Why we built Morrowyn: gifts deserve better than a search box

July 1, 2026 · 5 min read

Every AI gift tool works the same way. You open it in a mild panic, answer a quiz about someone you love, and it hands you a list. Next occasion, you start from zero: same quiz, same generic candles, same quiet feeling that the tool doesn’t actually know anything about the person you’re buying for.

We think that’s backwards. So we built the other thing.

The search box was never the problem

Ask anyone why their last gift was disappointing and they’ll rarely say “I couldn’t find anything”. The internet has no shortage of things to buy. What went wrong happened earlier, in one of three quieter places.

You didn’t remember the date in time. Not “forgot it entirely”, usually. You remembered on the Thursday for a Saturday, which is the same thing as forgetting, just with more guilt attached.

You didn’t remember the person. Not who they are, obviously. What they’re into right now. The class she mentioned starting in March. The fact that his headphones died. What you already gave them last year, which you will absolutely re-gift if nobody stops you.

You didn’t have an idea that fit. Not an idea in general: an idea for them, at the amount you actually wanted to spend. “Gifts for her 2026” is not that. It has never once been that.

Notice that none of the three is a search problem. Two of them are memory problems and one is a judgement problem, and a search box is structurally incapable of helping with any of them. It doesn’t know your calendar. It has never met your sister. It has no idea what you spent last time.

Why gift finders can’t fix it

This isn’t a criticism of anyone’s engineering. It’s the shape of the product.

A gift finder is stateless on purpose. It has to be: it wants to serve a stranger who arrived from a search result, convert them into a click, and take a cut. Everything about that design is optimised for the panicking visitor, which means it can only ever meet you at your worst moment. The tool is at its most useful precisely when you have the least time to use it well.

And because it forgets you the moment you close the tab, it can never get better at you. Ten visits over three years and it knows exactly as much about your father as it did the first time, which is nothing. The tenth list is no smarter than the first.

What we built instead

Morrowyn starts from the opposite end. You add your people once: a name, the relationship, their dates, a few things they love. It takes a few minutes, and it’s the only real work the app ever asks of you.

Then it keeps everything. The hints they drop, the sizes, the thing they said in passing that you jotted down in April. What you gave, and how it landed. Which ideas you liked and which made you wince at the price.

And it watches the calendar so you don’t have to. Weeks before each occasion, not the night before, it comes to you: “Sam turns 30 on the 14th. Here’s what caught our eye.” Five to seven ideas, each with the reason it fits, tied to something you actually told it. Not “a scented candle”. Something more like: “she mentioned getting into pottery in March, and this is a beginner’s wheel in the range you usually spend.”

The reason it can do that is not that our AI is cleverer than anyone else’s. It’s that ours is the only one that has been listening all year.

The part that compounds

Here’s the bit we care about most, and the reason this is a memory and not a finder.

Year one, Morrowyn is a good reminder with decent ideas. Fine, useful, not magic. Year two is different, because by then it knows that your brother hated the “experience day” and loved the obscure vinyl, that your budget for friends is real and your budget for your mother isn’t, that you already gave your father a whisky set and must never, ever suggest another one.

Year three it is quietly better than you are. Not because it’s smart, but because it remembers everything and you remember about sixty percent, which is the honest number for all of us.

Most software gets worse as it fills up with your stuff. This is one of the rare cases where the accumulation is the product.

What we deliberately didn’t build

A few things we’ve said no to, so you know what you’re getting:

• No shopping agent. Morrowyn suggests and explains; it doesn’t buy things on your behalf. Handing your card to an AI for a gift that matters is not a trade we’d make ourselves.
• No social layer. Your people are not a network to be grown. Nobody gets notified that you added them.
• No selling what it knows. What Morrowyn learns about the people you love stays yours: private by default, exportable and deletable any time, never sold, and never used to advertise to them. The whole product runs on your trust, so we don’t get to be careless with it.

The whole idea, in one line

We’re not trying to build a better search box for gifts. We’re trying to build the thing the genuinely thoughtful people in your life already have, which is not better taste and not more money. It’s a system: they wrote it down, and they started early.

Morrowyn is that system, and it remembers so you don’t have to.

We’re building it now, for anyone who’d like to be the person who always seems to know, without the scramble.

Read next

The difference between a thoughtful gift and an expensive one

It's tempting to treat price as a stand-in for effort. They're not the same thing, and mixing them up is the most common way a gift misses.

August 29, 2026 · 2 min read

Why "what do you want for your birthday?" is the wrong question

It feels like the safe, considerate question to ask. It's actually the one that guarantees a forgettable gift. Here's what to ask instead.

July 29, 2026 · 2 min read

Why the best gift-givers start three weeks early

The difference between a thoughtful gift and a panic buy usually isn't taste or budget. It's lead time. Here's what three weeks actually buys you.

July 4, 2026 · 3 min read

More in Thoughtful Giving›

‹ Back to all posts