Glassbox: Enterprise Customer Intelligence and Session Replay

My main use case for Glassbox is checking the interaction of customers with the application and identifying their pain points. I analyze if there are any issues, examine the logs, get the session ID, obtain the end-to-end transaction, and check API request responses for potential problems.

Glassbox has positively impacted my organization because it highlights issues that are not easily reproducible by users, the business, or any testing team. These issues are captured in Glassbox as certain users are only facing them. This helps the business target those particular customers in order to fix the issues for those specific users. It also helps perform impact analysis to determine how critical an issue is and to establish a timeline for fixing it.

The best features Glassbox offers include performing impact analysis. There is also an AI feature that has been added which is helpful for asking any questions I need regarding that particular session. The user journey from start to end, showing where they started and where they are stuck, and all the page details captured in the Glassbox session are important.

Additionally, the API details, code, and request response information are valuable.

One specific outcome I can share about Glassbox is a decrease in customer complaints because as soon as I catch an issue, as soon as there is an event where a particular issue is occurring and it crosses the threshold, the system sends an event and I can check that user session to see what is happening and address that issue immediately and resolve it. This decreases customer complaints. Sometimes there are dead-click issues that are not easily reproducible by the testing team or any other team in the organization, and Glassbox captures those as well. Glassbox helps detect and resolve issues by capturing dead clicks, page not working issues, and reloading issues. Those details help an organization resolve issues.

If I had to think of one area where Glassbox could be even better, I would suggest improving the AI.

Currently, I do not have additional pointers for how Glassbox can be improved.

I have been using Glassbox for two years.

I find Glassbox stable in my experience.

The customer support for Glassbox is good because I did not face any issues while using the application, and any issues that did arise were resolved quickly.

Before Glassbox, my organization was using ContentSquare, then they switched to Glassbox and later to Quantum Metric.

In my day-to-day work with Glassbox, I use it to check where the customer is facing an issue and where the error is located. I check the end-to-end transaction that has been captured in Glassbox. I check the API and error code. I also perform impact analysis if there is any error occurring or if I need to provide user impact information. I check the apps being used to determine if the issue is happening on mobile or desktop, or which platform it is happening on, including Android or iOS. These details help in debugging an issue.

A specific example of how I have used Glassbox to solve a problem is whenever an issue occurs, I check the session ID and retrieve the Glassbox session for that particular affected user. I check what the pain point was and what was causing the issue in the application. I check the timeline and logs in Kibana to understand what is causing the issue and what error the customer is facing, then debug accordingly to resolve the issue.

I typically use the AI feature based on the issue that I am debugging, and it has changed the way I work. I ask questions about that particular issue. If a session is significantly longer than expected, I would ask for steps to replicate and I would also replicate it on my end to see what exactly the customer did and if I am able to reproduce it. I ask for a summary of what the user did in that particular session or where the user encountered an error.

Regarding Glassbox's AI capabilities, I have not tried asking any security or governance-related questions in the AI, so I cannot comment on that capability.

Regarding the accuracy and reliability of the AI output, it is good in my experience.

I give Glassbox a rating of eight out of ten because when I was using Glassbox and then I switched to Quantum Metric, I preferred Glassbox over Quantum Metric.

My advice for others looking into using Glassbox is to familiarize yourself with each and every feature because it is helpful. If you use each and every feature, it will definitely benefit your business.

I do not have any additional thoughts about Glassbox; I have covered everything about it.

My main use case for Glassbox is session recordings and struggle score.

For the session recordings, we take examples of user experience flows that we need to improve, and then we check some examples of the recordings. Crossing this with the struggle score, we understand the problems that users might encounter, and we try to enhance the user experience.

I believe that creating reports apart from the session recordings is necessary, since sometimes we cannot check recordings one by one. I think it is a bit of an issue in Glassbox because the way they are presenting the dimensions and metrics is not suitable for creating simple tables with dimensions and metrics. You have to filter or do some strange checks and cross-checks of dimensions just to reach a simple result. Therefore, I believe that in general, the user experience and also the user interface is not user-friendly.

The best features Glassbox offers include mobile analytics, session recordings, and struggle detection. However, these features are quite common in the industry for the same solution. The problem here is that it is not really user-friendly; it is not easy to reach the insights that you want to reach with Glassbox. For big companies such as mine, I feel that some of the configurations that might be scalable for many markets are done manually by the team. When we ask for standardizations of the pages, they are doing the classifications manually, which means we have to give them the Excel and then they have to cluster this manually. For me, this is concerning because you pretend to be a partner for enterprise companies, which of course have complex architectures with many markets and many platforms, so you need to have a more scalable backend.

I am not doing analysis; I am just delivering the solution. However, I am seeing that our markets are struggling with using Glassbox. The kind of struggles the markets are experiencing when using Glassbox is primarily about extracting insights, which is the main aim of a platform. As I have mentioned, it is not easy to create tables; the information is collected, but it is not exposed in a way that is friendly in the platform.

Glassbox can be improved by completely changing the user interface; I feel that there are a lot of duplications and you do not have the main scorecard at the beginning. The metrics that are shown should be simplified so you cannot have rage clicks and clicks exposed separately, but rather just have the total count of events, and then in the dimension, specify which kind of click you are counting, whether it is rage, error, or something else. Currently, everything is exposed as a metric, which means you cannot really have a table with a drill-down over a specific case. Another thing that can be improved is unifying the way you classify some of the dimensions across different markets and different platforms. The page definition is really a mess for us. Every time we have a new page, we have to submit the Excel to them, and they have to do it one by one for all the markets, and this is not really scalable.

There is one point about the connection with other solutions, such as Google Analytics. They need to understand that companies are not small and the solution is not ready to be connected with the complexity of Google Analytics that we have. The markets need to create segments from Google Analytics to Glassbox to split the analysis in Glassbox itself. Since they launched the integration two years ago, we could not link Google Analytics properly, and we are still waiting.

I work for Nespresso and I have been working there since 2021, but in the industry, it has been 15 years.

Glassbox is not stable in my experience. Glassbox often has downtime and technical issues; reports are not loading and there are login errors. Now it has been a bit better, but there is still a lot to do.

Glassbox does not meet my organization's scalability needs. A lot of work is manual, and this is unacceptable for a solution needed in a big company such as Nestlé.

The customer support is nice; we have a quite strict partnership with continuous feedback, workshops, and they are listening to us because the client is quite important, so it is fine.

Before Glassbox, we used SessionCam. SessionCam was acquired by Glassbox, which is why we made the switch.

I have not seen a return on investment from using Glassbox in terms of time saved, fewer employees needed, or money saved.

The pricing, setup cost, and licensing are managed by the mother brand, so I am not really involved; we are taking advantage of the global Nestlé contract.

I was not involved in the RFP; I do not know what other options were evaluated, but I believe ContentSquare was one of them because I saw that some markets decided to have a private contract with ContentSquare.

My advice to others looking into using Glassbox is to not stop at the capabilities provided in the user interface but keep going forward. I found this interview slightly repetitive because some questions are a bit repetitive; you do not analyze the previous answers, so you cannot understand that I have already replied. This can be improved to avoid wasting time. I give this review an overall rating of 6 out of 10.

My main use case for using Glassbox is to see the session activity that has been done by users. It helps me by using a correlation ID or a session ID to understand what actions users have performed. Glassbox records the full session, and through the session, I can trace the full behavior, including all calls that have been placed, which helps in understanding if it's a user behavior error, where the error has come from, or where the request has failed for better analysis.

Recently, I was working with the Marriott International group, where we noticed a lot of bot activity. I went to a particular browser and replicated some malicious behavior by opening the session repeatedly. I took a session ID from the production server, entered that session ID into Glassbox along with the proper timing to minimize search duration, and reviewed the recorded session. During the recorded session, I observed user behavior and traced various calls. I identified a user error when one of the users entered the credit card number incorrectly, helping me to recognize patterns associated with bot activity. Through those parameters, I identified various types of errors and user behaviors.

A second use case involves verifying if I am entering the correct credit card details or selecting the right country. I can review those use cases via recordings to assess sub-calls, which is essential because manual tracing can be challenging. Glassbox assists me in tracing the complete end-to-end flow, allowing me to identify where the error leak is.

Glassbox offers data recording and user behavior tracing within the application. Its features include AI struggle detections and session replays. It also provides insights through heat maps and journey mapping, allowing me to isolate errors, whether they occur in navigation pages, review pages, or find pages. This functionality helps in identifying drop-off points for errors, enabling me to mitigate those issues.

The AI struggle detection feature helps me identify user friction points by analyzing trillions of user interaction data points. It aggregates over 30 distinct behavioral and technical triggers into a singular struggle score, allowing me to know how many users faced issues. The AI algorithm flags patterns such as double clicks, session invalidations, rapid scrolling, and technical errors like Ajax failures and slow page loads. It helps me create alerts for the team, ensuring we address network abnormalities proactively and understand the potential business impact of financial loss.

Glassbox has positively impacted my organization by detecting bot traffic, which helps create better features to prevent that traffic from hitting the application. It enables me to track application issues and convert complex user data into clear and prioritized insights while directly mapping digital frictions to financial losses. For example, we previously encountered a lot of 500 errors, and Glassbox helped us reduce those errors and improve page response times, particularly on a find.mi page that had a high response time due to those errors.

Glassbox can be enhanced by adding direct alerting features that effectively pinpoint user pain points found in GW crowd reviews, better training the system based on past behaviors, and improving session loading speeds, as loading sessions can take too long. Additionally, refining UI navigation to ease usability and reducing lag during complex session replays would be beneficial, along with implementing data retention limitations for improved data processing.

I choose a seven because, despite helping in user behavior and error counting, session replays have been a significant pain point for me, and UI navigation can become complex. Clear documentation with a guided experience would help users understand the features better.

I have been working in my current field for the last seven years.

Glassbox is stable, providing predictable sessions and strong compliance structures, along with a secure and tamper-proof vault and native event web parity.

Glassbox's scalability has been good, effectively managing a significant amount of sessions.

Customer support is occasionally beneficial, offering 24/7 technical troubleshooting, dedicated help infrastructure, and strategic optimization consulting, all governed by strict service level agreements for prompt response times.

Previously, I used Dynatrace but switched to Glassbox for its more interactive views, which aid in user behavior analysis, especially since Dynatrace was cost-prohibitive.

My team found it easy to learn and adopt Glassbox because of its user-friendly interface, requiring us only to input the session ID to navigate and get the needed information easily.

We purchased Glassbox through its marketplace.

I have noted a return on investment since employees previously took two days to manually identify data, but now they complete the same work within four hours thanks to Glassbox, which reduces working hours and translates into cost savings for us.

Glassbox is priced as a premium enterprise platform featuring multi-tiered licensing structures. The enterprise entry point usually ranges from $10,000 to $50,000 per year, offering robust support, session volumes, data retention periods, and deployment models, which justifies our choice of its enterprise version.

Before choosing Glassbox, we evaluated options including Quantum, LogRocket, and ContentSquare among others.

Glassbox facilitates effective communication within the team through numerical reports that developers and analysts can use to improve application performance and draft strong KPI results.

We have a governance agreement governing our use of Glassbox and ensuring compliance with regulations, allowing us access to data under that guidance.

Glassbox ensures data privacy by employing built-in privacy guardrails that prevent collection or exposure of personally identifiable information. By default, it masks highly sensitive information.

Glassbox helps me prioritize which issues to fix first by providing an overview of errors, such as internal server errors and application failures, allowing me to segregate error types and focus on addressing critical issues promptly.

Glassbox integrates by sharing session data and connecting with tools like Dynatrace and other platforms through pre-built connectors and open APIs to facilitate behavior analysis.

Glassbox updates its platform through patching with regular release cadence, featuring weekly improvements and significant feature launches. This process makes it easy for us to stay informed about new installations.

Glassbox supports mobile analytics using a single lightweight SDK, capturing 100% of user sessions on both iOS and Android, mapping behavioral data alongside technical performance metrics without requiring manual event tagging. It also uses native SDK wrappers like Flutter and React Native.

I created a before-and-after matrix to measure the reduction in 500 errors and page response time. Previously, I had around 3,000 errors on my find page, but after using Glassbox to analyze and implement the right blockers, that number reduced to 500. This reduction significantly helped prevent Ajax and network data errors and quantified the revenue lost, making it clear how I improved error management and user experience.

I advise others to understand their needs first, particularly if they focus on user recording behavior within enterprise applications like hotel booking or grocery systems. They should read the Glassbox documentation before starting, which will enhance their experience with the platform. I gave this review a rating of seven out of ten.

Our main use case for Glassbox is for session search and replay, and we also use it for creating dashboards and getting analytics out of it.

We use session search and replay for our customer call center, and we use it for troubleshooting customer issues. We use it for finding production incidents and quantifying the impact, such as how many customers are impacted by any production incident. Our product team uses dashboards when they do any new releases or new features; they want to see the usage, how many customers are clicking or going on a particular flow.

Technology teams use Glassbox for troubleshooting; they are more interested in crashes, struggles, and pain points. Products and business teams are more interested in conversion rates, drop-offs, and when they release a new product. The fraud and investigation team is also interested in identifying bad actors.

The best features Glassbox offers are session monitoring, session replay, and session search.

What makes those features stand out for me is that the user interface is good; it is easy to use. Glassbox has a lot of good training materials on that, and it is very intuitive. It is all web-based, so it is easy to train, and it is easy to find sessions. Glassbox makes everyone's life easy.

Glassbox has positively impacted my organization by playing an important role in production incidents. It helps in quantifying the impact, how many users are impacted or how many users faced a particular problem. The marketing and product team also use it for identifying different scenarios, how the product is being used, and how different customers are using it when they release a new campaign. Glassbox helps in identifying the campaign and how it is being used.

A specific outcome where Glassbox made a difference is that it has reduced the turnaround time for our customer support team. The customer support team extensively uses Glassbox when they are on call with customers, so they are immediately able to pull up their session. They can see the pain points where customers are struggling. Glassbox has reduced the turnaround time of customer incidents, and it is saving a lot of time and money for us.

Glassbox has enabled new business opportunities and improvements in customer experience by providing insights into what the customer is doing, their pain points, and their struggles. Glassbox is a very useful and informative product.

Glassbox has helped us identify trends or patterns in customer behavior that we would not have seen otherwise. It is a session replay tool and focuses on customer journey and customer path. Glassbox helps us in identifying what customers are doing and what their struggles and pain points are.

Glassbox is not the best when it comes to back-end data and capturing things such as server time and response time. If Glassbox could improve on capturing back-end data and server-side data, that would be really helpful.

I believe Glassbox could definitely combine features such as funnels and reports. A few of the data points cannot be captured retroactively, so if Glassbox could increase that, it would help. There are also some limitations with dashboards, such as how many widgets can be added, how many days of data can be retrieved, and how many rows in a table can be inserted. There are little improvement scopes in Glassbox.

Regarding Glassbox's AI capabilities, Glassbox has started their AI model called Gia, which is still catching up; it is getting there. It is useful and informative, but still not up to the best of standards. Sometimes it hallucinates and goes in a loop and does not give the desired result. However, it is definitely helpful in finding results and giving desired outcomes.

I would rate the accuracy and reliability of Glassbox's AI output an eight; it is reliable, durable, and accurate, but still at times it is not as usable and gives a result that is not relevant. Glassbox needs to work on it, and we have notified Glassbox vendors about it; they are working and trying to improve. There is definitely scope for improvement.

I have been working in my current field for five years with Wells Fargo, and with Glassbox, I have been working for almost ten years.

Glassbox is stable.

Glassbox's scalability is very nice because it is on the cloud; it is all managed by the vendor, so it is very easy to scale and adapt.

Customer support is good; we like it. The engineers are very helpful, nice, and knowledgeable.

We previously used Tealeaf before Glassbox; we switched to Glassbox because it is a more modern product with better pricing, better UI, and better training material. Tealeaf was getting obsolete and outdated, and it was getting pricier as well, so we switched.

Onboarding new users to Glassbox is pretty easy. The documentation is good, the training provided is good, and the vendor team is knowledgeable and very helpful.

Glassbox handles updates and maintenance with some challenges, but they are not a lot, and most of them are handled by vendors. They take good care of it; they do regular patching and upgrade activities and they let us know when they do it.

I have seen a return on investment with Glassbox. We have been able to save some time and money because it has reduced the turnaround time for incidents. It has also helped preventively to create some issues and notified us earlier, so it is a very helpful tool.

My experience with pricing, setup cost, and licensing has been positive; we have been a customer for a few years, so we were definitely able to get some good pricing and contract renewal. We are happy with what we are paying to Glassbox.

Before choosing Glassbox, we evaluated other options such as Quantum Metric and Tealeaf.

Glassbox integrates well with other tools or platforms in our organization; we have integrated it with other platforms such as Splunk, Grafana, and AppDynamics because every product has its own limitations and strong suits. Not everything is good in all the products, so we need to create a bigger picture and get the best of all the worlds.

Glassbox has definitely helped my team collaborate more efficiently and share insights across departments; it works with our different teams and also works with other monitoring tools to give a bigger picture and better result. We have implemented it across different teams and the organization.

Glassbox handles compliance requirements for our industry very well; we have identified a lot of PII information and we mask all the fields such as account number, credit card number, social security, and date of birth. Glassbox does a very good job in masking that and not capturing that.

My advice to others looking into using Glassbox is definitely to consider their use cases and thoroughly go through the strengths and weaknesses of Glassbox. It is good for the banking industry, and if you are looking for server-side applications or server-side data, you should try different products. Just focus on strengths and map it with your requirements. I would rate Glassbox an eight overall.

My main use case for Glassbox is customer journey mapping. I use Glassbox for mapping the customer journey by looking between starting pages and end pages, finding where customers veer off, closing gaps, and identifying friction points they may experience, especially through what we call our buying process from start to finish.

The best features Glassbox offers include being able to view sessions in real time, locating errors whether they are script errors or pain points, and finding those issues easily to see where they pop up on the site, then working with development to close and fix those.

During a launch, I use real-time session viewing to make sure that we are aware of customer activities. During launches, I see what customers are doing and if they are taking the paths that we need them to take; I notice situations where we expect them to navigate independently, yet they click for chat or click on un-clickable elements, leading us to make adjustments, especially regarding pricing where people continuously click on the pricing.

I also appreciate that if there is an error or an issue, I use custom segments to find where the problems are and pull up numerous sessions based on those segments in Glassbox. Glassbox is a tool I use all the time; it is one of our most popular tools. It definitely helps save time and improves errors by enabling me to locate issues I was not previously aware of. I can definitely locate issues quicker, especially during tests; I know what to look for and expect, and Glassbox plays a major role in that, particularly during launches when I have people watching sessions in real time rather than waiting to see what happens and looking for it in other analytics tools such as Adobe Analytics.

One improvement I want to mention is that the way I had Glassbox set up was through the data layer; I am a technical person, so I used a Google Dev screen to search within Glassbox for segments rather than using the Adobe API connection, which I think would have been preferable since others have reported better success with that. I give Glassbox a seven because I would have preferred it connected through Adobe rather than through the data layer, which complicates training for non-technical staff to build custom reports; having the Adobe API set up would have made finding Adobe segments easier.

I have been using Glassbox for five years.

Glassbox is stable.

Glassbox is very scalable.

Customer support is great, whether through email or direct interactions with our reps, and they respond almost immediately, even when not located in the US.

I previously used SessionCam, which turned into Glassbox.

They continually improve Glassbox, and I have been happy with the updates.

I purchased Glassbox directly from Glassbox through a representative.

Definitely time was saved, but I would not say I needed fewer employees.

Pricing was fair; while I did not directly manage the budget, we continually renewed at a great rate.

I evaluated other options before choosing Glassbox, but I do not remember which ones since it has been quite a while.

Regarding Glassbox's AI capabilities, I think its governance and security are as good as any other AI; AI is essentially trained by people repeatedly, using data from millions of users, and whoever is training Glassbox's AI has done a decent job. Its accuracy and reliability of output are fairly accurate; I would say mostly accurate, but you always need a human to verify results, as sometimes the outputs are not what I expect, such as the time it pointed me in a different direction instead of where I should have been looking.

My advice for others looking into using Glassbox is to ensure it serves your needs; it is a massive product that provides a wealth of data, so make sure you are looking for and using the right data, as well as understanding how it will shape your site and overall customer experience, and I recommend using the Adobe API if it is available. I gave this review a rating of seven out of ten.