The Comparison Tool is the easiest way to compare similar images and determine which performs best.
- Go to https://app.neuronsinc.com/predict to launch Predict.
- Navigate to the images you want to compare.
- Select assets by ticking the checkbox either on the top left corner of the thumbnails or on the left side of the row, depending on whether you are in grid view or list view.
- Tip: You can also select all items in a folder by clicking on the checkbox on the left side of the folder toolbar. However, keep in mind that you can only compare up to 6 images at a time!
- As soon as you select any items with the checkbox, new options will appear in the middle of the toolbar. The numbers on the buttons show you how many items you have selected.
- Click on the Compare button in the middle of the toolbar to compare the selected images. In the following example, we are comparing 2 variations of the same ad.
- Your selected images will open in a detailed comparison view. Here you can compare cognitive scores, heatmaps, and custom AOIs, and determine which asset performs best.
You can also create a comparison by opening up the prediction details page of an image and clicking on the Compare button in the top right corner of the screen!
Adding and removing compared assets
If you have already created a comparison, but decide you want to compare a different set of assets, there's no need to start over! Follow the steps below to change which items to compare for your current comparison:
- Click on the X Files Selected button in the top right corner of the comparison page.
- In the window that opens, you can navigate around by clicking on folders or the breadcrumbs in the top left corner. Select and deselect which assets to compare by clicking on them.
- Click on the Compare X Files button to compare the newly selected assets.
- In the comparison page, you can also remove images from the comparison by clicking on the button in the top right corner of the asset.
The comparison tool is currently only available for image assets only.