Run several PDF operations in one pass
Run several PDF operations in one pass, without downloading in between.
Input
Add several if the workflow starts by merging them
Workflow
No steps yet. Start from one of these, or add an operation below.
What to know about PDF Workflow
Most PDF jobs are not one operation. Combining four scans, stripping the metadata and shrinking the result means three tools, three downloads and three uploads of a document you would rather not send anywhere. Here you build the sequence once and run it once.
Each step hands its result to the next one. The document never leaves the browser tab between steps — there is no intermediate file to download, re-upload, keep track of, or forget to delete afterwards, which is the part that actually matters when the document is a contract or a set of medical records.
Steps run in the order you put them in, and the order is yours to change: move a step up or down, remove one, or add another, and the sequence you see is exactly the sequence that runs. If a step cannot follow the one before it — merging halfway through a workflow, or doing anything after the document has been encrypted — the builder says so while you are building rather than after you have run it.
If a step fails, the workflow stops there and tells you which one. It does not skip ahead and hand you a document that looks finished, because a file that quietly missed a step is worse than no file at all.
Fourteen operations can take part. The ones that cannot are the ones whose settings only make sense while looking at the finished document — drawing a redaction box or dragging pages into order cannot honestly be decided for a document three steps from now — and those stay as standalone tools.
How to run several PDF operations in one workflow
- 1
Choose your files
Pick one PDF, or several if the first step is going to merge them. They stay on your device.
- 2
Build the sequence
Add the operations you need and configure each one. Move them up or down until the order is right — the workflow runs exactly as shown.
- 3
Run it once
Every step runs in turn, with progress shown for each. One PDF comes out at the end, and nothing was uploaded at any point.
What PDF Workflow accepts and produces
- Input
- One PDF, or several if the workflow begins by merging them.
- Output
- A single PDF with every step applied, in the order you arranged them.
- Limits
- Fourteen operations can be chained. Redact, Crop, Add Text, Highlight and Organize cannot, because their settings need the finished document in front of you. Merge must come first and Protect must come last.
- Offline
- Yes, once the engines a workflow uses have been fetched. A workflow that includes OCR also needs that language model stored first.
Why PDF Workflow here
- One download at the end instead of one after every operation
- Intermediate documents never touch a disk or a server
- Invalid sequences are refused while you build, not after you run
- A failed step stops the workflow and says which one, rather than returning a half-finished file
Where your file goes
Nowhere. This tool runs inside your browser, and the page is served with a policy that forbids it from sending your document anywhere. You can confirm that yourself: open your browser’s developer tools, watch the Network tab, and process a file.
Questions
What is a PDF workflow?
A list of operations applied to a document one after another. Instead of merging a set of scans, downloading the result, uploading it to a compressor, downloading that, and uploading it again to strip the metadata, you describe the whole sequence once and run it once. The document passes from step to step inside the browser tab.
Which operations can I combine?
Fourteen: merge, extract pages, delete pages, rotate, OCR, fill forms, sign, flatten, watermark, page numbers, remove metadata, compress, unlock and protect. Each one is the same engine the standalone tool uses — the workflow arranges them rather than reimplementing them, so a step behaves exactly as its own tool does.
Why can I not add Redact or Crop to a workflow?
Because their settings cannot be chosen honestly in advance. Redaction needs you to see the words you are removing, cropping needs you to see the page you are trimming, and organising needs you to see the pages you are reordering — and at the point you are building a workflow, the document those steps will actually receive does not exist yet. Offering a rectangle drawn over the original while the step will really act on a merged and rotated document would be a lie. Those tools stay standalone, where you are looking at the real page.
Do I have to upload the file again after each step?
No, and removing that is the point of the feature. The output of each step is passed straight to the next one in memory. There is no intermediate download, no re-upload, and no file left in a downloads folder for you to clean up later.
Does the workflow really run on my device?
Yes. Every step is the same in-browser engine the individual tool pages use, running in a Web Worker in your tab. Nothing is transmitted at any point — you can watch the Network tab while a workflow runs and see for yourself, and the page is served with a policy that forbids the browser from sending your document anywhere even if the code tried.
Will a whole workflow run with no connection?
Yes, once the engines it needs are stored on your device. Open the tools you intend to use once while connected, and the sequence will then run with the network switched off entirely — every step of it, not just the page loading. A workflow containing OCR also needs that language model fetched beforehand, because it is a separate multi-megabyte download.
Can I reorder the steps?
Yes. Each step can be moved up or down or removed, with buttons rather than dragging, so the whole builder works from the keyboard. The order shown is the order that runs — there is no hidden reordering or optimisation behind the scenes.
What happens if one step fails?
The workflow stops at that step and tells you which one it was and why. Earlier steps are shown as completed and later ones as not run. You do not get a download, because handing back a document that quietly missed a step is worse than handing back nothing — you would have no way of knowing which one it missed.
Can I combine merge and compress?
Yes, and it is the most common reason people come here. Merge has to be the first step, because it works on the files you started with rather than on a document a previous step produced. After that, compress, remove metadata, watermark, number the pages or protect the result in whatever order suits you.
Can a workflow include operations that destroy something?
Yes, and it tells you before it runs. Flatten removes the editable form fields, remove metadata deletes document properties, and deleting or extracting pages discards the rest. Each of those is the purpose of the operation rather than a side effect, so the builder lists them plainly before you start rather than after you have a file you cannot undo.
Are my files or my workflow stored anywhere?
No. The documents stay in the tab and the sequence you build is held in the page's memory, not sent anywhere and not written into the address bar. Reloading the page clears both. Nothing about what you built or what you processed reaches this site's servers, because for this feature they do no processing at all.
Worth reading about PDF Workflow
- Preparing a scanned document to sendA scan hides text you cannot see and metadata you did not add. What to check, and in what order, before a scanned document leaves your hands.
- A black box is not redactionA black rectangle drawn over a name leaves the name in the file. Learn how PDF redaction really works, how leaks happen, and how to check your own document.
- Why is my PDF so large?Find out what is really taking up space in a PDF — scanned images, embedded fonts, duplicated resources — and which of them compression can and cannot recover.