[97] Can Squarespace Do That? How to Analyze What's Possible in an AI Design

It's a Tuesday afternoon, and there's an email at the top of your inbox from someone you'd genuinely like to work with. One sentence, one link: Can you just build this in Squarespace for me?

You click. What opens was very obviously generated with AI, but that doesn't automatically mean slop. It might be a pretty good design. A big hero section with a background image, some overlapping shapes, typography sitting somewhere you wouldn't have thought to put it. The color contrast isn't perfect, but it looks finished. It's not a napkin sketch or a Pinterest pin from 2013. It looks like a real website, and your client made it in a couple of minutes.

There's one word in that email doing an enormous amount of work, and it's just. Because what they sent you is an idea, not a plan. Your job isn't to create the idea anymore. Your job is to figure out what's possible in Squarespace and what isn't, and that takes far more time than your client realizes.

Here's what's actually happening inside those AI tools. When you ask one to design a homepage, it's trying to give you something that looks right. It arranges text, images, shapes, color, and spacing into something that reads as a finished website, and it's gotten good at that. But unless you specifically tell it to design within the constraints of Squarespace, it isn't thinking about how the design gets built. It's not thinking about Fluid Engine. It's probably not thinking about page sections at all, or content blocks, or the fact that your mobile layout isn't a smaller version of desktop; it's its own arrangement with its own rules. So it puts text where text looks good. It overlaps two elements because the overlap looks interesting. It creates space that works beautifully in the image it generated, even if that space doesn't exist inside the platform you're about to use.

When AI has to guess, you end up with a mess.

Now here's the part I want you to sit with. Most of what AI sends your way probably can be created in Squarespace, especially once you start customizing with code. But being able to recreate a design doesn't mean you should recreate every detail exactly as AI drew it. Squarespace has a structure. Some of it exists for accessibility, some for responsive design, and some because there are design principles and technical considerations underneath a website that an AI mockup simply isn't accounting for. Those limits are good. Your client looks at what AI made and thinks that looks great, and they might be right. Your job is to ask whether it will work great. Very different question.

So before I reply to the email, before I quote anything, and definitely before I open Squarespace and start clicking around a blank template, I sort every element of that design into one of three buckets.

Built in is everything Squarespace already does. A hero section with a background image and text on top. An image block, a text block, a button, a gallery, an accordion, a basic form. A Fluid Engine section that looks unusual because one image is offset or the typography is bigger than you'd normally go. This is almost always your biggest category. Look at the mockup with content blocks in mind and ask what already exists that gets you most of the way there.

Built custom is possible, but it takes extra work. A hover effect Squarespace still doesn't handle natively, like changing a button's color on hover. A layout that's almost a carousel list section, except the arrows sit somewhere else. A custom background treatment, a border effect, text behaving in an unusual way. This is the fun category if you're comfortable with CSS, and it's also where you need discretion. Does rotating that image three degrees meaningfully improve the design? Does that corner treatment matter? Does that typography animation still look good on mobile, and if not, is it worth an hour building a separate mobile solution? Sometimes yes, those details are what make a design feel special. But sometimes AI made something unusual because it had no reason not to, and you're about to spend two hours reproducing a decorative detail nobody cares about, including the client.

Built somewhere else is the one that derails projects. This isn't I need to learn more CSS. It's functionality that requires an entirely separate system. A client sent me a mockup with a booking calendar: visit the site, pick a date, pick a time, get a confirmation. Visually that's a fancy form, and AI can draw it in seconds. But it's an appointment system. It has to know your availability, hold a time slot so two people don't book the same one, send confirmations, handle reminders and cancellations and reschedules. You can absolutely have that on a Squarespace site, but it means another tool, likely another ongoing subscription, testing to make sure it works, and styling to make sure it matches. To the person looking at the mockup, that calendar feels exactly like the hero section above it. Same finished look, wildly different invoice.

Never let your first response be "Squarespace can't do that." Even when it's technically accurate, it isn't helpful, and what your client hears is she doesn't know how or she doesn't want to. Reflect the three categories back to them instead. Here's what's already built into Squarespace and is straightforward. Here's what needs custom development. Here's the functionality that needs a separate tool or integration. Then explain what each one means for the timeline, the budget, and any subscription they'll be carrying, or why one small decorative detail isn't worth the development time. That conversation is where your experience becomes visible.

Because that's what they're paying for. Not pixel-matching. They're paying you to look at a picture and understand what it actually takes to turn it into a working website: to recognize the difference between something that looks simple and something that is simple.

I'm not down on AI-generated design. I use these tools nearly every day, and I don't think a single client is wrong for sending one. There's something genuinely exciting about someone who can't design or code finally getting the idea out of their head and onto a screen. That's a great starting point. It's just a starting point.

They have the idea. You create the plan.

Next
Next

[96] Before You Cancel Squarespace: What It Really Takes to Host a Site Yourself