What CECO PDF is, and who runs it

CECO PDF fixes the wrong word in a PDF or a scan, in the document's own font, and checks that nothing else on the page changed.

What it does

CECO PDF corrects text inside a PDF: a typo, a misspelt name, a wrong date or number, in a document of your own, or one you are entitled to alter. It works on scans, where the words are ink in a picture of the page, and on PDFs whose words are real text, and it runs in the browser, with nothing to install.

Its point is what happens after each correction. The page is rendered again, before and after, and compared pixel by pixel, and the corrected file cannot be downloaded until nothing outside the corrected area has changed. How it works follows one correction from start to finish. Its capabilities are on one page too: every tool, format, limit and refusal, each figure read from the service that enforces it.

Each way people ask for a correction has a page of its own: editing the text in a PDF in the same font, editing a scanned PDF without retyping it, changing a date or a number you typed wrong in your own document, and removing text from a PDF for real, rather than covering it. Editing a PDF on a Mac, or on an iPhone or iPad, has a page too. So do Edit PDF on Windows and Edit PDF on Android.

With a Business plan, it can also be used from your own software, with an API key; the API documentation says how.

Adding to a page, and taking out for good

It can edit a PDF by laying text boxes, images, shapes, highlights and notes over its pages, each a real PDF annotation that can be taken off again, with the page's own content kept as it was; adding text to a PDF has a page of its own. A box that would hide words on the page is refused: changing those words is a correction. It can also redact a PDF, removing words, areas and images from the file itself, not just covering them, and reading the file back before it is kept. A region that holds a watermark, a signature, a seal or a security feature is refused. In the same editor it can crop a PDF, by its margins or to a box, keeping what lies outside the crop in the file and repair a damaged PDF, listing anything that was lost.

Filling in a form, and signing it

In the workspace where you correct a document, Fill & sign can fill in a PDF form. It outlines the blanks it finds, you can draw a box where it missed one, and nothing is written until you have looked over every answer. The result is either a finished copy, each answer set in type that matches the form, or a copy with working form fields over the blanks, for a PDF reader to fill. Either way the page is checked as a correction is. A field it could fill only badly, such as one too small for its answer, is refused with its reason, and the rest of the form is still filled. A blank that already holds ink on the page is refused too, and never written over. A form in the XFA format, or a document already issued rather than a blank form, is refused as a whole.

It can also sign a PDF with your own mark, once you have confirmed that the mark is yours: one you draw, your name or initials in a standard italic type, or an image of your signature that you upload. No handwriting is made up for you. A place on the page that already holds ink or a signature is refused, and so is a PDF that has already been signed digitally. The record says where each answer and each mark went, without the answers themselves.

What it does with whole pages

It can also merge PDFs, split one, extract pages, delete pages, put them in a new order and turn them. Each makes a new file, carries every page across exactly as it was stored, and says in the file which pages came from where.

Marks and passwords

It can add a watermark, page numbers, headers and footers and Bates numbers, each drawn over the page and never into it, and checked so that nothing outside the mark changes. It can protect a PDF with a password, and take off a password you know — only with that password: it is not a password cracker, and it never removes a watermark.

Comparing two versions

It can compare two PDFs and show exactly what changed between them, with the same pixel-for-pixel comparison it holds its own corrections to: every changed place numbered, the words that changed beside it, and a page added or taken out shown as that.

Making a file smaller

It can compress a PDF too, into up to four smaller files to choose from, each with its exact size and a close look at its smallest text beside the original's, or into a file just under the size an upload form allows, keeping as much detail as fits: a hundred kilobytes, two hundred or a megabyte. It never makes a page harder to read to reach a size; when a size cannot be reached, it says how close it gets and offers to split the file instead.

Reading a scan

It can make a scanned PDF searchable: the words read on each page go under its picture, each where its ink is, and the picture is left exactly as it was. It can give the words back as text too, laid out as the page has it if you like, and a page that is clearly a table as a spreadsheet. A word it is not sure of is left out rather than guessed, and a page it cannot read with confidence is named as unreadable, not given a wrong reading.

Turning other files into PDFs

It also works the other way round: it can convert other files to PDF — pictures from a camera or a phone, as JPG, PNG or HEIC; documents from Word, Excel and PowerPoint; plain text and Markdown; and HTML. Each conversion tells you how faithful it is. A JPG's compressed picture travels into the PDF unchanged. A HEIC photo cannot, because PDF has no room for that format, so it is re-saved, and you are told. Office files are drawn by LibreOffice instead of Microsoft's own software, so lines and pages may break elsewhere; whatever difference it can detect is named. A web page file is built from its own contents, and nothing it refers to online is downloaded.

Turning a PDF into other files

From a PDF, it can make other files: a picture of each page as PNG or JPEG, at the resolution you choose; its words as plain text, or as text laid out as the page has it; a Word document; an Excel workbook; or an archival PDF/A copy. A scan is read the way the OCR tool reads one, and an uncertain word is left out there too. Word gets text you can edit only from simple prose: if a page holds a picture, a drawing, a table or columns, every page goes into Word as a picture of itself. Excel takes a PDF only when every page is clearly a table, and keeps each cell as the text it holds, without formulas. The archival copy is made of pictures of the pages, and is handed over only once a validator has passed it. Each export comes with a record of what its format kept and what it lost.

What it leaves alone

CECO PDF corrects words; lays text boxes, images, shapes, highlights and notes over a page; redacts; fills in forms and signs them with your own mark; merges and splits PDFs; extracts, deletes, reorders and rotates pages; adds watermarks, page numbers, Bates numbers and passwords, and takes off a password you know; compares two versions; makes files smaller; reads scans and makes PDFs from other files, and other files from PDFs. It does not retouch photographs or drawings, and it does not ask other people to sign. Besides the corrected words, the one thing it adds to a file it corrects is the record of what changed. When a correction on a page of real text needs a whole paragraph set again to fit, the result says so rather than doing it quietly.

What it holds to

• Every correction is checked before it can be downloaded.
• There is no replace-all. Each correction is made where you point, and checked on its own.
• It says what it did: whether the new words were typed as text, moved from the page's own ink or drawn to match it, and when a paragraph had to be laid out again.
• A refusal is an answer. A correction that cannot be made well is refused, with its reason, and the document stays as it was.
• Your file's name stays in your browser. Only its length and its extension reach the service, unless you choose to print a name on a certificate.
• No part of a document is kept for any other purpose unless you choose to report a region of it, and nothing chooses for you: no setting, and no box that starts ticked.
• Uploaded documents are deleted after 90 days, and you can erase everything yourself, at any time, without asking anyone.

Who runs it

CECO L.L.C. (Kosovo, Business ID 812382810) runs CECO PDF and is responsible for the documents you upload to it.

To get in touch, write to admin@cecopdf.com.

Elsewhere

CECO PDF is also on YouTube, LinkedIn, X, TikTok, Instagram and Threads, each under the name cecopdf.

CECO PDF is listed on SaaSHub.

Where its rules are written

The terms say what a verified correction means, who may alter a document, and how payment and cancellation work. The privacy notice says what is kept about you and your documents, why, for how long, and how to erase it. The pricing page lists what each plan costs and allows, and the refund policy says how to ask for a refund within 14 days of a purchase.