Eye tracking is a method of measuring eye positions and eye movement. This is of great interest because there is a direct correspondence between where people look and where they focus their attention. In research, eye tracking is used to provide you with insights into users' attentional processes and delve into their subconscious response to your products. It allows you to experience your product through the eyes of your customers.
We combine artificial intelligence with considerable eye tracking experience to offer the best smartphone eye tracking available on the market for UX and market research. Our eye tracking is based on the analysis of images from customers selfie cameras and does not require any specific hardware.
Not at all! Oculid's mission is to make eye tracking accessible and affordable to everybody. That is why, with our software-only solution, you won't have to buy any additional eye tracking devices. Oculid's eye tracking technology is currently available for all Android phones with Android 8 or higher.
Positioning is a crucial step when getting ready for an eye tracking study. The participant is asked to keep the phone in a position similar to taking a selfie. Keeping the face in the frame of the phone enables the tracking of their eyes at all times. It takes only a few seconds. Similar to calibration, the purpose of positioning is to increase the accuracy of the eye tracking technology. Participants should maintain that position throughout testing. As the positioning can be perceived by some users as uncomfortable, we recommend testers to start the test in a comfortable, seated position and hold the phone in a natural way as much as possible.
During calibration, participants are asked to follow a moving dot moving across the screen. This takes about 10 seconds. The simple procedure allows Oculid to learn more about the characteristics of the persons' eyes, and is necessary for the eye tracking technology to work as accurately as possible.
A few conditions can negatively impact the quality of the eye tracking data, namely wearing glasses with shiny frames and completing the test under particularly bright lighting conditions that lead to squinting. Fortunately, these are easy to avoid. To optimize your eye tracking results, participants are advised to take off their glasses (when possible) and complete the test in a place with indirect natural or artificial light.
A fixation can be defined as a state when the eye remains relatively still over a period of time, for example, when the eye temporarily rests on a syllable during reading. Fixations are highly correlated to a person's focus of attention. Therefore, longer fixations on a specific object will indicate that user's attention has been caught, whereas shorter fixations will reveal areas of your product that are of lesser interest to the tester.
Eye-tracking metrics should always be interpreted with respect to the task that your testers accomplished. In order to help you with the interpretation, here are some examples to get you started:
- In general, longer fixation durations indicate deeper mental processing and more effortful cognitive processes, whereas shorter fixations indicate higher mental workload if people fail in accomplishing a task.
- Many short fixations can be indicative for poor design, as users are searching for something specific, but can’t find it.
- More fixations within a specific area could indicate the high semantic importance of that area. Or conversely, in a search website task, this might indicate that your testers have trouble interpreting the information in that area.
We recommend to compare the results between your tasks to get a better idea which tasks might have caused more trouble than others. Additionally, it is very useful to compare the results between participants, before diving into a deeper analysis using the video overlays of especially high or low performing testers.
You can test anything that has a URL including mobile websites, web versions of mobile apps, click prototypes like Figma or InVision.
You can test a Figma, InVision, Sketch prototype or any other prototype that can be accessed via a web link. We plan to integrate with these design tools in the near future to further improve your experience in prototype testing.
Simply click on "Create test," either from your 'Overview' page or from the sidebar. A few fields are mandatory to create a test: The test name (can’t be the same as an existing test), the number of participants, at least one task filled with a task title, a task description and a valid selected URL or media (image or video). You can create several tasks for the participant and organize them in the order you wish them to complete them. You can choose whether you want to ask your participants to think aloud or not, meaning what they say will be recorded during the test. After clicking on “Create Test”, you will be redirected to the Overview where you will see your newly created test.
The number of tests that you can create depends on the pricing plan that you choose. Please contact our sales team at firstname.lastname@example.org to find the perfect plan for your needs.
Yes, if you are unsure of the final details of a test you can already write the information contained in your test. All information and media will be saved in a draft version. You can come back at anytime to finish and create your final version of the test. Every new test will be auto-saved as a draft, so in the case you go back inadvertently to the overview page, the test will be easily recovered.
Yes, if for any reason you would like to wait for new participants to access the test, you can “pause” the chosen test, making the test key inactive until you resume the test. Note that when created, a test is automatically active and new participants can access it with the test key you sent them.
A test will automatically be Completed when your maximum set number of participants uploaded their results. If you have already enough participant results and want to close the test, you first have to set it to Paused in the test overview section and then you will be able to Close it. You will then receive a confirmation message that the test key has been made inactive for all new participants trying to complete this test.
Every task, description or source included in a test can be edited as long as no participant took part in your test. The test has to first be set to inactive in order for you to make any changes. This way no participant can access the test you want to change until you reactivate it. To set your test to inactive, open the settings menu of the test three dots and click on Pause. When entering again the settings of an inactive test you can now click on Edit to make changes to your test.
The number of participants per project depends on the plan that you choose. Please contact our sales team at email@example.com to find the perfect plan for your needs.
The recording time per project depends on the plan that you choose. Please contact our sales team at firstname.lastname@example.org to find the perfect plan for your needs.
We recommend you to break your test down into meaningful tasks and using specific instructions such as "navigate to your account page and find your purchase history" rather than providing broad and general statements "explore the website." Keep the task descriptions short and easy to understand for the testers and avoid ambiguous wording. This way, it will be easier for you to analyze and interpret your data with respect to the specific tasks and to compare the results between testers.
You can define as many tasks as you want, but keep in mind that the recording time per project is limited and depends on the plan that you purchase. The longer the test, the longer the it will take for the participant's results to upload.
Each task can consist of a different URL, image or video. You an set one URL for a whole test or a series of tasks including multiple URLs.
You can also add a link to an external questionnaire (Typeform, SurveyMonkey, Netigate, Feedier, SurveyGizmo) that the participants will be asked to complete at the end of the test. The participants will be shown the link after uploading the Oculid test results. In order to use this feature you have to paste your existing link when creating the new test. This feature is just a redirection and there is no automatic mapping between Oculid dashboard and the questionnaire results. If you have further questions on this, please write us an email at email@example.com.You can include a Quick Survey to start the test. Set three demographic metrics: age, location and gender, or create your own custom question to ask your testers.
Yes, with our A/B testing feature you can define which tasks of your test you want to separate into two groups of participants. In order to get the same testing conditions within the two groups, your participant don’t know they are participating in an A/B test. The tasks are named the same and have the same description. The only difference between the A and B task is the URL or media source you will provide.
To have the optimal experience using the web interface, please use Google Chrome or another Chromium-based browser (e.g. Brave).
The following image formats are supported: .jpg .jpeg .png .gif .svg.
You can upload images up to 5MB and videos up to 20MB.
For each task we enable a “pause recording” feature. It means you can (and probably should!) create a dedicated task that won’t be recorded, to protect the data of your tester. On the app side, the tester is showed during the test that this specific task is not being recorded. If a task is marked not to be recorded, the “Thinking out-loud” feature can however still be active to record the microphone during the task. Gaze position and tap metrics will be available in the dashboard for the task, but you will neither see the screen recording of the task in the video replay, nor the heatmap of the task.
In general, other apps can be opened in parallel. However, they must not actively use the selfie camera, so for example running Zoom in parallel is not possible.
Once you have finished creating your test, you will be redirected to the Test Overview section where you can click to see the invitation (under Actions and then See Invitation). The invitation text contains all information that your participants need to log in and complete your test. Simply copy it and send the invitation to your participants. To begin the test, participants will have to download the Oculid app and enter the code from your invitation. Each test starts with a calibration.
You can set the test as inactive by pausing the access to your test. The test code will then be made inactive. Any participants in the midst of a testing session on the Oculid app will still be counted as a tester as soon as they upload their test results. Please note that you can resume and activate your test at any time. You can also change the status of a test to Complete by closing it in the settings menu of a Paused test. This action is irreversible and the test cannot be resumed at a later point; the test key will not work anymore.
You can’t make any changes in a test after the first participant has taken part in it. If you don’t want to delete the whole test and start over, you can delete the participant/s in the Dashboard so that you have 0 participants, then set the test to inactive and edit the test. Please note that you will lose the results from those deleted participants.
The app can be downloaded from the Google Play Store here . The app is currently available in the following countries: Austria, Belgium, Cyprus, Czech Republic, Denmark, France, Germany, Greece, Ireland, Italy, Lichtenstein, Luxembourg, Malta, Netherlands, Poland, Portugal, Spain, Sweden, United Kingdom, and the United States.
All Android smartphones with Android 8 or higher are supported.
We highly recommend that participants ensure a stable high speed internet connection when completing the tests. Wi-Fi is not strictly necessary, but is recommended as the results data volume will scale up depending on the duration of your test.
With Oculid's testing platform, you can analyze the following data:
- Task Duration (seconds), Device that was used, number of Taps, Fixations, Mean Fixation Duration and when the action was created
- Eye-tracking metrics in the form of gaze replays (i.e. overlays of the eye positions on screen recordings) and quantitative metrics of visual attention and search
- Heatmaps on images
- Screen recordings of every user who uploaded the test results
- Behavioral metrics like taps/swipes, time on task, task success
- Thinking aloud recordings (audio recordings of what test participants said while completing your tasks)
- Task ratings
- Demographics answers from a quick survey
Unfortunately not, but right after your testers finish their test, their screen recordings will be available with the option to overlay gaze and taps for visualization.
Upon collecting data, you will have access to it in Oculid's dashboard where you will receive a qualitative and quantitative summary of your data. You can also download the quantitative data and visualizations to use them outside the dashboard.
We are committed to making your analyses as easy and meaningful as possible. That is why it is possible to directly compare behavioral and eye tracking data in the dashboard by creating interactive charts that can be downloaded to be used in reports or shared with others in your team. You can also download that quantitative data itself.