Engauge AI

A one-sentence summary

Engage AI is a browser-based SaaS tool designed for advertising creative testing and market research; it uses the camera on the participants’ devices to analyze their viewing behavior, demographic characteristics, and emotional reactions on a second-by-second basis, and then presents the results in the form of actionable reports.

Tool Introduction

This product was developed by XCINEX Corporation, and its target users include brand owners, content creators, and market research firms. It focuses not only on metrics such as views, click-through rates, or completion rates, but also on the changes in expressions and attention patterns that may occur at different moments during a video.

Researchers create activities and upload the videos to be tested; participants view these contents after granting permission for use of their camera in a compatible browser, and the platform then aggregates data on emotions, demographics, and level of engagement. Such results can help in refining creative ideas, but they should not be regarded as an accurate indication of an individual’s true psychological state.

Main functions

Advanced video content analysis

Users can upload advertisements, promotional videos, or other video materials to create testing campaigns. The system links the viewing process with a timeline, enabling teams to determine the audience’s reaction to specific shots, text captions, or changes in rhythm.

Emotional engagement per second

The statuses listed on the official website include neutral, happy, sad, scared, angry, surprised, disgusted, looking away, and scrolling forward. The results are presented in seconds, making them suitable for comparing segments such as opening sequences, key features, transitions, and call-to-action messages.

Demographics and viewer count detection

The platform can use cameras to determine the number of viewers and analyze demographic factors such as age, gender, and ethnicity. These demographic characteristics are inferred by algorithms rather than through any form of identity verification, and their accuracy and reliability can be affected by lighting conditions, camera quality, the composition of the sample, as well as differences among various groups.

Event management

The team can organize research activities, manage videos and the participation process, and conduct tests based on various ideas or target audiences. The price list on the official website lists items such as participation restrictions, access controls, age limits, role management, and rewards; however, the public page does not clearly show whether each of these options is selected or not.

Customizable surveys

The official FAQs list customizable survey options as part of the analysis capabilities. Researchers can combine camera behavior signals with survey responses; however, the specific types of questions, logical transitions, and export formats are determined by the settings page available after logging in.

Interactive report

The report is used to observe changes in emotions and level of engagement along the video timeline, as well as to identify segments that require further verification based on the results of the analysis. The public version does not disclose information regarding the download of raw data, statistical tests, or the format in which the reports can be exported; it is necessary to check these details according to the actual research procedures before making a purchase.

Brand and team control

The pricing section lists features such as custom thank-you messages, brand identifiers and themes, user roles, and browser statistics. As the icon status in the currently available table is incomplete, these items are not assigned to any specific package in this document.

Analyzable signals

Complete workflow

1. Register a customer account and select a trial plan or a package that suits the number of team members, video quality requirements, and point needs.
2. Create a research activity, upload videos for which testing and display rights are available, and set the conditions for surveys, access, and participation.
3. Send a participation link to the target sample, and before starting, clearly explain the purpose of the camera, the type of data collected, the methods for withdrawing from the process, and the reward conditions.
4. Participants use a compatible browser to authorize the camera and view the content in its entirety, with the system performing real-time monitoring within the browser.
5. After the event, review the timeline, mood, demographic, and engagement reports to identify sections that need to be revised or retested.
6. Repeat the testing with new creative versions and similar samples, and verify the conclusions by using questionnaires, interviews, conversion data, or traditional research methods.

Usage tutorial

Create the first creative test

1. Log in to the customer portal and create an activity, assigning a clear name to the project as well as defining the research hypotheses.
2. When uploading a video, make sure that its clarity does not exceed the limit set by the current plan, and check the playback duration as well as the orientation of the video.
3. Select the indicators and survey questions that need to be examined, in order to avoid testing too many hypotheses at once within a single activity.
4. Set the eligibility criteria, age requirements, access controls, and details on any acknowledgments or rewards upon completion.
5. First, invite a small number of internal users to conduct a trial run in order to check whether camera authorization, video loading, and the report timeline are functioning properly.
6. After official release, monitor the valid sample size and remaining points; analysis will begin once the target sample size is reached.

Interpreting second-by-second emotion curves

1. First, mark the key time points such as product exposure, price information, plot turns, and calls to action.
2. Observe whether multiple participants exhibit consistent changes during the same period of time; do not draw conclusions based on a single abnormality.
3. Examine the behavior of diverting the gaze or scrolling separately from the video rhythm, page interactions, and technical issues.
4. Compare the differences using the predefined sample groups, and record the sample size and missing data for each group.
5. Cross-verify the model labels with survey responses, open-ended feedback, and actual business metrics.
6. After modifying a key variable, retest it to avoid being unable to determine which change is responsible for the difference.

Conduct compliant participant testing

1. Provide a concise consent statement before turning on the camera, outlining the purpose of the monitoring, the way in which the data will be used, and the expected duration.
2. It explains that participants can withdraw by closing the survey window, and it should be made clear in advance whether such withdrawal will affect any rewards.
3. Minors must have the consent of a guardian and meet the age requirements of the relevant jurisdiction.
4. Demographic data or emotional inferences must not be used for high-risk automated decision-making, discriminatory screening, or the creation of expanded personal profiles.
5. After the event concludes, the scope of data retention is controlled in accordance with the research objectives, and requests for access, correction, or deletion are addressed.

Which users are it suitable for

• Brand and advertising team:Test different versions of the opening statement, key features, and call to action before increasing media spending.
• Market research firms:Include the browser camera feed in remote creative testing and analyze it together with survey data.
• Content creator:Identify the sections in which the audience might lose interest, and then optimize the pacing of the narrative and the order in which information is presented.
• Product Marketing Team:Compare the audience reactions to demonstration videos, landing page videos, or concept videos for new products.
• Creative agency:Provide customers with the pre-test results of various ad versions along with suggestions for improvements.
• Research and Insights Team:Explore the relationship between expressions, attention behaviors, and subjective feedback, but this requires a rigorous research design.

Typical use cases

• Compare the differences in response regarding brand exposure, key selling points, and the call to action between the two ads.
• Identify the segments in long videos where viewers look away or scroll forward more frequently.
• Before official release, conduct a preliminary test on small samples of different versions in terms of editing, soundtrack, subtitles, and rhythm.
• Trends in different demographic groups are used as clues for research, with reasons explained through questionnaires or interviews.
• Generate timeline insights for research reports to help creative teams focus on actionable improvements.
• Use repeated tests to verify whether the revisions have improved the target segment, rather than relying solely on the team’s subjective evaluations.

Product advantages

• Participant testing can be carried out directly in the browser, without the need for participants to install any dedicated desktop software.
• Mapping the reactions to specific time points in the video makes it easier to identify the segments that need to be modified, compared to relying solely on the overall score.
• It covers multiple dimensions of observation, including emotions, attention and behavior, number of viewers, demographics, and survey information.
• It provides activity management and interactive reports, facilitating the transition from material distribution to result analysis.
• The trial version provides access to the enterprise suite, allowing users to test whether the actual workflow is suitable for their team before making a payment.
• The official website clearly states that, by default, no images, videos, or audio of the participants are captured or stored, and it says that the detection takes place within the browser.

Limitations and interpretation of results

• Facial expressions do not correspond one-to-one with true emotions; model labels can only serve as research signals.
• Inferences about age, gender, and ethnicity can be affected by training data, skin color, lighting, camera angle, and cultural differences.
• Camera quality, browser permissions, the network connection, and the participants’ environments can affect the proportion of valid samples.
• The method used to recruit samples may introduce selection bias, and the results obtained from a small sample cannot be taken as representing all consumers.
• The public page does not disclose the validation data, confidence intervals, error rates, or results of independent audits for each model.
• For some of the advanced features listed in the price table, only the names of those features are shown; the current page cannot reliably determine the relationship between the different options.
• The official website only reveals the monthly pricing; although the terms mention annual payment, the exact amount for annual payment must be confirmed on the settlement page.

Privacy, Consent, and Data Training

According to official policy, testing for participants is carried out in real time within the browser by default, and images, videos, or audio are not transmitted to servers by default. Customer information, payment details, platform usage data, anonymized survey data, and data processed by service providers all fall under the category of data that needs to be managed.

It is important to note that the terms also retain the right to save anonymized data, as well as images or videos, in order to improve the model. By default, such data is not saved, provided that this does not conflict with the collection of specific training data; however, the organizers must define the conditions under which this action is taken, the mechanisms for obtaining consent, the duration for which the data is retained, and the procedures for deleting it.

Prices and packages

As of August 22, 2026, the public areas of the official website display the following monthly pricing in US dollars. The trial period is either 7 days or 15,000 points, depending on what is available first, and it allows access to all enterprise features; the final costs and taxes are indicated on the customer portal’s settlement page.

Cancellation and refund

A monthly subscription can be canceled at any time; once canceled, it remains valid until the end of the current billing cycle, with no refund for any unused months. An annual subscription can also be canceled, and according to the official terms, the service continues until the end of the current month, with a proportional refund for any unused months.

Key criteria for choosing a package

• Estimated number of activities, sample size involved, and points required for each analysis.
• Is it necessary for uploaded videos to have a resolution of 720P, 1080P, or 4K?
• It is the number of team members who need access to the backend, not the number of survey participants.
• Is brand customization, access control, roles, rewards, and integration with third-party surveys required?
• Is an annual contract, customized events, or additional points needed?

Platform and usage conditions

APIs, SDKs, and open-source status

As of the time of verification, the official website did not provide any API references, official SDKs, a developer platform, or a verifiable official GitHub repository. The code repositories, management tools, and browser extensions that appeared in search results under similar names belong to other projects and cannot be used as evidence of Engauge AI’s open-source nature.

Engage AI is a commercial SaaS product; its source code and emotion recognition models are not made public, so it cannot be classified as an open-source tool. If a company needs the ability to export raw data, integrate it with other survey platforms, use single sign-on, or take advantage of automated interfaces, it should request technical details before making a purchase.

Differences from traditional creative testing

Basic information

Recommendation score

Recommendation score: 4.0 / 5. It is suitable for brands and research teams that need to quickly identify key moments in video creativity; the second-by-second timeline, browser integration, and multi-dimensional reports offer significant practical value.

The deductions stem mainly from the limited availability of data on model accuracy, the risks associated with the fairness of demographic inference, and the need for further clarification regarding the relationship between local processing and the retention of training data. Serious research should use it as an auxiliary tool for measurement, rather than as the sole source of conclusions.

Frequently Asked Questions

Is Engage AI free?

It is not a tool that remains free forever. New users can try it out for 7 days or by using 15,000 credits, depending on which condition is met first; after that, a subscription or an additional purchase of credits is required.

What features are included in the trial version?

The official website states that a trial access allows use of the full enterprise suite, but the actual points, restrictions on offers, and available regions are still determined by what is displayed on the portal after registration.

Do participants need to install software?

No special applications need to be installed; the service operates through compatible browsers. Participants must allow access to the camera and keep the survey window open.

Are the participants’ videos saved?

Officials state that, by default, images, videos, and audio are not captured or saved, with the detection taking place within the browser; however, the terms reserve the right to save anonymized data as well as certain images or videos for model training, and event organizers must confirm the specific settings and obtain consent prior to conducting tests.

Is microphone analysis already available?

The official specifications describe the microphone as a feature that might be added in the future; the current version uses a camera instead. Audio emotion analysis cannot be listed as a feature that is already available.

Are the results of emotion recognition reliable?

The results can help identify times points that warrant further investigation, but facial expressions do not equate to true emotions. Important decisions should be made by taking into account sample design, questionnaires, interviews, as well as actual performance data.

Is Chinese supported?

The current official website and public materials are primarily in English; there is no confirmation of a complete Chinese version of the backend or Chinese-speaking customer service. Whether Chinese-language videos can be used properly is a separate issue from the language of the interface, and it is advisable to test this through trial offers before making a purchase.

Are APIs or SDKs available?

As of the time of verification, no public APIs or official SDKs were found. Teams that need programmed data import, export of raw data, or integration with third-party datasets should contact the authorities for confirmation.

Is Engage AI open source?

No. This product is a commercial, closed-source SaaS solution; no official public source code, SDKs, or model weights have been found.

Can it be used in research on minors?

The participation requirements state that participants must be at least 18 years old; those under 18 need the consent of their parents or guardians. Organizers are also required to comply with the stricter regulations regarding child privacy and biometric data in the location where the study is conducted.

What rights are required to upload an advertising video?

Customers must ensure that they have the permission to test and share content, music, portraits, and other materials, as well as obtain any necessary approvals from third parties featured in the videos. The platform’s terms of service place responsibility for these matters on the uploader.

Summary

Engage AI combines remote video testing, camera behavior analysis, demographic inference, as well as survey and interaction reporting within a browser-based workflow, making it suitable for identifying creative issues before a formal launch. There are three subscription plans based on points, clarity, and the number of team members, and limited-time trials are also available.

Its results should be viewed as clues for further research, rather than an indication of what someone is thinking or a basis for individual judgments. Before using it, the team must obtain informed consent, assess algorithmic biases, verify the training data, and design the research properly; furthermore, cross-validation should be carried out using questionnaires, interviews, and business data.