Do I have to rebuild my form from scratch?

No. Scan it once, then place fields on top of the scan. The form keeps the layout everyone already recognises, and the completed PDF looks like the sheet you have always filed.

Does a scan have any fields to detect?

No. A scan is an image, so there is nothing to read automatically. You place each field where the answer belongs, working down the page.

The same scanned form with the answers printed onto it, as a PDF. One per response.

What resolution should I scan at?

200 to 300 dpi. Below that, small print goes mushy on a phone screen. Much above it and the file gets large for no visible gain. Scan in colour or greyscale, not black and white, if the form has shading or a logo.

No. OCR adds a searchable text layer, which is useful for finding old files but does nothing for form filling. Fields are placed by coordinates on the page, not by reading the words underneath them.

Can I photograph the form instead of scanning it?

Yes, but use a document-scanning app rather than the plain camera, so the edges are detected and the perspective corrected. Save the result as a PDF, one page per page.

How to turn a paper or scanned form into an online form

Scan the form once, upload the scan, and place a field wherever an answer belongs. The layout everyone already knows stays exactly as it is, but people can now complete it on a phone, and what comes back is that same form with the answers filled in.

Start with one good scan

Everything downstream is a picture of this scan, so it is worth two minutes of care.

Scan at 200 to 300 dpi. Below that, small print goes mushy on a phone screen; far above it, you get a large file for no visible gain. Scan in colour or greyscale rather than pure black and white if the form has any shading or a logo — bitonal scanning turns light grey boxes into nothing.

Lay the sheet flat and square. A skewed scan is the single most common cause of fields that look misaligned later, because you will place your boxes against a page that is rotated a degree or two, and the printed answers will sit at an angle to the ruled lines. If the original is a photocopy of a photocopy, find a cleaner parent copy first. You are about to make this image the canonical version of a form you may use for years.

If you have no scanner, a phone photograph will do, but use a document-scanning app rather than the plain camera so the edges get detected and the perspective corrected. Then save the result as a PDF, one page per page.

Do I need to run OCR on it?

No. OCR recognises the characters in an image and adds an invisible text layer behind the picture, which makes the file searchable. That is genuinely useful for an archive. It does nothing for form filling. Fields sit at coordinates on a page; whether there is recognised text under them changes nothing about where an answer prints or how it behaves. Skip the step unless you want the archive benefit for its own sake.

Placing fields on an image

Upload the scan and mark a rectangle wherever an answer goes. Give each one a label — "Participant name", "Date of activity", "Emergency contact number" — because the label is what the recipient reads on their phone, not your scanned artwork. Choose a type for each: a line of text, a date, a number, an email address, a choice from a fixed list.

Run the scan through the free fillable PDF checker once, mostly to confirm what you have. It will report no fields and no selectable text at all, because the entire page — headings, ruled lines, your logo, the crease down the middle — is one flat image. That is the expected answer for a scan, not a fault in the file. Why can't I type in this PDF? explains how a scan differs from a flat PDF and why neither accepts input.

Placing the fields is the same job it would be on any other PDF: you draw each box where the answer goes. A one-page form is usually eight to fifteen of them and takes a few minutes. The only real difference with a scan is that you cannot copy a printed label off the page to paste into your field name, so you retype them and check your spelling as you go. Two habits make it quicker: work top to bottom in reading order, so the on-screen form comes out in the order the paper does, and make each box slightly more generous than the printed line, so a long name has somewhere to sit.

What about a multi-page form?

Scan every page into one PDF rather than one file per sheet, and keep them in the paper order. Fields are placed per page, so a four-page enrolment pack becomes one link and returns one four-page completed PDF. Splitting it into four forms gets you four submissions to reunite by hand, which is the problem you started with.

How do I check the alignment?

Send yourself a test submission before you send the link to anyone. Fill in your own form with obviously fake values — "XXXXXXXXXX" in every text box is better than a real name, because it shows you where a long answer would overflow — look at the returned PDF, and nudge anything that sits high or low. Do it once, and the next two hundred come out right.

Why keep the familiar layout?

Because people who use a form regularly do not read it — they navigate it.

You could rebuild the form from scratch as a clean web form. Sometimes that is the right call. Often it is not, and the reason is recognition. A trustee who checks waivers at a gate knows the date is bottom-left. A site foreman knows the third block is the one that matters. Redesign the form and everyone who was fluent in it becomes a beginner again, and you spend a season fielding questions that the old sheet answered by shape alone.

There is also the question of what other people accept. If a form goes to a committee, a governing body, an insurer or a landlord's agent, the version they know is the version they will process without asking. A tidier reworking gets queried.

And there is a version-control benefit that only shows up later. Keeping the scan as the master means there is exactly one form. The paper in the drawer and the link you send are the same document, so nobody is ever comparing a 2023 paper copy against a redesigned online one and wondering which clause applies.

What your recipients see

They see a web page with your labels laid out top to bottom, sized for their screen. They do not pinch and zoom around a photograph of a sheet of paper, and they do not need to — the scan is the output format, not the input format.

They complete it in a browser, on a phone if that is what they have, with no app to install and no printer involved. When they submit, their answers are drawn onto the scan at the positions you marked, and you get a PDF that looks like the form you have always filed, with the boxes written in.

One completed PDF per response, collected in one place, so a stack of waivers from a Saturday event ends up as a stack of PDFs rather than a folder of phone photographs. How to collect form responses as completed PDFs covers what happens to them after that, including the CSV export.

Can I still use the paper version?

Yes, and most organisations run both for a while. The scan is the master, so a sheet handed over at a desk and a link tapped on a phone are the same document with the same wording. File the paper copies as you always did and the returned PDFs alongside them. What you should not do is maintain two different layouts, because the moment the wording drifts you have two forms and no way to tell which one a given response answered.

If your form is not on paper

If you are digitising a process rather than a specific sheet — you have always done it by hand and there is no canonical form — do not scan something just to have an image. Start from a form that already exists. The schools and childcare templates cover the paperwork that goes home in a bag and comes back late, including student enrolment and emergency contacts, with the fields already placed. Send a link the same afternoon.