PDF tools

How to Fill In a PDF Form and Flatten It

How to fill a PDF form in your browser, why some PDFs have no fields to fill, and what flattening actually does to the answers you typed.

6 min readUpdated Aug 23, 2026

Two things go wrong with PDF forms. The first is that you cannot type into them at all. The second is subtler: you fill one in, send it off, and are told to resend it "not editable", because every answer is still a box someone could change. Both come down to the same piece of PDF machinery - what it is, how to work it with Fill PDF Form, and what flattening really does.

Why some PDFs can be typed into and others cannot

A fillable PDF carries an AcroForm: a list of named fields, each with a type, a value and a rectangle on the page where it is drawn. Click one in a reader and you are editing that field's value, not the page. The page underneath never changes.

Most documents that look like forms have none of this. The boxes and ruled lines are printed artwork - part of the same drawing instructions as the headings - so there is nothing to click and nothing to fill. A scan of a paper form is the extreme case: it is a photograph of a form, and no amount of software turns it back into one.

There is a third kind, XFA, built with Adobe LiveCycle: the fields live in an XML layout the PDF merely carries, which is why those files show "Please wait... if this message is not eventually replaced" in anything but Acrobat. Fill PDF Form detects XFA and says so rather than showing an empty list.

Filling a form in four steps

  1. Open Fill PDF Form and drop your PDF on. It is read in your browser tab, not uploaded.
  2. Check the count - it reports how many fillable fields it found, and any values the form already holds are shown filled in, so you only change what needs changing.
  3. Work down the list. Text boxes, tick boxes, dropdowns and multi-select lists each get the right kind of input.
  4. Decide whether to flatten, then click Fill and download.

Each field is labelled with its tooltip where the form has one - on a well-made form that is the actual question, "Full name as it appears on your ID", rather than the database-ish name the designer gave it. Underneath sits the technical detail: the field's real name, which page it is on, whether it is required, and any character limit. Fields the form marks read-only, like a pre-printed reference number, are shown greyed out with their value visible so you can read them but not break them.

What flattening actually does

Tick Flatten and each field's drawn appearance is painted into the page itself, then the interactive field is deleted. The words look identical. They still select, copy, search and print. But there is no longer a box to click into, so nobody can type over your answers or clear them by accident.

The clearest way to see the difference is to extract the text. Fill a form without flattening and pull the text out of the page: the answers are not there. They were never on the page - they live in the field objects alongside it, which is why a text extractor, a printer with an old driver, or a viewer that renders fields differently can all disagree about what the document says. Flatten the same file and the answers come out with everything else, because now they genuinely are the page.

That is what an office is asking for when it says "not editable" - but be precise about what it is not. Flattening locks the text, not the file: anyone can still open it. To keep it shut without a password, add one with Protect PDF; the two are different jobs and they compose fine.

Signature fields are left alone

If your form ends with a signature box, it is listed but never filled. A digital signature is a cryptographic operation performed with a certificate only you hold; nothing running in a browser tab can produce a valid one on your behalf, and a tool that drew a picture of your name in that box would be handing you something that looks signed and is not.

Flattening removes the empty box rather than faking it. If you need a real signature, sign the flattened file in whatever tool holds your certificate - that order matters, because signing locks the document and any fill afterwards would break the signature.

The rupee sign, and other characters that cannot be drawn

Field appearances have to be redrawn whenever a value changes, and the built-in font used for that is WinAnsi-encoded. It covers Latin letters, accents and most European punctuation - Jose Muller, cafe, an em dash and a euro sign are all fine. It does not cover the rupee sign, Devanagari, CJK or emoji.

Type one of those and the field warns you, naming the exact characters. The value is still written into the file, and readers with real font support - Acrobat, and browsers' built-in viewers - draw it correctly from the form's own fonts. What you cannot do is flatten that field, because flattening bakes in the drawing and there would be nothing to bake: the answer would be flattened away to nothing. So the tool refuses rather than silently handing you a blank box, and tells you which field is holding things up.

If a flattened file is non-negotiable, the workaround is to enter something the font can draw - "Rs 1,250" rather than the rupee symbol - or to fill the form in Acrobat, which embeds a font that can. It is a limit of drawing without a Unicode font, not a fault in your file.

A worked example: a 12-field application

Say a form has 12 fields: eight text boxes, two tick boxes, a country dropdown and a signature. You fill the eleven fillable ones. Without flattening the saved file still reports 12 fields, with the answers held in the field objects. With flattening it reports zero, the eleven answers have become page content, and the empty signature box is gone.

The file size barely moves either way - the answers are a few hundred bytes against a document that was already there. If the form has to fit an upload cap, that is a separate job for Compress PDF to Size, run after flattening rather than before.

What is left untouched

The pages, text, images and page sizes are unchanged; only field values, and on flattening the field objects, are rewritten. Links, bookmarks and comments come across unmoved. The title, author and dates are left as they were rather than re-stamped with today's date and the name of the library that saved the file - a detail most PDF tools get wrong, and one you can check with PDF Metadata before you send it.

All of it happens in your own tab. The form is read, filled and saved on your machine and nothing is sent anywhere, which matters more here than for most tools: the fields on a form are, by definition, the parts holding your name, address, income or claim number. Load Fill PDF Form, disconnect from the internet, and fill a form. It still works.

Frequently asked questions

Can I undo flattening?
Not from the flattened file. Flattening paints the answers into the page and deletes the fields, so there is nothing left to click into and no record of which words used to be which field. Keep the unflattened copy if you might need to change an answer later - fill the form, download it without flattening, and only flatten the version you are actually sending. The tool names the two files differently for exactly this reason.
Why can I type into my PDF in Acrobat but this tool shows no fields?
Because Acrobat can add fields, and can also let you place free-floating text on a page that has none. Neither is the same as the document having a form in it. If Acrobat is showing you a typewriter-style text tool rather than highlighted boxes, the PDF is flat and there are no fields for any other tool to find. The other possibility is an XFA form, which Acrobat fills through its own XML layout - this tool detects those and tells you rather than showing an empty list.
Will flattening make my form print correctly?
It usually helps. Field values are drawn from a separate appearance stream that some print pipelines and older drivers regenerate, ignore, or render with different spacing - which is the usual cause of a form that looks right on screen and prints with blank or clipped boxes. Flattening removes that whole layer of indirection: the answers become ordinary page content, so anything that can print the page can print them.