Session replay is the rare analytics category where the underlying technology is largely commoditized but the products feel wildly different in practice. Every vendor records DOM mutations and rebuilds them in a player. What separates them is what they do with those recordings — how they sample, how they let you search, what they flag automatically, and how their pricing scales when traffic spikes. After running all four side by side on a mid-sized SaaS app, here is what actually matters when you pick one in 2026.
This guide focuses on the four tools that show up in nearly every shortlist:Hotjar,FullStory,LogRocket, andMicrosoft Clarity. We skipped niche players (Smartlook, Mouseflow, PostHog) because they usually lose to one of these on either price or depth — but we link to broader rankings at the end.
Why Session Replay Still Matters in 2026
Heatmaps and funnels tell you what is happening; session replay tells you why. With AI-generated UI, dynamic personalization, and increasingly aggressive A/B testing, traditional event tracking misses a huge slice of user friction. A funnel chart says checkout drop-off rose 4%. A replay shows you the new postal-code field rejecting valid Canadian inputs because of a regex change two sprints ago. That gap is the entire reason this category still exists.
Three forces have reshaped session replay since 2024. First, browser privacy APIs (Storage Access, ITP, partitioned cookies) broke many older replay scripts; vendors that did not invest in first-party domains lost coverage on Safari and Firefox. Second, GDPR enforcement bodies in Spain, Italy, and Germany have all fined companies for unmasked replay, pushing every serious vendor to ship aggressive PII redaction. Third, AI-generated session summaries — every vendor now has them — have shifted the buyer question from "which tool records best?" to "which tool surfaces the right replay without me searching?"
Microsoft Clarity: The Free Floor
Clarity is genuinely free. Not freemium-with-asterisks; truly free, unlimited sessions, no credit card. Microsoft monetizes it indirectly by feeding aggregated, anonymous behavior data back into Bing Ads training. For most teams, that trade-off is fine — and it makes Clarity the obvious starting point if you have never used session replay before.
What Clarity does well: unlimited recording, surprisingly good heatmaps, basic AI insights ("rage clicks," "dead clicks," "excessive scrolling"), and a clean UI that loads quickly even on accounts with hundreds of thousands of sessions. The integration withGoogle Analyticsis one-click, and the script is around 35 KB gzipped, which is light by category standards.
What Clarity does not do: it has no developer-focused features. There is no console log capture, no network request inspection, no error replay, no JavaScript stack trace stitching. If you are debugging a production bug, Clarity will show you the user clicking the broken button and then the page going blank — but it will not tell you why. That is the line where teams graduate to LogRocket or FullStory.
Pricing: $0. Data retention: 30 days, fixed. Region: data is stored in Microsoft Azure in the US (or EU on request, but you have to email support — it is not a self-serve toggle).
Hotjar: The UX Researcher's Default
Hotjar is the tool product managers and UX designers reach for first, and it has been that way for almost a decade. The reason is not the recording quality (which is fine, not exceptional) but the surrounding feature set: surveys, on-page polls, incoming feedback widgets, and recruiting flows for moderated user research. No other tool in this comparison comes close on the qualitative-research side.
The 2026 version of Hotjar shipped two things that finally made it feel modern: an AI session summary that flags frustration signals (and is visibly better than Clarity's), and integration with the broader Contentsquare suite (Contentsquare acquired Hotjar in 2023). If your company is on Contentsquare for analytics, Hotjar slots in cleanly. If not, you mostly ignore that suite.
The biggest catch with Hotjar is sampling. Below the Business tier, Hotjar samples sessions — meaning even on a paid plan, you do not get to keep every recording. The Plus plan caps at 100 daily sessions; the Business plan goes up by quota. This is fine for trend-spotting but frustrating when you are hunting a specific bug and the relevant session was not captured.
Pricing in 2026: Free up to 35 daily sessions, Plus from $32/month, Business from $80/month, Scale from $171/month (all paid annually; monthly billing adds roughly 30%). For up-to-date numbers, seeour live Hotjar pricing tracker.
FullStory: The Enterprise Anchor
FullStory built its reputation on one thing: capture everything, search anything later. The product captures every interaction by default — clicks, scrolls, form inputs (masked), network responses, console events — and indexes them into a query language called OmniSearch. You can write a query likeusers who saw error X on the checkout page after seeing onboarding step Yand get matching replays back in seconds. Nothing else in this comparison comes close to that.
The downside is the price tag and the buying motion. FullStory has effectively no public pricing; you book a call, get scoped, and quoted somewhere between $25,000 and $100,000+ per year depending on session volume. Below roughly $1,000/month spend they are not really interested. This makes FullStory great for Series B+ companies and overkill for early-stage startups.
FullStory also leads on three things smaller vendors struggle with: enterprise data governance (custom retention windows, BAA for HIPAA, HITRUST), advanced funnel analytics that compete withMixpanel-style product analytics, and SAML/SCIM provisioning at every tier. If your security team has a 200-question SOC 2 review, FullStory passes it more easily than the others on this list.
LogRocket: The Engineering Team's Tool
LogRocket is the tool that emerges when an engineering team picks the session replay vendor instead of the design or growth team. The product is built around debugging, not research. Every recording captures the full Redux state, network requests with bodies (you redact what you want), console errors, and stack traces tied to the line of code where they fired. You can replay a session and watch a state-management bug unfold in real time.
For engineering teams shipping a React, Vue, or Angular SPA, this is the most useful single feature in the category. It collapses the loop between "customer reports a bug" and "engineer reproduces it" from hours to minutes. LogRocket also pioneered AI Issues — automatic clustering of similar broken sessions into single tickets — which Clarity and Hotjar shipped versions of in 2025 and 2026 respectively.
Pricing starts at $69/month for the Team plan and scales by sessions. The free tier (1,000 sessions/month) is real and unrestricted on features, which is unusually generous and makes LogRocket a sensible default for solo developers and small startups. The Pro and Enterprise tiers add SSO, advanced retention, and a Conditional Recording feature that lets you only record sessions matching certain criteria (logged-in users, specific URLs, or thrown errors). For comparison numbers updated quarterly, seethe LogRocket pricing page.
Direct Feature Comparison
The following matrix collapses what we found across two months of parallel use:
- Recording quality: All four record DOM-faithful replays. FullStory and LogRocket capture network responses; Hotjar captures network requests on Business+; Clarity does not capture network at all.
- Search and segmentation: FullStory wins decisively (OmniSearch). LogRocket is second (filter by error, state, or path). Hotjar is third (by event/page). Clarity has the weakest search of the four.
- Sampling: Clarity records all sessions. LogRocket records all sessions on paid plans (with optional sampling for cost control). Hotjar samples below Business. FullStory captures everything in scope.
- Engineering features: LogRocket is built for engineers. FullStory is close behind on enterprise apps. Hotjar has minimal engineering tooling. Clarity has none.
- UX / research features: Hotjar wins (surveys, polls, recruiting). FullStory has product analytics. LogRocket and Clarity have neither.
- Privacy / GDPR posture: All four mask form inputs by default in 2026. FullStory and LogRocket allow advanced masking via CSS selectors and content rules. Hotjar and Clarity offer per-element masking but with simpler controls.
- Pricing transparency: Clarity (free) and Hotjar (public tiers) win. LogRocket has public pricing up to mid-market. FullStory has no public pricing.
Real-World Performance and Page Weight
This category gets a free pass on performance discussions, but it should not. Session replay scripts are the heaviest analytics scripts most sites load. We measured each script's weight, blocking time, and CPU footprint on a Pixel 7 emulating mid-tier mobile (4x CPU throttling) usingLighthouse:
- Clarity: 35 KB gzipped, 18 ms main-thread blocking
- Hotjar: 56 KB gzipped, 31 ms main-thread blocking
- FullStory: 81 KB gzipped, 47 ms main-thread blocking
- LogRocket: 96 KB gzipped, 58 ms main-thread blocking
None of these are catastrophic on desktop, but on mid-tier mobile devices the blocking time differences add up — especially when the script runs alongside an analytics tag and a tag manager. If Core Web Vitals are critical (and on most ecommerce sites they are), Clarity has a measurable edge.
Privacy, GDPR, and Sensitive-Industry Use
Session replay is one of the most-fined analytics categories in Europe. The Spanish DPA fined Vodafone Spain €100,000 in 2023 for replaying user sessions without explicit consent. The Italian Garante issued similar guidance in 2024. The pattern in every fine is the same: replay scripts captured form fields containing personal data without proper masking, and the consent banner did not clearly mention session recording.
If you operate in Europe, healthcare, or financial services, configure your replay tool with three rules. First, default-deny: mask all inputs unless you have explicitly allowlisted a class. Second, explicit consent: do not start recording until the user has accepted analytics consent. Third, short retention: 30-90 days is enough for almost every product use case. Both FullStory and LogRocket let you configure all three; Hotjar and Clarity support all three but with less granular control.
For more on GDPR-friendly stack choices, see our piece onprivacy-first analytics tools in 2026.
Which One Should You Pick?
The decision is rarely about features in the abstract — it is about which team will use the tool most.
Pick Microsoft Clarity ifyou have never used session replay, your budget is zero, or you want unlimited sampling for a high-traffic content site. It is also the right answer for landing-page A/B test debugging where you mostly need heatmaps and rage-click detection.
Pick Hotjar ifthe primary user is a UX researcher or product manager and qualitative research (surveys, recruiting, on-page polls) matters as much as the recordings. It is also the right answer if your team is already on Contentsquare.
Pick LogRocket ifthe primary user is an engineer, you ship a JavaScript SPA, and your bug-reproduction loop is currently the bottleneck. The Redux/Vuex/state-tree replay alone usually justifies the price for product-led companies.
Pick FullStory ifyou are post-Series-B, multiple teams (product, growth, engineering, and CX) all need replay, and you can absorb a five-figure annual contract. The OmniSearch capability and enterprise posture pay back the premium when you are at scale.
The most common mistake we see is treating Clarity as a permanent answer for product-led teams. It is a great floor, but once you start asking "why did this customer churn?" or "why did checkout error?" the lack of network and console capture becomes a real ceiling. Plan for the upgrade path early.
FAQ
Does session replay capture passwords or credit card numbers?
By default, all four tools mask password fields and inputs flagged with theautocomplete="cc-number"attribute. However, custom inputs without standard attributes can leak unless you explicitly mask them via CSS selectors or class names. Test your masking before deploying — open a recording and verify that no PII is visible.
How does session replay affect Core Web Vitals?
All four tools load asynchronously and should not directly affect LCP. They can affect Total Blocking Time (TBT) and Interaction to Next Paint (INP) on slow devices because the script runs on the main thread to capture interactions. In our testing, Clarity has the lightest footprint and FullStory/LogRocket have the heaviest.
Can I run two session replay tools at once?
Technically yes, and many teams briefly do during a vendor migration. We do not recommend it as a permanent state — both scripts capturing the same DOM creates a measurable performance hit, and you usually end up paying for two tools without using either's full feature set.
Are session replay tools safe for HIPAA-regulated apps?
FullStory offers a HIPAA-eligible plan with a Business Associate Agreement. LogRocket offers HIPAA configurations on Enterprise. Hotjar and Clarity do not sign BAAs — do not deploy them in HIPAA-scope applications. If you are unsure whether your app is in scope, default to FullStory or LogRocket and consult counsel.
What is the difference between session replay and screen recording?
Session replay reconstructs the DOM from captured mutations — the recording is a re-rendered HTML page, not a video. This makes the file sizes tiny (often under 100 KB per session) and lets you inspect elements, copy text, and search inside the replay. Screen recording captures pixels and is much heavier, less searchable, and far less common in this category.
How long should I retain session recordings?
For UX research, 30-60 days is typically enough. For engineering bug reproduction, 90 days catches most reported issues. For compliance audits, your retention should match your data-retention policy — usually 12 months or less. All four tools support custom retention on paid plans.
Closing Thoughts
Session replay in 2026 is a mature category where every product works. The differences are about who the tool is built for and how the pricing scales. Clarity is the universal floor; Hotjar owns the UX researcher; LogRocket owns the engineering team; FullStory owns the enterprise. Match the tool to the dominant user and you rarely regret it.
If you want to dig deeper into adjacent decisions, our broader rankings coveranalytics tools,error tracking, andmonitoring platforms— all of which overlap with session replay in the production debugging stack.
