Esquerre Design

This is the first of two bathrooms on our main living floor, and it might be the smallest room in the entire house. I can stand in the middle of it and touch both walls at the same time — one hand on each side, no stretching required. It’s less a bathroom and more a closet that got a bathtub and a toilet instead of shelving.

It also doesn’t have a window, which is a big part of why it feels so closed in. There’s no natural light source at all, just an overhead fixture, so even during the day it reads more like an interior storage room than a place anyone spends real time. One idea I keep coming back to is a skylight — it wouldn’t solve the square footage problem, but it would completely change how the room feels, and it’s one of the only ways to get real light into a room with no exterior wall to put a window in.

The bones of the room aren’t doing it any favors either. The tile is original to the 1960s — a pale pink that the previous owners painted white at some point, and that paint is already starting to peel in spots, which tells you everything you need to know about painting over tile as a long-term fix. At some point someone also laid plank flooring directly over whatever was underneath rather than dealing with it, so the floor has that slightly-off, not-quite-level feeling that comes with covering a problem instead of solving it. And the vanity, which is a fairly standard size, still manages to eat up a huge chunk of the available floor space in a room this small.

None of that sounds like the setup for a favorite room, and yet this ended up being mine, because it’s the only bathroom in the house with an actual bathtub. I love to soak, and this is the only option I have. The problem is that the tub is original to the house and quite shallow, which works fine for bathing my son but not so much for me — the second I try to actually submerge and relax, the water starts pouring down the overflow drain before I’m anywhere close to covered. It’s a bath in theory more than in practice.

Zoom out a little and the bathroom math in this house gets more interesting. The primary bedroom on this floor is genuinely huge, but there’s no en suite attached to it at all — you have to leave the bedroom to get to any bathroom, this tiny one included. Upstairs, there are two more bathrooms, and both of them only have a single sink. The back one is my husband’s — it has a window, actual elbow room (I can do a full spin with my arms mostly extended and not hit anything), but it still has the original 1960s shower, a narrow, submarine-hatch-style stall that’s clearly next on the list. The previous owners made a lot of real progress in the five years they owned this house, but the bathrooms weren’t part of it — which, along with the kitchen, makes them the biggest open opportunity left in the house.

So here’s the idea I keep circling for this particular bathroom: push the wall it shares with our bedroom out just a little. There’s enough room on the bedroom side to let the door move back and still leave the room feeling normal, which would free up just enough extra square footage to actually reconfigure this bathroom instead of just patching it. Combine that with a skylight where there’s currently no window at all, and a tiny, closet-like bathroom could genuinely start to feel like a real room.

The complication, of course, is that nothing in an old house is ever just cosmetic. We have hot-water baseboard heat running through this section of the house, and if I actually expand the bathroom into the bedroom wall, there’s a heating pipe running through that area that would need to be rerouted around wherever a new door ends up — either under the floor or some other path. From what I understand, it’s the last leg of that loop before it circulates back to the boiler, so it’s probably not as complicated as it sounds, but it’s still uncharted territory for us.

I’ll be honest, I’m still pretty new to all of this — figuring out what’s actually possible in a room this small, versus what just sounds good in my head, is its own process. But between this bathroom and the kitchen, it’s clear where the next real chapter of this renovation needs to go, even if I don’t have the full plan yet.

One thing that’s been genuinely useful throughout this process is using AI to actually see an idea before committing to it, physically or financially. I ran this same bathroom through an AI redesign, keeping the tile the same color as the walls instead of treating it as a separate element, and even though it’s the same footprint, it barely reads as the same room anymore.

The skylight is doing a lot of the work here. It solves the no-window problem without giving up any privacy, which matters a lot in a bathroom, and there’s something appealing about being able to lie in the tub and watch or listen to rain instead of staring at a blank ceiling. It’s exactly the kind of light source I was picturing when I first started thinking about this room.

If I’m being honest about the render, it’s missing a warm wood tone — everything reads a little cool and showroom-perfect right now, which is a good reminder that a render is a starting point, not a finished room. What it’s actually useful for is showing how much you can change a space just by rethinking the placement and finish of what’s already there, without inventing something completely new. The personal effects — towels, art, the actual stuff of living somewhere — are what will eventually turn an AI image into a room that feels like ours.

That first version leans into the fantasy a little — it’s basically what this room could become if we pushed the wall and started over. But I also generated a second version that’s a lot more useful right now, because it doesn’t require touching a single wall. Same footprint, same layout, just a new vanity, an updated tile pattern, and that same skylight dropped in. It’s a good example of what’s actually achievable today, without opening up the bedroom wall or rerouting the heat loop at all.

Between the two, I finally have a real range to work from: a smaller, realistic first phase I could actually tackle now, and a bigger vision to grow into later if the wall move ever happens. That’s the real value of doing this in AI first — I’m not guessing anymore, I’m choosing.