How to spotPublished Mateusz Smoczyński

Five things fake applications get wrong, and the one thing they get right

Fabricated applications are built to pass keyword filters, not to survive a careful look. Five checks any recruiter can do in a minute, and why the sixth one is the trap.

A fabricated application is optimised for one thing: getting through the first filter. That filter is usually a keyword match, sometimes a recruiter with ninety seconds per CV. Everything else about the application is an afterthought, and that is where it shows.

Here are five things we see over and over in real pipelines. None of them is proof on its own. Two or three together are worth a closer look before you book the interview.

1. Text you can’t see, but the ATS can

Open the PDF, press Ctrl+A. If a block of text lights up where you saw empty space, you have found hidden text: white font, tiny font, or text placed outside the page. It is almost always a list of technology names and job titles, put there so the CV scores well on keyword filters.

Why it matters: nobody does this by accident. It is a deliberate attempt to pass a machine check, which tells you something about the rest of the document.

2. A phone number that isn’t a phone

Many fabricated applications use VoIP numbers: virtual lines that can be created in bulk, forwarded anywhere and dropped after use. A quick carrier lookup tells you whether a number is a mobile, a landline or a VoIP line.

Why it matters: a VoIP number is not wrong on its own. Plenty of consultants use them. But it is common in applications submitted on someone else’s behalf, where the person you will talk to is not the person who will show up to work.

3. Two full-time jobs at the same time

Look at the dates. Two roles marked full-time that overlap for a year are either a copy-paste mistake or a work history assembled from somewhere else. Ask about it. A real candidate explains it in one sentence. A fabricated one gets vague.

4. The file says someone else wrote it

PDF and Word files carry metadata: author name, creation date, the software used. When the author field says one name and the CV says another, there is often an innocent reason (a friend’s laptop, an agency template). There is also often a not-so-innocent one: the same person produced the file for several “candidates”.

Why it matters: this is one of the strongest signals when it appears together with others, because it points at an operation rather than an individual.

5. The time zone doesn’t match the story

The same metadata shows when the file was last edited, in the editor’s local time. A CV that states Warsaw as the candidate’s location but was last saved at 3 pm UTC+8 is not lying yet. It is inconsistent. Inconsistencies are what you are looking for.

The one thing they get right

Keyword match. A fabricated application usually matches the job description better than a real one, because it was written from the job description. If your screening starts and ends with “how well does this CV match the role”, the fake wins and the real candidate with a slightly off CV loses.

That is the trap. The better a fraudulent application gets at looking good on the surface, the more the surface stops being useful. Which is why the checks above look at how the application holds together, not how well it sells.

Doing this at scale

Each check above takes a minute by hand. On a pipeline of a few hundred applications a month, that minute is the whole problem. TalentShield runs these checks, and about seventy others, on every application as it arrives inside your ATS, and shows recruiters the ones that fired, with the reasons. It never rejects anyone. It tells you where to look first.

If you’d like to see what it finds on your own applicants, book a demo.

See it in your ATSTalentShield runs candidate fraud detection inside Greenhouse, Teamtailor, Lever and any ATS with an API. Book a demo or talk to Mateusz.

Keep reading

Fake candidates 10127 September 2026

What is a fake candidate? A plain-language guide for recruiters

Fake candidates are not one thing. From AI-polished CVs to stolen identities run by state operators: what the term covers, how big the problem is, and why one signal never settles it.

Read →
Fake candidates 10127 September 2026

Why fake candidates are rising now, and what to do about it from Monday

Remote hiring, generative AI and industrial-scale fraud explain why fake candidates went from rare to routine. Here is the logic behind it and a three-step plan that costs nothing to start.

Read →

See it on your own pipeline.

Leave your email and the ATS you use. We come back within one business day to book a 30-minute demo, on your own applicants if you want. Just have questions? Talk to Mateusz, 30 minutes, no slides.

Book a demo