What are AOIs
‘Areas of Interest’ refer to any visually important element within a piece of creative content, such as titles, claims, calls to action, etc. AOIs help measure user engagement, visual hierarchy, and effectiveness in guiding attention. Tagging the key informational areas helps to understand more about them and their impact within the creative.
Creative content can have many AOIs meaning manual identification and tagging for analysis requires effort, focus, accuracy and consistency.
Given the manual effort involved it can be time consuming to get this right - in the Dragonfly Studio we are automating the process as well as giving you all the previous functionality.
How it works
You might be wondering how it is able to detect AOIs automatically. Well, it is a 2-step process.
1. AOI Identification
The element detection model scans the uploaded asset, recognising distinct visual components such as text, logos, and product imagery.
This step ensures that key areas of interest are isolated without applying predefined labels.
2. AOI Classification
Once detected, AOIs are sent for further analysis, where they are categorised based on a structured taxonomy.
This classification informs predictive insights, ensuring meaningful evaluation of how audiences engage with different elements.
How to Manually Add an AOI
Three simple steps!
Click the “+ AOI” button in the top left corner of the review pane.
Choose from the taxonomy list which AOI best represents your missing elements.
Reposition and resize the new AOI to fit the creative you are working with. Try and keep the edges tight around the element.
How to Delete, Edit, and Rename AOIs
When reviewing the elements that have been detected, you may notice an AOI is irrelevant to the test you are conducting, in the wrong region, or that it has the wrong label. In which case you can find the relevant button on the right hand side review pane.
Polygons Vs Rectangles
The latest iteration of Studio offers rectangles as the sole means of highlighting Areas Of Interest. Why?
Minimal Score Differences
Studies comparing rectangular and polygonal AOIs have found that fixation and dwell time metrics remain largely consistent, with only minor variations that do not significantly affect conclusions.
Holmqvist et al. (2011) in "Eye Tracking: A Comprehensive Guide to Methods and Measures" highlight that small deviations in AOI boundaries do not substantially impact aggregated eye-tracking metrics.
Consistency and Reproducibility
Rectangular AOIs reduce variability in results between different users and studies.
Research from Tobii and SR Research suggests that using simpler AOIs, such as rectangles, improves comparability across studies because they standardise how data is captured and interpreted.
Practical Usability Over Precision
While polygons allow for a more precise fit, they introduce subjectivity and added complexity in defining AOIs, leading to inconsistencies across studies.
By using rectangles, we enhance the reliability of our tests by reducing variability between users and studies. It also simplifies the process, making AOI selection faster, more intuitive, and less prone to inconsistencies. Ultimately, this means cleaner data, smoother workflows, and results that remain actionable and comparable across different tests.
Bergstrom & Schall (2014) in "Eye Tracking in User Experience Design" argue that for most UX and marketing studies, the goal is to capture general engagement patterns rather than pixel-perfect AOI delineation.