Online, humans act like wild animals — they sniff for the prey and track it. This is the “information scent” theory of human behavior, which states the conversion rate of website visitors is dictated by their user experience. If they find the information they’re looking for, they’ll stick around; if not, they’ll move on. Hotjar to the rescue! It is a website analytics tool that gathers user feedback and provides behavior analytics. Sounds too complicated? Read this Hotjar review to the end, and you’ll figure it all out.
Ease of use
Value for Money
Starting from: $39/month
Pricing Model: Subscription
Yes, has free trial
Yes, has free version
- A unique insight into user behavior
- A delight to use
- Honors the Do-Not-Track browser setting
- Lackluster surveys
- Results depend on the data set
- Doesn’t support mobile apps
- Small business
- Medium business
- Large enterprises
- Cloud, SaaS, Web-Based
Hotjar — understanding your users’ behavior
Website visitors are swimming in information and stimuli. They have a fraction of a second to choose where to go and what to do. Unfortunately, they make these choices without realizing how, so asking them can’t reveal the truth. Instead, website owners need to employ a tool like Hotjar, which provides a visual way to understand web analytics while also providing heat maps and user feedback tools. By understanding their visitors and customers, website owners can remove sources of frustration and make for a better user experience.
What is Hotjar?
Hotjar is a user data aggregation tool that uses website heatmaps and user activity recordings to understand what drives user engagement. The data is then used by product managers, designers, and researchers to remove hurdles in the user experience.
The shining example is the below-the-fold principle. Researchers have long known that 80% of website visitors make their judgment of a website without scrolling down the page. First impressions matter; in the case of a website, it’s whatever is in the top left corner of the page. Interestingly, visitors from countries using right-to-left languages have a mirrored scan pattern — they start in the top right corner and scan the contents right to left, going down the page. If the landing page does not have the needed information, the visitor will move on without a second thought.
A Hotjar heatmap will reveal which website areas aren’t attracting enough attention. You can then create a survey to gain more insight into user behavior and narrow down why.
How to use Hotjar?
Sign up with an e-mail address or a Google account. After confirming the 6-digit PIN, you will be asked to provide your organization name, size, and your role in it. There are seven organization size tiers, with the largest intended for 5,000+ employees. There are also seven user roles, such as analyst and store owner. Past the Hotjar login, enter a web address you own and install the Hotjar code into the corresponding page. You will get three options:
- enter code manually
- install on the platform
- Share the code
The code has 11 lines. First, you install the tracking script by copying/pasting this code into its HTML’s <head> section. After the installation, you can have Hotjar verify it’s properly installed. This automatically opens the Hotjar dashboard and starts the 14-day free trial, but the first results will only come after seven days of monitoring.
Now let’s discover some of its best features, and then you can decide if this is the tool you need or not.
Upon logging into Hotjar, the user is welcomed by the dashboard. It has six sections in the left-hand sidebar:
Overview and Highlights are for reviewing all the data captured. Heatmaps and Recordings review the involuntary user interactions, while Incoming Surveys are for voluntary user feedback.
Unique insight into user behavior
To understand a person, we should walk a mile in its shoes. Hotjar lets you experience a first-person view of what the user did, revealing cursor movement, mouse clicks, and the feedback sent as if you’re the user. This is provided under the tab “Recordings” as a video file with regular video controls, allowing you to rewind, fast-forward, speed up, or slow down the playback.
User activity is captured by recording cursor position on the web page 10 times a second and when a click happens. In addition, Hotjar uses the MutationObserver API, a feature in all modern browsers, to check for changes on the web page, parse, and send them to the WebSocket.
The system can detect “rage clicks,” multiple fast clicks on the same area when the user is stuck and about to give up. The rage click moments and other notable events are denoted with appropriate emoticons in the progress bar in the recording.
Interacting with the recording is delightful and easy, like using a YouTube video. Watching one is mesmerizing and feels like spying on a wild animal with bated breath.
There is no way to download these recordings; you can only export a list of their URLs in the .CSV or .XLSX file format. The limit is 100,000 recording URLs per file. In addition, Hotjar supports tracking on single-page applications (SPA) that infinitely load content as the user scrolls and other custom-made web pages that tinker with the URL.
Custom trigger events
By default, Hotjar captures only sessions longer than 30 seconds, but you can toggle this off, and the capture can be set to trigger when an event happens. To do this, you will have to add an event in the Hotjar dashboard and a corresponding snippet of code in the HTML of the target web page. For example, triggers can be clicking a survey, starting, or completing checkout, and so on. People can use Google Tag Manager to set them.
Limitations and restrictions
Hotjar will not capture sessions where:
- the web browser is not supported
- cookies are disabled or blocked
- Do-Not-Track is turned on
- the browser is in private/incognito mode
- browser’s local Storage or session Storage is disabled
- the browser uses script blockers or adblockers, such as Ublock Origin
- the internet connection to Hotjar is blocked or interrupted
When a user lands on a web page, Hotjar takes a full-size screenshot measuring 15,000 pixels high. Any page that stretches down further will only be rendered up to 15,000 pixels. Heatmaps are also limited when the page has a scrolling element on the DIV container. Hotjar cannot detect the scrolling element or properly track the interaction with contents inside the container.
Pages with iFrames and Adobe Flash content will also not generate a heatmap. Finally, it will not generate the heatmap if the web page is over 10 MB in size due to code and/or images.
You can create surveys in Hotjar by hand, which is a lot of work, or by using the provided survey templates. You can also feature your template or request Hotjar make a new one. Surveys can be added to your workflow through Zapier. Surveys feel like they don’t belong in Hotjar; they’re just plain form fields where the user types in answers.
For such an amazing service focusing on visual feedback, I would expect Hotjar surveys to be more spectacular, visual, and dynamic. But, unfortunately, visitors must be coaxed into filling surveys because they’re a chore, and Hotjar does nothing to change that.
The Incoming Feedback widget lets you ask users for feedback and send you a screenshot of the area they’re referring to. This is a brilliant and intuitive twist on the old x-out-of-five-stars system, making me wish Hotjar innovated in the survey department as well. Unfortunately, for a company dealing with user feedback, Hotjar seems to have trouble understanding what its users want.
Within two minutes of signing up, there were already two Hotjar surveys nudging me to complete them. One even promised a $50 gift card if the answers were valuable enough for Hotjar to warrant a follow-up survey.
Results depend on the data set
This is a flaw you don’t often see with services — Hotjar cannot guarantee any results since they exclusively depend on the data set your users provide through their activity. If there is bias in the data, Hotjar will produce a biased result, and nobody will be the wiser. Without giving serious consideration to your data, you’ll reach plausible but nonsensical conclusions.
The root cause of this is how human behavior changes over time. Individual behavior depends on and needs to be looked at in the context of contemporary trends that shift like dunes.
Results you get through Hotjar may not make sense or may only make sense within a certain period or due to a certain trend. All psychological research has the same problem of being notoriously hard to replicate outside of its societal context. People have free will too, so if they know that you’re using Hotjar and how it works, they may sabotage it by intentionally providing incorrect feedback or acting erratically.
In short, you must do big boy science to get big boy results. You must gather as much data as humanly possible and use a statistician and/or psychologist to interpret it. However, Hotjar feels like a kid’s toy in this regard. It lets you tinker with pretty colors, but that’s not good enough for serious research.
If you install another snippet of code in the <head> section of your HTML below the tracking one, you can assign user attributes. Some of them are:
- last purchase category
- total purchases
- last purchase date
- last refund date
They let you set apart visitors or customers that did an action during a certain time frame, such as refunding within the last 30 days. You can have up to a hundred attributes on any website, targeting any number of users.
This is called “user segmentation” and runs under Hotjar’s Identify API, only available for Business and Scale pricing plans. The user attribute data from your website is passed on to Hotjar via an asynchronous AJAX call, where it’s matched against a user ID. Hotjar lets you sift through users by their IDs or by their attributes.
Hotjar lets you gather anonymous data and then assemble it into a personal profile without any personally identifying information (PII), which formally satisfies GDPR and other privacy-related legislation. The tool was initially staunchly against gathering any PII but eventually caved in and left its customers a loophole — use our API and promise you won’t gather any PII while we look away.
Hotjar can gather the following data from a website where its code snippets are installed:
- user identifiers
- name, e-mail address, and other similar data
- protected classifications (race, gender, etc.)
- records of services or products bought
- user interactions (buttons clicked, mouse movements, etc.)
- current or past job history and/or performance assessments
- education data
The data collection is done through cookies, so if you have disabled them, Hotjar will take the hint and won’t gather any data. Hotjar also respects Do-Not-Track headers, which is a welcome surprise.
Do-Not-Track is a setting in modern browsers that lets websites and web analytics services know the user would like to stay anonymous. Of course, there is no such thing as the Internet Police, so nobody enforces it, and websites have no reason to honor it, but Hotjar does.
Privacy legislation allows web users to request and receive all personal data gathered by a website. Hotjar gives you access to the User Lookup tool if you get one such request. Through it, you can access user information using their e-mail addresses and/or user IDs.
Learning user secrets
So, how much can be revealed by just tracking customers’ shopping history? Well, quite a bit. 2012 The New York Times article titled “How Companies Learn Your Secrets” reveals how Target used the shopping records of female customers to determine which ones were pregnant, even when they didn’t know it themselves. In the early 2000s, this was without much computing power or fancy artificial intelligence, only using math and stats.
It turned out pregnant women buy one or more products out of 25 depending on how far along they are. By examining which products, the women who acknowledged they were pregnant for perks and benefits bought, Target statisticians could determine the probability of pregnancy and the due date of any woman shopper, even if she had never shopped at Target before.
Today, marketers and statisticians can access and crunch more data than ever before but so can anyone with internet access and some determination. That became painfully obvious in 2014 when the Cambridge Analytica scandal revealed just how much of what we like reveals who we are, even if we don’t consider it PII. In that instance, researchers correlated liking certain Facebook pages with life choices and events.
A user liking two Facebook pages, one dedicated to a certain band and one dedicated to a certain comic book, could be a teenager whose parents have or are about to get divorced. As more and more likes are known, the correlation becomes more precise. Researchers could guesstimate the person’s past, present, and future sexual orientation, drug use, and mental illnesses are given enough likes.
As for Hotjar’s session recordings, they could conceivably be used to estimate the likelihood of dementia since reaction speed and click precision are ostensibly correlated to the state of the person’s nervous system. The data could also show if the person is left-handed and reveal various personality traits, such as the propensity to get enraged and/or emotional.
There are four Hotjar pricing plans:
Plan prices are $39–$389 a month, with a 20% discount when billed annually, payable through credit cards and PayPal. There is a 14-day free trial that requires no payment method that allows capturing up to 500 sessions daily. In addition, there is a 30-day money-back guarantee for all plans.
- Up to 35 daily sessions
- Automatic data capture
- Unlimited heatmaps
- Up to 1,050 recordings/month
- Create and store 3 surveys & incoming feedback widgets
- Everything in Basic, and:
- Up to 100 daily sessions
- Powerful filters for recordings
- Trigger recordings by page or event
- Up to 3,000 recordings/month
- Unlimited surveys & incoming feedback widgets
- Everything in Plus, and:
- Up to 500 daily sessions
- Seven custom-built integrations
- Analyze key user cohorts
- Rage clicks and u-turns
- Removal of Hotjar branding
- Everything in Business, and:
- Unlimited daily sessions
- Dedicated customer success manager
- SAML single sign-on (SSO)
- Enough data for the largest sites
- Full access to every feature
See all Features
Hotjar Business and Scale plans support seven integrations:
- Google Optimize
- Zapier integrations
Google Optimize is a tool that tests the performance of HTML code in web pages. It is bundled with the Google Marketing Platform, an online website analytics suite. HubSpot is a marketing platform that helps you find new customers, draw them in, sell them stuff, and see if they’re happy about it. Provides features for marketers, developers, and the support staff.
Omniconvert is a growth platform for eCommerce businesses. From acquisition to conversion and retention, Omniconvert helps you define and build your ideal audience, improving the conversion rate for your website(s). In real terms, it lets you cut down on your marketing costs.
Optimizely is a user experience platform that lets you experiment with data and test your ideas. You can introduce multiple calls to action to a single product, observing which converts more and why. The platform also lets you manage your digital content.
The segment is a product and customer analytics tool. Applied to marketing, Segment lets you track the customer’s journey across products, platforms, and devices.
Slack is a mass messaging platform for tightly woven teams. Integrated with Hotjar, Slack can funnel feedback to you straight away. Zapier is an automation and integration platform. It links together systems that would normally be incompatible.
Now let’s discover some alternatives for HotJar.
QuestionPro is the heavyweight champion in the survey department. Granted, it does require a large team/budget and a lot of fiddly labor, but the results can pass scientific scrutiny. QuestionPro will support as large a survey project as you need without blinking.
Google Analytics is a web analytics tool inside the Google Marketing Platform. It is widely used and compatible with other marketing tools, though it doesn’t have the muscle enterprises need to grow.
Is Hotjar better than Google Analytics?
No. I would say that Hotjar is in a web analytics category of its own and could never compete with Google in traditional website analytics or platform integration.
Hotjar Review Conclusions
Hotjar is a fine behavior analytics tool that works well with other analytics tools, such as Google Analytics, but doesn’t have much apparent utility on its own. Moreover, Hotjar seems to bear all the trappings of the modern web, namely being obsessed with immediate presentability rather than long-term usability. So you get the pretty colors, recordings, and heatmaps, but unless you have the scientific inclination to dig until you find the context, they’re more likely to mislead you than lead you.
Striving to understand others is a worthwhile goal, especially if your business depends on goodwill and customer satisfaction. Observing others without interfering with their choices reveals who they are and where they’re headed. However, Hotjar is by no means enough for this and can at best whet your appetite for research.