Best Session Replay Tools in 2026: PostHog, FullStory, LogRocket, Hotjar, and Microsoft Clarity

This guide is for a team that already knows it wants session replay - watching real user sessions to debug a broken flow or understand where people get stuck - and is choosing between PostHog, FullStory, LogRocket, Hotjar, and Microsoft Clarity. It is not a feature-by-feature spec sheet. It is built around one question: given what each tool actually costs and what changed in 2026, which one should a specific team pick.

The category looks stable from the outside, but one of these five vendors quietly restructured its entire pricing model this year, and another quietly removed its free plan. Both changes matter more than any feature comparison, so they come first.

Quick Decision

There is no single best session replay tool in 2026 - there is a best tool per constraint. If the constraint is budget, Clarity wins outright. If the constraint is "we already pay for PostHog," PostHog wins because replay rides on the same usage-based bill. If the constraint is "our engineers need error and log context alongside the video," LogRocket wins on capability but demands real budget from day one. If the constraint is "marketing needs heatmaps and surveys," Hotjar still wins on feature fit, but a buyer needs to understand it now bills through Contentsquare, not as one flat plan. FullStory wins for teams that want AI-assisted search across huge session volumes and are fine going through sales to get a number.

What Changed in 2026: Hotjar Now Bills as Three Separate Contentsquare Products

The biggest shift in this category this year is not a new feature - it is a billing structure. Hotjar was acquired by Contentsquare, and by July 2025 the transition into Contentsquare's own product and billing structure was complete. The evidence is easy to reproduce: fetching hotjar.com/pricing today returns a permanent HTTP 308 redirect straight to contentsquare.com/pricing. Hotjar as a brand still exists inside Contentsquare's product suite, but a buyer no longer lands on a dedicated Hotjar pricing page.

What replaced the old single Hotjar plan is a three-way split, according to Contentsquare's own migration documentation as tracked by third-party pricing analysts: Experience Analytics (heatmaps and session recordings - the closest match to classic Hotjar), Voice of Customer (on-site surveys and feedback), and Product Analytics. Each module is reported to carry its own Free, Growth, Pro, and Enterprise tiers and, in the new structure, its own separate bill. A team that used to write one check for heatmaps, recordings, and surveys together may now be managing two or three Contentsquare subscriptions to get the same coverage it had under one Hotjar invoice.

This is worth knowing before evaluating Hotjar on features alone, because the practical buying question is no longer "what does Hotjar cost" - it is "how many of these three modules does our team actually need, and does consolidating them under Contentsquare still make financial sense compared to a single-purpose tool like Clarity or PostHog."

Pricing, Tool by Tool

PostHog: usage-based, no seat charge, replay priced separately from analytics

PostHog does not sell a flat "Session Replay plan." Replay is one of several products inside PostHog, each metered independently after its own free monthly allowance. The web replay allowance is 5,000 recordings a month free (2,500 for mobile), and every seat on the workspace is included at every usage tier - team headcount does not change the bill. Past the free allowance, PostHog's official pricing page lists a declining per-recording rate: roughly $0.0050 per web recording from 5,000-15,000 recordings a month, stepping down to $0.0035, then $0.0020, $0.0017, and finally $0.0015 per recording at 500,000-plus recordings a month. Mobile replay is billed on a similar but roughly double-priced ladder. Recordings on the free plan are retained for about a month; paid plans get 90-day default retention, extendable up to 5 years on Enterprise add-ons.

The full cost-modeling breakdown - including what a 1, 5, and 10-seat team actually pays once product analytics, flags, and surveys are added on top of replay - lives in ToolPick's dedicated PostHog pricing guide. The short version for replay specifically: a team sampling replay lightly can stay inside the free allowance indefinitely, while a team recording most or all sessions at scale will feel it as one of the faster-growing line items on the bill.

FullStory: no public pricing, but a genuinely generous free tier

FullStory does not publish prices for its Business, Advanced, or Enterprise tiers - every paid plan on FullStory's pricing page routes through a "request pricing and demo" flow, and the page does not list a per-seat or per-session dollar figure for any of them. That opacity is the single biggest practical friction point for a small team evaluating FullStory: there is no way to self-serve a real number before talking to sales.

What FullStory does publish is its free plan, called FullstoryFree, and it is one of the more generous free tiers in this category: 30,000 sessions a month, 12 months of data retention, and access to core capabilities including session replay, basic analytics, and debugging tools for up to 10 users. That is a real product a small team can run on indefinitely for evaluation or even light production use before deciding whether the paid tiers are worth a sales call.

LogRocket: the free tier is gone - it is a 14-day trial now

This is the other major 2026 shift worth flagging directly: LogRocket's pricing page and its own signup page both currently describe access to the full feature set for 14 days, with no mention of an ongoing free plan with a permanent session allowance. A team evaluating LogRocket today should plan around a 14-day evaluation window, not an indefinite free tier.

On the paid side, LogRocket's Core plan lists a starting price around $176 a month at a 25,000-session-per-month baseline, with analytics events, error events, and log events described as always unlimited on every tier - only session volume is metered. The Pro tier layers on AI features (session summaries, automatic issue detection, funnel insights, and an AI assistant for querying product data); below roughly 100,000 sessions a month it is billed as a separate add-on, and above that volume LogRocket's pricing page describes it as included with Core at no extra charge. Enterprise is custom-priced and adds unlimited seats, streaming data export, and self-hosted options. For teams that do not want every session captured, LogRocket also offers a Conditional Recording option; its own pricing calculator shows one example landing around $765 a month for capturing roughly a quarter of total sessions - useful context for how steeply the bill moves with sampling rate, similar to the sampling advice that matters for PostHog's replay pricing above.

Hotjar / Contentsquare: reported pricing, not confirmed on a live invoice

Because Contentsquare's pricing page is rendered client-side and did not return tier-level numbers on a direct fetch, the figures below are sourced to Contentsquare's own support documentation as aggregated by a third-party pricing tracker, not independently recomputed by ToolPick from a live invoice - treat them as reported rather than verified firsthand, and confirm current numbers directly with Contentsquare before budgeting against them.

With that caveat: Experience Analytics, the module that covers heatmaps and session recordings, is reported to include a free tier around 200,000 sessions a month on a single project, with a Growth tier scaling from roughly $49 a month at 7,000 sessions up to around $739 a month past 200,000 sessions. That $49 figure specifically is the annual-commitment rate paid by credit card or PayPal - buyers who purchase via invoice (the standard path for corporate procurement) do not get this discount and pay more per year for the same tier, according to the same tracker. Voice of Customer (surveys) is reported to start around $99 a month for 500 responses, scaling toward roughly $1,479 a month at higher response volumes. Product Analytics is also reported to include a free tier - 10,000 sessions a month with 6 months of data retention - and only its paid Growth, Pro, and Enterprise tiers are custom-priced/contact-sales, matching the pattern of the other two modules. If those numbers hold, Hotjar's Experience Analytics free tier is actually more generous in raw session count than PostHog's or FullStory's free replay allowance - the trade-off is that a team wanting the full Hotjar-style bundle now needs to evaluate and likely pay for more than one Contentsquare product to get there, and needs to confirm whether it is buying on the discounted annual/card rate or the higher invoice-billed rate.

Microsoft Clarity: free forever, genuinely no cap, but narrow in scope

Microsoft Clarity has no paid tier. Its own homepage describes it as free to keep using indefinitely, built to grow with a business, with no traffic limits - and there is no pricing page listing dollar figures because there is nothing to charge for. This is the one claim in this article that does not need a hedge: Clarity's cost model has been unchanged and unambiguous for years, and nothing in 2026 changed it.

The real trade-off is retention and scope, not price. According to Microsoft's own data-retention documentation (last updated 2026-06-30), session recordings - what Clarity calls Playback Data - are retained for 30 days. If a session is labeled or favorited before that window closes, it is preserved for 9 months instead, and the same 9-month window applies to click and heatmap data. Clarity also has no feature flags, no experiments, and no in-product surveys - it is purpose-built for replay and heatmaps, not a PostHog or Amplitude replacement. A team that outgrows Clarity is usually outgrowing its scope, not hitting a paywall.

Comparison Table

Who Should Pick What

• Solo developer or pre-revenue team with no budget line for tooling. Microsoft Clarity. There is no cost trade-off to reason about, and 30-day replay retention is plenty for debugging a specific broken flow.
• A team that already runs PostHog for product analytics, flags, or experiments. PostHog session replay. It is one more metered product on a bill the team is already managing, not a second vendor relationship, and the free 5,000-recording allowance covers most early-stage usage.
• An engineering team that treats replay as a debugging tool, not a UX-research tool. LogRocket. Unified error, log, and network data alongside the replay video is the differentiator, but budget for it deliberately - there is no free tier to fall back on if the trial runs out mid-evaluation.
• A marketing or UX-research-led team that wants heatmaps and on-site surveys together. Hotjar via Contentsquare, with eyes open about the module split. Confirm with Contentsquare directly which modules a specific use case actually requires before assuming one subscription covers everything Hotjar used to.
• An enterprise team evaluating AI-assisted session search across very high session volumes, with time to go through procurement. FullStory. Use the 30,000-session free plan internally first to validate the workflow before starting the sales conversation for a paid tier.

Honest Cons, Tool by Tool

Every tool on this list has a real limitation worth knowing before committing, not just after the first invoice or the first missing feature shows up.

• PostHog: the bill is unpredictable by design. A team that forgets to sample replay and records close to 100% of sessions at scale will see recording costs grow faster than almost any other line item on the account, because replay is priced roughly an order of magnitude higher per unit than raw analytics events.
• FullStory: the free plan is generous, but there is no way to self-serve a real paid-tier price. A small team that just wants a number to compare against LogRocket or Hotjar has to go through a demo call to get one.
• LogRocket: the 14-day trial is a hard clock. A team that wants to validate LogRocket against a real usage spike - a launch week, a marketing campaign - needs to time the trial deliberately, because there is no ongoing free plan to fall back on if 14 days is not enough.
• Hotjar / Contentsquare: what used to be predictable (one Hotjar invoice) is now, by the vendor's own restructuring, potentially three invoices across three modules. A team that liked Hotjar's simplicity before the merger should re-confirm current packaging directly with Contentsquare rather than assuming legacy pricing still applies.
• Microsoft Clarity: free comes with real scope limits, not just retention limits. No feature flags, no experiments, no in-product surveys, and no product-analytics funnels - a team that needs any of those will outgrow Clarity by design, not by accident.

Migration and Buying Checklist

• Measure current or expected session volume before comparing any usage-based tier (PostHog, LogRocket, or Hotjar's Experience Analytics) - every one of these prices per session or per recording, so a guess at volume produces a meaningless price comparison.
• Set a billing limit before turning on replay broadly on a usage-based tool. PostHog's own limits mechanic stops ingesting data once a cap is hit rather than over-charging - a hard stop, not a throttle, so pick a ceiling deliberately rather than discovering it during a traffic spike.
• If evaluating Hotjar or Contentsquare, ask directly which of the three modules (Experience Analytics, Voice of Customer, Product Analytics) the account actually needs, and get a combined quote rather than assuming a single Hotjar-style price still exists.
• If evaluating FullStory, run the 30,000-session free plan internally first and document the specific workflow gaps before starting a sales conversation - it gives the team a concrete list of what to ask pricing for, instead of an open-ended "what does it cost."
• If evaluating LogRocket, confirm the 14-day trial window will actually overlap with a representative usage period (not a quiet week), since there is no fallback free tier once it ends.
• Before leaving any of these tools, confirm data export options and retention windows in writing - Clarity's 30-day default retention in particular means exported evidence needs to be pulled promptly if a session matters for a later incident review.

Bottom Line

The category did not get more expensive across the board in 2026 - it got more structurally uneven. Microsoft Clarity is exactly as free as it has always been. PostHog and LogRocket both meter usage, but LogRocket now demands a decision within 14 days where it used to offer an ongoing free tier. FullStory's generous free plan is unchanged, but its paid pricing remains deliberately opaque. And Hotjar, the tool most people still picture as one simple plan, now runs its pricing through Contentsquare's three-module structure. None of that makes any single tool wrong - it makes "what does session replay cost" a question that needs answering per vendor, per module, and per team's actual usage pattern, not from memory of how these tools used to be priced.

Related ToolPick Decisions

• PostHog Review 2026: Product Analytics, Feature Flags, and Session Replay Fit - The feature and workflow comparison this guide's PostHog section leans on.
• PostHog Pricing 2026: What Usage-Based Pay-As-You-Go Actually Costs - The full multi-product cost model, including replay volume scenarios at 1, 5, and 10 seats.
• Best Product Analytics Tools in 2026 - How PostHog's usage-based model compares to seat-and-MTU-capped analytics competitors, for teams deciding whether to bundle replay with analytics at all.