more WhatsApp accounts. It also has a chatbot, a shared inbox for every account, and Store Orders for online shops. It installs from the Microsoft Store, and the full app is free for 14 days, then $50 a year. Free web tools, browser extensions and a Wix app are listed on the Free Tools page.
Test It on Your Own List
LeadTrapper for Windows reads every business website for emails, phones and WhatsApp numbers and tells you how sure it is about each one. $50/year, 14-day free trial.

