Session Replay vs Heatmaps vs Product Analytics: What to Use

Every product team hits the same wall eventually. You look at your funnel and see a 40% drop-off between signup and activation. You know that it happens, but you have no idea why. Is the onboarding flow confusing? Is a button broken on Safari? Are users reading the instructions or scrolling right past them? The answer depends on which tool you reach for.

Session replay, heatmaps, and product analytics are not competing solutions. They are three completely different lenses on user behavior, and each one reveals things the other two cannot. Product analytics tells you what happened and how many users it affected. Heatmaps show you where attention and interaction concentrate on a page. Session replay shows you why a specific user struggled, in full context, click by click.

The mistake most teams make is picking one tool and assuming it covers everything. A startup installs Mixpanel, tracks events religiously, and still has no clue why their pricing page converts at 2%. An e-commerce team watches session replays for hours and can describe individual user frustrations but cannot quantify whether those frustrations affect 5% or 50% of visitors. A marketing team stares at a heatmap showing nobody clicks their hero CTA but has no way to understand the journey that led users to ignore it.

This guide breaks down what each tool type actually reveals, the leading vendors in each category, the privacy and performance trade-offs you need to understand, realistic costs at different scales, and a recommended stack for startups, growth-stage companies, and enterprises. If you are already leaning toward product analytics, our PostHog vs Amplitude vs Mixpanel comparison goes deeper on that specific decision.

Session replay records a user's entire interaction with your product and lets you play it back like a video. Under the hood, most tools use DOM reconstruction (via libraries like rrweb) rather than actual screen capture. The browser's DOM mutations, scroll positions, mouse movements, clicks, and input events are serialized into a lightweight data stream and reconstructed in a player on the other end. This means the "recording" is typically 5 to 50KB per minute, not the megabytes you would expect from video.

The real power of session replay is qualitative insight. You stop guessing what users did and start watching it. A user hovers over a tooltip, reads it, then immediately closes the modal and leaves. A user fills out a form, hits submit, sees a flash of red text, and starts the form over. A user clicks a disabled button six times in a row, gets frustrated, and bounces. These micro-interactions never show up in analytics dashboards, but they explain the numbers you see there.

Leading Session Replay Tools

FullStory is the most mature dedicated replay platform. It targets enterprise product and UX teams with deep search capabilities, frustration scoring (rage clicks, dead clicks, error clicks), and robust behavioral analytics layered on top of replay. FullStory's DX Search lets you find specific sessions using natural language queries like "show me users who hit an error on the checkout page and then left." Pricing is enterprise-only, typically starting around $300 to $500/month for small teams and scaling significantly with session volume. Expect to pay $2,000 to $10,000/month at growth stage.

LogRocket sits at the intersection of replay and debugging. It captures Redux, Vuex, and Zustand state alongside DOM replay, logs all network requests with payloads, and surfaces JavaScript errors in context. For engineering teams that need replay primarily for debugging rather than UX research, LogRocket is often the best fit. Plans start at $99/month for 10,000 sessions and scale to $500+/month for 100,000 sessions. Their free tier gives you 1,000 sessions/month.

Hotjar is the entry point for most teams. It bundles basic session replay with heatmaps, surveys, and feedback widgets in a simple UI designed for non-technical users. Hotjar's replay quality is good enough for UX insights but lacks the deep technical debugging features of LogRocket or the enterprise search of FullStory. The free plan includes 35 daily sessions, the Plus plan ($32/month) includes 100 daily sessions, and Business plans ($80+/month) scale from there. Hotjar was acquired by Contentsquare in 2023, so expect deeper integration with enterprise experience analytics over time.

PostHog includes session replay as part of its open source product analytics suite. You get replay alongside event analytics, feature flags, and A/B testing without needing a separate vendor. The free tier includes 5,000 recordings/month, and paid recordings run $0.005 per recording. For teams already using PostHog for analytics, this is the most cost-effective replay option. For a detailed breakdown of PostHog and its competitors, check our Sentry vs LogRocket vs FullStory comparison.

When Session Replay Shines

• Bug reproduction: A user reports "the checkout button didn't work." Replay shows you the exact sequence of actions, browser state, and error that caused the failure.
• Onboarding optimization: Watch new users go through your activation flow. See where they hesitate, misclick, or abandon.
• Rage click identification: Surface sessions where users repeatedly click non-interactive elements, revealing UX confusion.
• Qualitative validation: You already know 30% of users drop off at step 3. Replay shows you the specific friction causing it.

Heatmaps aggregate interaction data from hundreds or thousands of users and overlay it on a visual representation of your page. Instead of watching individual sessions, you see patterns. Where do most people click? How far do they scroll? Which parts of the page attract mouse movement (and presumably attention)? Heatmaps turn individual behavior into statistical patterns you can act on.

Types of Heatmaps

Click heatmaps show where users click or tap. Hot zones (red/orange) indicate high-click areas, cold zones (blue/green) indicate low interaction. This immediately reveals whether your CTA buttons are getting attention, whether users are clicking non-interactive elements (a sign of UX confusion), and whether important navigation items are being ignored. Click heatmaps are the most actionable type because clicks represent intentional decisions.

Scroll heatmaps show how far down the page users scroll before leaving. This is critical for long-form content, landing pages, and product pages. If 80% of your users never scroll past the fold, your below-the-fold content is invisible. Scroll heatmaps often reveal that the most important information on a page is placed where nobody sees it. They also expose "false bottoms," where the visual design of a page makes users think they have reached the end when more content exists below.

Movement (hover) heatmaps track mouse cursor position over time. The theory is that cursor position correlates with eye gaze, and research from Google and Carnegie Mellon generally supports this for desktop users. Movement heatmaps show which headlines users read, which images they examine, and which sections they skip entirely. On mobile, these are less useful since there is no persistent cursor.

Leading Heatmap Tools

Hotjar is the most popular heatmap tool by install base. It generates click, scroll, and movement heatmaps with minimal setup (one script tag). The free plan supports 35 daily sessions for heatmaps, and paid plans start at $32/month. The UI is simple enough for designers and marketers to use without training.

Microsoft Clarity is completely free, backed by Microsoft, and surprisingly capable. It offers click and scroll heatmaps, session replay, and "smart" insights that automatically flag dead clicks, rage clicks, and excessive scrolling. There is no session limit. For budget-constrained teams, Clarity is hard to beat. The trade-off is that Microsoft uses the anonymized data to improve their products, which may be a concern for privacy-sensitive industries.

Contentsquare (which now owns Hotjar) offers enterprise-grade experience analytics with AI-powered zone-based heatmaps, revenue attribution per page zone, and journey analysis layered on top of heatmap data. Pricing is custom, typically $30,000 to $100,000+/year for enterprise contracts.

Lucky Orange is a mid-market option that bundles heatmaps, session replay, surveys, and live chat. Plans start at $32/month for 5,000 sessions. It is a solid all-in-one choice for small e-commerce teams that want heatmaps without enterprise pricing.

When Heatmaps Shine

• Landing page optimization: See exactly which elements get attention and which are ignored. Rearrange content based on real interaction data, not assumptions.
• CTA placement testing: Before running an A/B test, use a heatmap to confirm users are actually seeing your CTA. No amount of copy optimization helps if nobody scrolls to it.
• Content strategy: Scroll heatmaps reveal whether users read your long-form content or bail after the first paragraph.
• Design validation: Confirm that clickable elements look clickable and non-clickable elements are not attracting clicks.

Product analytics is the quantitative backbone of any data-driven team. While session replay shows you individual user stories and heatmaps show you page-level patterns, product analytics answers the structural questions: How many users completed onboarding this week? What percentage of trial users convert to paid? Which feature correlates most strongly with 30-day retention? Where in the funnel are you losing the most revenue?

At its core, product analytics tracks events (user actions) and properties (attributes of those actions and the users who performed them). You define an event taxonomy, instrument your code to fire events at the right moments, and then build funnels, retention charts, cohort analyses, and dashboards from that data. The quality of your analytics is only as good as your event taxonomy. Garbage in, garbage out.

Leading Product Analytics Platforms

PostHog is the open source platform that bundles event analytics, session replay, feature flags, A/B testing, surveys, and a data warehouse. You can self-host it for complete data ownership or use their cloud product. The free tier is generous: 1 million events/month, 5,000 session recordings, and 1 million feature flag requests. Paid pricing is usage-based, starting at $0.00031 per event after the free tier. For startups that want one tool instead of five, PostHog is the most cost-effective choice and the one we recommend most often.

Amplitude is the enterprise analytics platform built for product teams that need deep behavioral analysis. Amplitude's behavioral cohorts, predictive analytics, and "Amplitude Experiment" (their A/B testing product) are best-in-class for large organizations. The Starter plan is free up to 50,000 monthly tracked users. The Growth plan runs approximately $49/month and up, while the Enterprise plan is custom-priced. Amplitude is the right choice when your analytics team needs advanced segmentation, causal analysis, and SQL-level flexibility.

Mixpanel is the original event analytics platform, now repositioned as a leaner, more developer-friendly alternative to Amplitude. Mixpanel's query interface is fast and intuitive, their SDKs are well-documented, and their pricing is straightforward. The free plan includes 20 million events/month (very generous). The Growth plan starts at $20/month. Mixpanel dropped support for "people profiles" as a separate billing dimension, simplifying their pricing model significantly.

When Product Analytics Shines

• Funnel analysis: Track conversion through multi-step workflows. Identify which step loses the most users and quantify the revenue impact.
• Retention tracking: Measure whether users come back after day 1, day 7, and day 30. Identify which actions during the first session predict long-term retention.
• Feature adoption: Ship a new feature, track how many users discover it, use it, and return to use it again. Kill features nobody uses.
• A/B test analysis: Measure the statistical significance of experiments and make confident decisions about which variant to ship.
• Revenue attribution: Connect user behavior to revenue events. Understand which user segments and acquisition channels generate the most lifetime value.

If you are building a mobile product, the analytics requirements shift. Native SDKs, offline event queuing, and device-specific metrics become critical. Our mobile app analytics guide covers the platform-specific considerations in detail.

Every user behavior tool collects personal data. The question is how much, where it goes, and whether you have given your users adequate notice and control. Getting this wrong is not just a compliance risk. It is a trust risk. One poorly handled data incident can undo years of brand building.

Session Replay Privacy Risks

Session replay is the highest-risk category because it captures the most data. A replay can inadvertently record passwords, credit card numbers, personal messages, health information, and any other content displayed on screen. Most tools offer masking options, but they require configuration. FullStory and PostHog mask all text by default and require you to explicitly opt in to capturing visible text. LogRocket and Hotjar capture text by default and require you to opt out of sensitive fields. The "mask by default" approach is significantly safer for GDPR and CCPA compliance.

Under GDPR, session replay typically requires explicit consent (not just a cookie banner that mentions "analytics"). The replay captures user behavior in enough detail that it constitutes profiling under Article 22. You need a clear privacy policy disclosure, a consent mechanism that allows users to opt out, and a data processing agreement with your replay vendor. Under CCPA, you must disclose the collection in your privacy notice and honor "Do Not Sell" requests if the replay data is shared with third parties.

Heatmap Privacy

Heatmaps are lower risk because they aggregate data. A click heatmap does not show you what any individual user did. It shows you what 10,000 users collectively did. However, the data collection process still captures individual interactions before aggregating them. Most heatmap tools use the same JavaScript recording technology as session replay, just with different output. Microsoft Clarity and Hotjar both capture individual interactions to generate heatmaps, which means the raw data carries the same privacy considerations as replay. The aggregated output is safer, but your data processing activities still need to comply with consent requirements.

Product Analytics Privacy

Product analytics captures structured events, not visual recordings. This makes it easier to control exactly what data you collect. You define the event taxonomy and the properties attached to each event. If you never send a user's email address as an event property, it never reaches the analytics platform. PostHog's self-hosted option gives you complete data sovereignty, since the data never leaves your infrastructure. Amplitude and Mixpanel are both SOC 2 Type II certified and offer EU data residency.

Practical Compliance Checklist

• Consent: Implement a consent management platform (OneTrust, Cookiebot, or similar) that blocks session replay and heatmap scripts until users opt in. Product analytics with anonymized, non-cookie-based tracking may be permissible under legitimate interest, but get legal advice specific to your jurisdiction.
• Data masking: Enable "mask all text" mode in session replay by default. Only unmask elements you have explicitly reviewed for PII risk.
• Data retention: Set retention policies on replay and analytics data. Most tools default to 90 days for replays and 12 months for analytics. Shorter is safer.
• DPA: Sign a Data Processing Agreement with every vendor. FullStory, LogRocket, PostHog, Amplitude, and Mixpanel all provide standard DPAs.
• Privacy policy: Disclose session replay, heatmap, and analytics tools by name in your privacy policy. Generic "we use analytics" language is not sufficient under GDPR.

Every script you add to your site costs something in performance. Session replay, heatmaps, and product analytics all inject JavaScript that runs on every page load, captures events, and sends data to a remote server. The performance impact varies significantly by tool category and vendor.

Performance Impact

Session replay has the heaviest footprint. The recording library must observe DOM mutations, serialize changes, capture mouse movements at 20 to 60fps, and batch-upload the data. FullStory's JavaScript bundle is approximately 60 to 80KB gzipped. LogRocket is around 70KB. PostHog's recorder is roughly 50KB. On a modern device with a fast connection, the impact is negligible (2 to 5ms per frame of additional overhead). On a low-end Android device with a 3G connection, replay can add 200 to 500ms to initial page load and increase CPU usage by 5 to 10%. If your user base includes emerging markets or budget devices, test replay performance carefully before enabling it for 100% of sessions. Most teams sample replay at 10 to 50% of sessions to balance insight with performance.

Heatmaps use similar recording technology but typically send less data per session because they only need interaction coordinates, not full DOM reconstruction. Hotjar's script is approximately 30KB gzipped. Microsoft Clarity is around 20KB. The performance impact is lighter than replay but not zero. Expect 1 to 3ms per interaction event.

Product analytics SDKs are the lightest. PostHog's JavaScript SDK is about 20KB gzipped (without replay). Amplitude's is approximately 35KB. Mixpanel's is around 15KB. These SDKs fire events only when specific user actions occur, so ongoing CPU overhead is minimal. The performance impact of a well-implemented analytics SDK is effectively invisible to users.

Cost Comparison at Scale

Let's compare costs for a SaaS product with 50,000 monthly active users and 500,000 monthly sessions.

• Session replay (100% capture): FullStory: $3,000 to $8,000/month (enterprise pricing). LogRocket: $500 to $1,500/month (depends on plan). PostHog: approximately $2,500/month (500K recordings at $0.005 each). Hotjar: $128+/month (Business plan with 500 daily sessions, which is roughly 15,000/month, not full coverage).
• Session replay (10% sampling): Reduce the above costs by roughly 90%. PostHog drops to approximately $250/month. LogRocket drops to $99 to $300/month.
• Heatmaps: Microsoft Clarity: $0 (free, unlimited). Hotjar: $80 to $128/month for Business plans. Contentsquare: $30,000+/year (enterprise).
• Product analytics: PostHog: Free up to 1M events/month, then approximately $0.00031/event. At 5M events/month, roughly $1,240/month. Mixpanel: Free up to 20M events/month, then Growth plan at $20+/month (Mixpanel's free tier is remarkably generous). Amplitude: Free Starter plan for up to 50K MTUs, Growth at approximately $49+/month, Enterprise at custom pricing.

Total Cost of a Full Stack

A startup running PostHog for analytics and replay (10% sampling) plus Microsoft Clarity for heatmaps pays roughly $250 to $400/month total. A growth-stage company running Amplitude for analytics, LogRocket for replay, and Hotjar for heatmaps pays $500 to $2,000/month. An enterprise running FullStory, Contentsquare, and Amplitude is looking at $5,000 to $20,000/month or more. The cost difference between budget-conscious and premium stacks is 10x to 50x, and the premium stack is not 10x better. Choose deliberately.

There is no single "best" tool combination. The right stack depends on your team size, technical maturity, budget, and what questions you need to answer. Here is what we recommend based on the hundreds of product teams we have worked with.

Pre-Product-Market Fit (Seed Stage, Under 1,000 Users)

At this stage, you need qualitative insight more than quantitative rigor. You do not have enough users for statistical significance in A/B tests or meaningful funnel analysis. Watch users. Talk to users. Understand what they are trying to do and where they get stuck.

• Session replay: PostHog free tier (5,000 recordings/month) or Microsoft Clarity (unlimited, free).
• Heatmaps: Microsoft Clarity (free) or Hotjar free tier.
• Product analytics: PostHog free tier (1M events/month). Track activation events, core feature usage, and basic retention.
• Total cost: $0/month.

Your priority at this stage is speed. Do not spend weeks setting up a complex event taxonomy. Track the five events that matter most: signup, activation milestone, core feature usage, return visit, and payment. Watch 10 session replays per week to stay connected to the user experience.

Growth Stage (Series A/B, 10,000 to 100,000 Users)

Now you have enough data for quantitative analysis to be meaningful. You need proper funnels, cohort analysis, and experimentation. You also need replay for debugging and UX optimization, and heatmaps for landing page and conversion optimization.

• Session replay: LogRocket ($99 to $500/month) for teams that need debugging context. PostHog paid tier for teams that want one platform.
• Heatmaps: Hotjar Business ($80 to $128/month) for marketing and design teams. Microsoft Clarity as a free supplement.
• Product analytics: PostHog or Mixpanel for product-led growth companies. Amplitude Growth for teams with dedicated analysts.
• Total cost: $200 to $1,500/month depending on choices.

The critical investment at this stage is your event taxonomy. Spend time defining a consistent naming convention (we recommend `object_action` format, like `checkout_started`, `payment_completed`, `feature_used`). A clean taxonomy at 10,000 users saves you a painful migration at 100,000.

Scale Stage (Series C+, 100,000+ Users)

At scale, you need enterprise-grade tools with advanced segmentation, predictive analytics, data governance, and SOC 2/HIPAA compliance. You likely have a dedicated analytics team that needs SQL access to raw data, and your privacy and security team has specific vendor requirements.

• Session replay: FullStory for UX research teams. LogRocket or Sentry for engineering debugging. PostHog self-hosted for data sovereignty.
• Heatmaps: Contentsquare for enterprise experience analytics with revenue attribution. Hotjar/Clarity for lightweight page-level insights.
• Product analytics: Amplitude Enterprise for deep behavioral analysis. PostHog self-hosted for teams that need complete data ownership. Consider running both: Amplitude for business stakeholders, PostHog for engineers.
• Total cost: $3,000 to $20,000+/month.

At this stage, the tools are not the bottleneck. Your data culture is. The companies that get the most value from analytics are the ones where every product decision starts with "what does the data say?" and every launch includes success metrics defined before the code is written.

The core takeaway is straightforward: session replay, heatmaps, and product analytics are complementary, not competing. Product analytics tells you what is happening and how often. Heatmaps show you where attention concentrates on a page. Session replay shows you why individual users behave the way they do. You need all three perspectives to make truly informed product decisions.

Start with the questions you need to answer. If your biggest problem is understanding why users drop off in a specific funnel, start with product analytics to quantify the drop-off points and session replay to watch users who dropped off. If your biggest problem is low conversion on a landing page, start with heatmaps to see what users interact with and scroll heatmaps to confirm they see your CTA. If your biggest problem is bugs that users report but engineers cannot reproduce, start with session replay tied to error monitoring.

For most teams, we recommend starting with PostHog (analytics plus replay) and Microsoft Clarity (free heatmaps). That combination covers all three categories at zero or near-zero cost and gives you a foundation to build on. As you grow, you can layer in specialized tools like Amplitude for advanced analytics, FullStory for enterprise replay, or Contentsquare for experience analytics. But do not buy the enterprise stack at seed stage. You will spend more time configuring tools than talking to users.

The best analytics setup is the one your team actually uses. A perfectly configured Amplitude instance is worthless if nobody checks the dashboards. A FullStory account with 100,000 recorded sessions is worthless if nobody watches the replays. Pick tools that match your team's habits and workflow, invest in a clean event taxonomy early, and build a culture where data informs every product decision.

If you want help choosing the right analytics stack for your product, or you need a team to implement it properly, book a free strategy call with our team. We have set up analytics infrastructure for dozens of SaaS products and can help you avoid the common pitfalls that waste months of engineering time.