TL;DR, Quick Answer
8 min readSession replay data retention is how long a vendor keeps a recording playable before it deletes it. Published defaults run from 30 days (Microsoft Clarity, Datadog, Sentry Developer, PostHog Free) to 90 days (Sentry Team and Business) and a year (FullStory Free), with paid ceilings of 15 months per replay in Datadog and 5 years on PostHog Enterprise. Hotjar's help article shows two different rules to different readers, and LogRocket publishes no default at all. The GDPR sets no number: Article 5(1)(e) says personal data may be kept identifiable "for no longer than is necessary", and Article 13(2)(a) makes you tell visitors the period or the criteria behind it.
What is session replay data retention?
Every vendor sets its own session replay data retention, the period a recorded session stays playable before the vendor deletes it automatically. The window matters twice. It caps how far back a developer can go to reproduce a bug, and it fixes how long you hold personal data about every visitor you recorded.
The window is usually shorter than the one for analytics. A recording is a full reconstruction of a visit, with DOM snapshots, mutations and input events, so it holds far more about a person than a pageview count does. Several vendors split the two clocks on purpose. FullStory's help center names separate "Session replay retention" and "Product analytics retention" periods, and Datadog's extended retention "only applies to Session Replay and does not include associated events."
How long does each vendor keep session recordings?
Each vendor publishes a different number, and two of the seven publish no fixed default. The figures below come from each vendor's own documentation, read on 25 September 2026.
| Vendor | Default replay retention | Longest published | Where it is stated |
|---|---|---|---|
| Microsoft Clarity | 30 days from recording | Favorites and a random sample of recordings up to 9 months | Clarity FAQ, learn.microsoft.com |
| Hotjar | Two versions on one page: 365 days on all plans, or 1 month (Basic), 6 months (Plus), 12 months (Business) | 12 months, or 365 days | "How Long Does Hotjar Keep My Data?" |
| FullStory | 1 year on Fullstory Free | Not published for Enterprise, Business or Advanced | "Fullstory Plan Retention" |
| LogRocket | Not published; set per plan | Watched sessions saved up to 12 months (Pro, Enterprise) | "Extended Session Retention" |
| PostHog | Up to 30 days (Free) | Up to 90 days (Pay-as-you-go), 1 year (Boost, Scale), 5 years (Enterprise) | "Replay recording retention" |
| Sentry | 30 days (Developer) | 90 days (Team, Business, Enterprise) | "Data Retention Periods" |
| Datadog | 30 days | 15 months per replay with Extended Retention | Session Replay docs |
A few details sit behind those cells. Clarity's FAQ says "Clarity retains recordings for 30 days from the time of recording. However, Favorite recordings and randomly selected sample of recordings are retained for up to 9 months." The same FAQ answers a different question with a flatter rule, "The data is retained for the webmaster's consumption up to 30 days from the time of recording", and a few answers later says "You can access heat maps for up to 9 months." One page holds three retention answers.
PostHog frames every tier as a ceiling ("Up to 30 days"), so a project can pick a shorter window in the replay settings. Sentry states its periods as plain defaults per plan. FullStory says only that Enterprise, Business and Advanced customers can "specify how long to retain data for product analytics and session replay separately", and change it through their Fullstory account executive or the Adjust plan option under Settings > Account Management > Subscription.
LogRocket's pricing page says the "Final price depends on seats, retention length, and add-ons like Streaming Export", and its Plans page doc lists "Session Retention: how long your session replays will be kept" as a line on each customer's plan. Neither page names a number.
Why does Hotjar publish two retention rules on one page?
Hotjar's retention article carries both versions in its source and shows one depending on which pricing experiment the reader's account falls into. The HTML wraps each version in a block tagged data-experiment-id="pp_usage_based_pricing_phase_0_assignment", with a control variant and a variant1 variant.
The control text says Recordings and Heatmaps "retain data for 365 days from the date of capture. This applies to all accounts and plans." The variant1 text says retention periods "differ based on your plan" and lists "Basic plan: 1 month", "Plus plan: 6 months" and "Business plan: 12 months". A Basic customer who reads the control version expects a year and gets a month.
The rest of the page is the same in both versions. Surveys and Highlights are never deleted automatically. Engage audio and video "are deleted automatically after 2 years". Deleting data by hand "takes up to 30 days in total and all data is hard deleted." Neither of the two Recordings rules says which accounts it covers, so confirm the period for your plan with Hotjar before a DPIA or a privacy notice quotes it.

Can a single recording outlive the default?
Four of the seven vendors let some recordings outlive the default window, and each one picks them differently.
- Datadog. Extended Retention keeps a replay for 15 months. "The 15 months start when Extended Retention is enabled, not when the session is collected." Turn it off on a replay older than 30 days and "the replay immediately expires."
- LogRocket. Since 18 January 2024, Extended Session Retention saves every session "watched by a member of your organization, up to 12 months from recording date", capped at "5% of your session quota each month". A watched session means one played "for over 1 second". It is on by default for Pro and Enterprise, and disabling it leaves earlier saves archived: "Contact us if you need them deleted."
- Microsoft Clarity. Favorite recordings last up to 9 months, and so does a random sample Clarity picks itself.
- PostHog. "Export to JSON" downloads a recording, and "Downloaded recordings can then be imported back into PostHog for future playback, even if the original recording has expired."
LogRocket's rule has a privacy side effect. Watching a replay extends how long you keep it, so the recordings your team looks at most, often the ones showing a failure in checkout or signup, stay longest. The PostHog export route has a different problem. The JSON file leaves the vendor's retention policy entirely and lives wherever someone saved it.
When does a retention change take effect?
A retention change applies to new recordings only, at the two vendors that document the point. PostHog says "existing recordings keep their original retention period." Sentry says "Retention periods are set at the time data is ingested, based on the then-current plan", so a downgrade does not shorten the life of data you already hold.
Deletion also lags. PostHog warns that "Recordings may still appear for a short time after the retention period expires or when manually deleted via the UI." Hotjar gives a number for manual deletions: up to 30 days until hard deletion, backups included. Clarity offers no way to delete one person's recordings. Its FAQ answer to "Can I delete data for a specific user?" is "You need to delete the entire project to delete user's data."

Flowsery
Start Your 14-Day Free Trial
Real-time dashboard
Goal tracking
Cookie-free tracking
What does GDPR require for session replay data retention?
The GDPR sets no maximum number of days. It sets a test. Article 5(1)(e), the storage limitation principle, says personal data must be "kept in a form which permits identification of data subjects for no longer than is necessary for the purposes for which the personal data are processed."
Four other provisions turn that test into paperwork:
| Provision | What it requires |
|---|---|
| Recital 39 | "the period for which the personal data are stored is limited to a strict minimum" and "time limits should be established by the controller for erasure or for a periodic review" |
| Article 13(2)(a) | Tell the visitor "the period for which the personal data will be stored, or if that is not possible, the criteria used to determine that period" |
| Article 30(1)(f) | The record of processing lists, "where possible, the envisaged time limits for erasure of the different categories of data" |
| Article 17(1)(a) | A person can demand erasure once the data "are no longer necessary in relation to the purposes for which they were collected" |
Two consequences follow for replay tools. A vendor default is not a purpose. "30 days because the plan says so" does not answer "necessary for what?", so write the reason down, for example "replays are kept for one release cycle so reported bugs can be reproduced." And Article 13(2)(a) means the period belongs in your privacy notice, which is hard to write honestly if your vendor's help center shows two answers. Whether you need consent to record at all is a separate question, covered in whether session replay is GDPR compliant.
How do you choose a retention period for session replays?
Pick the shortest window that covers the time between a user hitting a bug and your team reproducing it. For most teams that means one or two release cycles plus the support queue's lag. Long windows rarely pay off for replay, because a bug reported several releases ago usually sits in code that has already changed.
Three settings reduce what the window holds before you shorten it:
- Mask at capture. A recording that never contained a card number or an email address has less to expire. The mechanics are in PII redaction for session replay.
- Record less. Sampling cuts storage and exposure together, with trade-offs explained in session replay sampling rates.
- Split the clocks. Keep aggregate analytics longer than raw recordings, the model FullStory and Datadog already document.
Flowsery follows the split model. Replays are kept for 3 months on the Team plan and 12 months on the Pro plan, and analytics data for 3 years on both. Its AI session replay reviews each recording and turns what is breaking into ranked, reproducible issues, so the evidence a ticket needs is pulled out while the recording still exists.
Frequently asked questions
What is the default session replay retention in Microsoft Clarity?
Clarity keeps recordings for 30 days from the time of recording. Favorite recordings and a random sample Clarity selects last up to 9 months, and heatmaps stay available for up to 9 months.
How long does Sentry keep session replays?
Sentry keeps session replays for 30 days on the Developer plan and 90 days on Team, Business and Enterprise. The period is fixed when the replay is ingested, so a later plan change does not alter it.
How long does PostHog keep session recordings?
PostHog allows up to 30 days on Free, 90 days on Pay-as-you-go, 1 year on Boost and Scale, and 5 years on Enterprise. A new setting applies only to recordings captured after the change.
Does GDPR set a maximum retention period for session recordings?
No. Article 5(1)(e) requires that identifiable data be kept "for no longer than is necessary" for its purpose. The controller sets the number, justifies it, and states it or its criteria in the privacy notice under Article 13(2)(a).
Can I keep one Datadog session replay longer than 30 days?
Yes. Enabling Extended Retention on a replay keeps it for 15 months, counted from the moment you enable it. The extension covers the replay only, not its associated RUM events.
Why does Hotjar's help center show different retention periods?
The article runs a pricing experiment. One version says 365 days for all plans, the other says 1 month on Basic, 6 on Plus and 12 on Business. Neither version says which one governs your account, so confirm the period for your plan with Hotjar before you quote it.
How long does FullStory keep session replays?
Fullstory Free keeps replays for 1 year. The help center publishes no period for Enterprise, Business or Advanced. Enterprise, Business and Advanced customers can set product analytics and session replay retention separately, through their account executive or the Adjust plan option in Settings.
Does LogRocket publish a default session replay retention?
LogRocket names no default number. Session retention is a line on each customer's plan, and the price depends on retention length. Extended Session Retention saves sessions watched by your organization for up to 12 months, capped at 5% of your session quota each month.
Can I delete one person's recordings in Microsoft Clarity?
Not for a single person. Clarity's FAQ says you need to delete the entire project to delete a user's data. Otherwise recordings expire after 30 days, and favorites and a random sample last up to 9 months.
Flowsery
Start Your 14-Day Free Trial
Real-time dashboard
Goal tracking
Cookie-free tracking
How do I choose a session replay retention period?
Pick the shortest window that covers the time between a user hitting a bug and your team reproducing it. For most teams that is one or two release cycles plus the support queue's lag. Write the reason down, because GDPR Article 5(1)(e) asks for data kept no longer than necessary, and state the period in your privacy notice.
Was This Article Helpful?
Let us know what you think!
See us more often in Google
One click marks Flowsery as a preferred source, so our articles sit higher in your Top Stories, AI Mode, and AI Overviews.
Before you go...
Flowsery
Revenue-first analytics for your website
Track every visitor, source, and conversion in real time. Simple, powerful, and cookie-free.
Real-time dashboard
Goal tracking
Cookie-free tracking
Related Glossary Terms


What rrweb Records, and How Its Privacy Options Work
How rrweb records a session as a DOM snapshot plus mutations, not video: its npm packages, MIT licence, and privacy options like maskAllInputs and blockClass.


What Session Replay GDPR Compliance Actually Requires of You
A session replay GDPR review turns on four things: an Article 6 lawful basis, Article 9 special category data, ePrivacy Article 5(3) consent, and a DPIA.


Where PII Redaction Session Replay Defaults Leave Data Exposed
PII redaction session replay defaults vary wildly: rrweb masks only passwords, Sentry masks all text, Clarity masks numbers and emails.


How Cookieless Analytics Counts Visitors Without an Identifier
A cookieless analytics tool counts visitors without storing an identifier in the browser. What that removes, what it costs, and why fingerprinting fails.


The Test That Settles Data Controller vs Data Processor
The GDPR test for data controller vs data processor is who determines the purposes and means. What each role signs, owes, and does when a breach hits.


How Session Replay Works, and What It Cannot See
A session replay rebuilds a visit from DOM mutations and input events, not video. See what it captures, what masking hides, and how it differs from heatmaps.
Related Articles


How a Google Analytics Opt Out Works, and What Each Method Blocks
A Google Analytics opt out can be a browser add-on, a ga-disable window flag or denied consent. What each one blocks, who controls it, and where it stops.


Where the Data Comes From in Session Replay vs Heatmaps
Session replay vs heatmaps: one visit against many, and how Hotjar, Datadog, PostHog, Clarity and FullStory build heatmaps from, or apart from, recordings.


How the Legitimate Interest GDPR Test Works Under Article 6(1)(f)
Legitimate interest GDPR rules in Article 6(1)(f), quoted from EUR-Lex, with the EDPB three-part test and why web analytics is still a contested case.

