I want to show you the thing that changed how I work more than any other single feature, and it is not a filter or a brush. It is the moment a design stops being a picture and becomes a form that fills itself in.
Here is the idea in one sentence: you build the layout once, you tell a few layers where to get their content, and from then on every post, product or page renders its own version. No copy and paste, no “let me just swap the headline”, no forty files called social-final-v3.
What a binding actually is
Take any text layer, open Properties and look for Dynamic content. Instead of typing words into the layer, you point it at a source. The post title. The excerpt. The category. The publish date. The author name and bio. The site name and tagline. If you run WooCommerce, the product price, the sale price, the discount, the product image. Every ACF field. And if none of those fit, any custom field at all through its meta key.
The layer keeps its font, its color, its effects, its position. Only the words change. That is the whole trick, and once it clicks you start seeing templates everywhere.
Image layers work the same way. Bind one to the featured image, the brand logo, or a generated background, and the picture arrives from the post instead of from your desktop.
The two problems everybody hits, and how they are solved
The first time people try this, two things go wrong within about a minute. I know because both of them went wrong for me.
Long titles break the layout
You design with a nice short headline. Then a real post arrives with fourteen words and the text runs off the canvas, or it wraps into five lines and eats the picture.
The fix is a minimum and a maximum font size on the bound layer. Draw the box the size the design needs, then say: this text may live anywhere between 34 and 72 pixels. Short titles get the big size and look confident. Long ones shrink until they fit. Anything that still overflows at the minimum gets shortened with an ellipsis rather than escaping the frame.
That single setting is the difference between a template that works on twenty posts and one that needs supervision on every single one. It is the least glamorous feature in the whole system and I would not give it up for anything.
Pictures arrive in the wrong shape
Featured images are portrait, landscape, square, and occasionally a panorama someone uploaded from a phone. Drop those into a fixed slot and you get squashed people.
Bound image layers have an Image Fit setting. Cover crops the incoming picture so it fills the box completely, keeping its proportions. Fit letterboxes it so nothing is lost. And an anchor decides which part survives the crop, so you can say “keep the top” and portraits stop losing their heads. The same setting doubles as a non-destructive crop on any ordinary image layer, which I use constantly.
The detail I love most: empty means invisible
A variable with no value hides its layer. That sounds like a footnote and it is actually a design tool.
Build a discount badge, bind its text to the discount percentage, and it appears only on products that are actually on sale. Everything else renders without it, automatically, with no conditional logic and nothing to maintain. Bind a subtitle to a custom field that only some posts fill in, and the layout closes up on the ones that do not.
Once you know this, you start designing templates with optional parts on purpose. One template quietly becomes four.
Preview with a real post while you design
Designing against dummy text is guesswork. View, then Preview with Post, drops actual content into your design while you keep working, so you can see what a genuinely long title does before you save anything.
It is non-destructive. Your placeholder text comes back the moment you leave the preview, and undo history is untouched. I flick it on and off constantly, usually against the worst post I can find, because a template that survives your ugliest headline survives everything.
When it holds up, File, then Save as Dynamic Template.
The same idea works for anything with a shape and a list, not just posts. The merge below takes a template with placeholders and a table of values and produces one result per row.
Placeholders take defaults and filters, a block can be made to disappear when its field is empty, and the whole set comes back as one text block or as separate files in a zip. It is the paper version of what a binding does inside the editor.
Template and data merge
Build the layout once, put your rows underneath, and get one finished result for every row. Everything happens in this browser tab: a dropped file is read on your own machine and nothing is uploaded.
Filters after the name: upper, lower, title, trim, slug, number, date (2024-05-17 becomes 17 May 2024), escape, attr, raw. number groups the English way, so 1234567.5, 1234567,5 and 1.234.567,5 all come out as 1,234,567.5: the last separator in a value is the decimal point when one or two digits follow it or when it is the only one of its kind, every other separator is grouping and is dropped, and a value that does not read as a number is left as it stands. Anything else after the bar is the default value for an empty field, as in {{price|On request}}. {{#column}} keeps a block only when that field is filled, closed with {{/column}}, and {{#}} is the row number.
or press Enter to pick one. It is read in this tab and replaces the rows above.
Columns found: none yet.
Detected: comma
One file per result in the ZIP.
The template looks like HTML, so values from the data are escaped.
The rendered preview sits in a sandboxed frame with scripts and network access switched off, so markup that comes in with your data can only ever draw inside that frame.
Then run it over everything you have
This is where it stops being a nice idea and starts saving days. Automation, Images, Featured Images: search and filter your posts, pages or products, select as many as you like, pick a template, press Process. If your content already lives in a sheet rather than in WordPress, the same thing works from a CSV.
The bulk run is the part that needs WunderPaint Pro. Building the template, binding the layers to your content and previewing it against a real post are in the free plugin.
For each entry the placeholders resolve, the design renders, and the result is set as the featured image. If you want, a background is generated from the content of that specific post, optionally steered by a style description, a reference image and your brand colors, so a hundred posts do not end up with a hundred copies of the same stock photo.
Three things about this run that I appreciate more every time I use it:
- The output keeps its layers. A generated image opens from the media library as a full document, not a flattened JPEG. If one of a hundred needs a nudge, you nudge it.
- Batches survive a reload. Progress is stored on the server, so closing the tab or a shaky connection does not mean starting over.
- Per-category template mapping. Assign a template per category once and a bulk run picks the right one by itself, so your tutorials and your news posts can look related without looking identical.
With Rank Math or Yoast active, the same run can set the image as the social preview and write an SEO title and description per post. And Also save as PDF drops a print-ready version next to the image, which turns out to matter more often than you would think.
Where I would start
Do not build the perfect template first. Build a plain one: a background, a bound title with a sensible size range, a bound featured image set to Cover, your logo. Preview it against three real posts, including the one with the longest headline on your site. Fix what breaks.
Then run it over ten posts, look at the results, and adjust. By the time you point it at two hundred, you already know exactly what it will do.
The payoff is not only speed. It is that your archive suddenly looks like it was designed by one person on one afternoon, because in a sense it was.
