You can now use self-service to produce high-quality image previews with optimized language. This capability works very similarly to generating text content, only you’ll start with an image first. You’ll use a full-screen workflow for uploading images, selecting touchpoints, and adding content inputs.
How to Generate Content Previews Using Imager
Click ‘New Content’ as usual in the upper righthand corner of Portal, and opt to ‘Create Content Yourself.’
Select the relevant marketing channel. (Imager is available for all channels except Email Subject Lines and SMS/EMS.) Then, opt to ‘Start with an Image.’
Upload an image in PNG format. You can do this by clicking to upload or dragging into the upload area.
Select your touchpoints. This is a key step as it helps our AI break down the types of language in your message and their purpose. To label a touchpoint, simply click and drag your mouse across the text.
A dropdown menu prompting you to ‘Select Touchpoint Type’ will appear. Click to designate if the portion of text is a headline, preheadline, CTA, etc.
When you’re done, your image will be tagged with all touchpoints and should look something like this. You can click ‘X’ to remove any touchpoint tags if you need to rearrange them. When you’re done, click ‘Next.’
Add additional detail. In the ‘Additional Refinement’ window, you can input information about the target segment, words and phrases our outputs should always use, and any other content considerations that add context. The more specificity you provide, the better our AI will be able to understand your desired message. Click ‘Generate’ when you’re done.
Review language options as you normally would. Your control image will appear at the top of the next screen with the other language options listed below. All options will have performance prediction, brand alignment, and readability scores as usual.
If you’d like to see more or different options, you can:
Update your ‘Content Refinement’ fields and ‘Generate’ again. You can also ‘Generate’ with the same inputs as many times as you want!
Update your ‘Style Guidelines’ to try out different style sets or tones and ‘Generate’ again. You can learn more about style sets here.
If your control image changes, keep in mind that you can edit your input image. This allows you to edit the text of your original control and regenerate options.
Last but not least, you can generate AI Alternatives to see even more options per touchpoint. First, highlight the desired text. Then click ‘AI Alternatives’ and new options will appear with their updated Performance Prediction scores. Click any option to swap it in as usual.
Make final image edits. To do this, click a touchpoint and then click the ‘Edit’ icon.
From here, the editor will appear so you can revise as you wish! Fonts are automatically matched to the closest Google Font. Formatting options include bulleted lists and background colors so you have even more flexibility in customizing your image.
‘Save’ all desired Variants to review with your team. Saved image variants now display previews alongside their corresponding touchpoint text. This allows users to proofread and provide feedback before approving Variants. Clicking an image preview opens a full-screen view, showing the language in full visual context.
If you want to share Variants with teammates who do not use Portal, you can click the ‘Download’ icon in the lower righthand corner of the image to download a PNG file.