Storage and delivery
Retention
How long a capture stays in your account. Set on the Web Profile, so it applies to every schedule using it — and worth deciding before a schedule has been running for six months.
The two settings
Keep captures for sits in the After the capture section of the Web Profile, and offers two choices. A new profile has neither chosen; you pick one.
| Value | Meaning |
|---|---|
Keep until I delete them | Captures stay indefinitely. Nothing removes them but you, or the bulk tools on the schedule's capture view. |
Do not keep after delivery | The capture is delivered to your cloud storage and then removed from here. It cannot be selected unless the profile has a destination. |
Each capture shows which applies to it: Kept until deleted, or Not kept after delivery. A capture made with an older retention value shows the date it is kept until.
Zero retention
“Do not keep after delivery” treats this account as a pipe rather than an archive. The capture is taken, written to your destination, and dropped here.
- It requires cloud storage. The profile form marks Deliver to as required for zero-day retention, because otherwise the capture would be deleted with nowhere to have gone. Cloud storage is included in every paid plan, not in the free trial.
- It keeps your storage use near zero, which matters on a high-frequency schedule.
- It puts the archive entirely in your hands. Whatever your bucket does with the file is now the whole story.
- The record stays; the file does not. Nothing is kept here - no thumbnail, no download, and the capture-ready email carries no preview. The capture still appears on the schedule with its status, timestamp and a per-destination result (Delivered or Failed, with the time it landed), and the API answers a download request for it with “This capture was delivered to your cloud storage and is not kept here.”
- Delivery is judged per capture. While it is on its way the capture reads Delivering. If no destination accepts the file it is removed anyway and the capture is marked Delivery failed; when at least one succeeds it is marked Delivered to cloud. A transfer that keeps failing is retried a few times and then treated as failed.
- No page source, no change detection. The page source is not delivered to cloud storage, and a comparison needs the previous capture, so neither works with this setting.
Check destination health if you rely on this
Storage is the real limit
With “Keep until I delete them”, nothing expires on a timer — but your plan includes a fixed amount of storage, and captures are not small. The brake is bytes, not days.
- A full-page PNG at 1920 wide is commonly several megabytes.
- A scroll video is larger again.
- A daily full-page capture of forty pages, kept forever, is a few hundred gigabytes over a couple of years.
Three ways to keep it under control, in the order worth trying them:
- Capture less often. A Trigger Window on the profile usually removes more volume than any other change.
- Capture smaller. JPG rather than PNG, viewport rather than full page, 1920 rather than 3840.
- Move the archive out. Deliver to your own bucket and use zero retention, where storage is yours to price.
Add-on storage exists
When storage is full
- New captures stop. Schedules are paused for you and read Paused - storage full. A capture refused for lack of space is not counted against your plan.
- They start again on their own once you delete captures or add storage.
- You are told before. Everyone in the workspace sees a storage warning in Notifications as storage fills up, and the owner gets an email.
Over the plan after a change
After a subscription ends
When a subscription ends, your schedules are paused and read Paused - subscription ended; every setting is kept. Captures you have stored are not kept forever: before anything is deleted, the owner receives an email naming the date. Subscribe again before then and nothing is lost — the schedules resume on their own.
Taking a copy before it goes
Before you shorten a retention setting - or simply because you want your own archive - a whole month of a schedule can be downloaded as a single zip. Open the schedule, press Download a month, and pick from the months that have captures; each one shows how many files it holds and how large it is before you commit to anything.
- Files are named by capture time, in the schedule's own time zone - so a month means your month, not a UTC one.
- The saved page source can be included, and lands under
page-source/. - Captures set to Do not keep after delivery are not in it: nothing of them is held here to put in an archive.
- A
SHA256SUMS.txttravels with the files. Runsha256sum -c SHA256SUMS.txt(orshasum -a 256 -c SHA256SUMS.txton a Mac) inside the unzipped folder and every file answers OK or does not — on any machine, years later, without us. The digests are the ones recorded the moment each file was written.
Captures taken before September 2026
SHA256SUMS.txt rather than listed with a wrong one. The rest of the archive is unaffected, and an archive in which nothing has a digest simply carries no checksums file.It cannot be paused or resumed
Changing retention later
Changing the profile setting affects captures from then on. To change what is already stored, use Retention on the schedule’s capture view, which applies to a date range you choose. Do not keep after delivery cannot be applied to captures that already exist — delete them instead.
- Shortening retention is always allowed.
- Extending it requires a running plan, since it commits storage.
- Favourites are skipped unless you tick Include favorites in the panel.
Older retention values
You may see 1 Month or 3 Months on an existing profile
Did this page answer your question?
Tell us what is missing