How to organize saved restaurants
One long list of places you want to eat is the same as no list at all. Paste yours in below, answer two questions about each one, where it is and what it is for, and watch it turn into three stacks you would actually open on a Friday.
You have a list of places you want to eat, and you are going to the usual one.
Not because the list is wrong. Because reading it takes longer than deciding not to.
A want-to-go list works beautifully while it is short and stops working once it is long. There is no exact number where that happens, and anyone who gives you one is guessing. The reason is simple: one long list makes you compare everything with everything, at the exact moment you are hungry and standing on a pavement. Nobody does that, so everybody goes to the usual place. The fix is not a better app and definitely not a longer list. It is splitting one list into three, using two answers about each place: roughly where it is, and what it is for. Below there is a sorter that does the splitting with you, one place at a time, in your browser. Then keep the three lists anywhere: notes app, map app, back of an envelope.
The sorter
Runs in your browser · nothing is sent anywhereTwelve is a good number to start with. If your list is enormous, take the top of it rather than all of it, because sorting eighty places in one sitting is how people give up on this.
HOW FAR AWAY IS IT
WHAT IS IT FOR
Answer both and it moves on by itself. Not sure about one? Pick "no idea", that answer is genuinely useful here.
The three stacks as plain text, ready to paste wherever you keep things.
Why the long list stops working
A want-to-go list is fine while it is short. You can hold a handful of places in your head and read them in four seconds. Then it grows. Somebody mentions a place, you add it. You screenshot something, you add it. Once the list has hundreds of names it no longer works like a shortlist, and nobody announced the change.
What it turned into is a list you can only read from the top. To use it you must check each place, work out where it is, and decide whether it suits tonight, from a name alone. That is real work, and you are doing it while hungry, with somebody waiting for an answer. So you go to the usual place, and the list gets a little longer next week.
Notice that the problem is not the number of places. It is that the list gives you no way to rule anything out. A hundred places sorted into near me and not near me is usable. A long list of names in one column is something you store, not something that helps you choose.
The four things that actually pick a restaurant
Watch anybody choose where to eat and the same four questions come up, roughly in this order.
The one that kills most options instantly. Not an address, just a rough answer: near us, other side of town, only when we are over there anyway.
A quick weeknight, a proper evening out, somewhere the children will survive, somewhere that takes eight people. A place that is perfect for one of those is usually wrong for the others.
Roughly. Everyday, treat, or special occasion. Three bands is enough, and it stops the awkward moment of opening a menu you were not ready for.
The one nobody writes down and everybody minds. Whether you can hear each other decides whether a place works for a birthday, a first date or your friend who hates noise.
Here is the hierarchy, because these numbers get muddled easily. Two answers sort the list, and they are the first two above: where it is, and what it is for. You can answer both about a name in under a second, and between them they rule out most of the list. Everything else lives in a one-line note next to the place, including cost and noise, and you only need it once a place has already survived the sorting. Two answers to file it, one line to use it.
Three lists instead of one
This is the whole system, and it is deliberately boring.
1. Go this month
Places near you, where you know what they are for. This list stays short on purpose, five or six names, and it is the only one you open when you are actually deciding. If it grows past ten it stops being a shortlist and you are back where you started.
2. When I am in that area
Good places in the wrong place. Nothing is wrong with them, they are simply not on your route. This list is not for choosing from, it is for checking when you already know you will be somewhere: near the station, in the other city, out by the park.
3. Needs a reason, or goes
You cannot remember why it is there, who mentioned it or what it is for. Before deleting, give it fifteen seconds: check the screenshot, the message or wherever the name came from. Quite often the reason is right there and the entry was simply written badly, which is a note problem rather than a bad restaurant. If fifteen seconds turns up nothing, delete it. A name with no reason attached was never information, and if the place is any good you will hear about it again.
Three lists work where one does not for a plain mechanical reason. Every question you can answer before you look, you no longer answer while you look. Deciding where to eat is now choosing from six nearby places rather than filtering two hundred names by hand.
Twelve places, sorted, start to finish
Faster to show than to explain. Here is a list somebody pasted in, the two answers they gave each place, and what came out.
| Place | Where | What for | Lands in |
|---|---|---|---|
| Trattoria Verde | Near us | Weeknight | Go this month |
| The Bread Shop | Near us | Quick lunch | Go this month |
| Kostas | Near us | With the kids | Go this month |
| Hana | Near us | Proper evening out | Go this month |
| The Anchor | Near us | Big group | Go this month |
| Sable | Other side of town | Proper evening out | When I am in that area |
| Cafe Nord | Other side of town | Quick lunch | When I am in that area |
| Pho 88 | Near the office | Quick lunch | When I am in that area |
| The Old Mill | An hour away | Proper evening out | When I am in that area |
| Bar Luz | Other side of town | No idea | Needs a reason |
| Nonna’s | No idea | No idea | Needs a reason |
| The Yard | Near us | No idea | Needs a reason |
What happened next, which is the interesting part. Five in the shortlist, four filed by area, three needing a reason. Of those last three, fifteen seconds in the messages app rescued two: Bar Luz turned out to be “Anna, the sherry place, go early” and The Yard was a colleague’s birthday suggestion. Nonna’s produced nothing at all and was deleted without ceremony.
Two extra rules the example throws up. A place you have already been to and want to return to belongs in the shortlist with the words “again” and what you had last time, because returning is a different decision from trying. And a place that needs booking two weeks ahead does not belong in the shortlist at all, no matter how near it is, since the shortlist is for tonight. Give it a trigger instead, like a birthday. Same for anywhere that has closed: delete it the moment you find out, or it will keep getting considered.
What to write down with the name
A name on its own has a shelf life of about two weeks. After that you are looking at a word. One line fixes it, holding up to four short things, written while the recommendation is still fresh. Add the important details early, so you are not researching every place at the moment you choose.
- Who told you about it. The single most useful word, because in six months you will be searching for "the place Anna went on about", never for the actual name.
- What to order. People do not recommend restaurants, they recommend a dish and then name the building it lives in. Losing the dish loses the recommendation.
- When it is bearable. Quiet at lunch, impossible on Saturdays, needs booking two weeks out. This is the fact that decides whether tonight is going to work.
- Roughly what it costs. One word: everyday, treat, occasion. Enough to stop you suggesting the wrong thing on a Tuesday.
Written out, a finished entry is one line and looks like this:
Trattoria Verde · near the park · weeknight, kids fine Anna, get the lemon pasta · everyday prices · quiet before 7
None of that needs a form or an app. It fits in the notes field of a pin, or on one line under the name in whatever list you already keep.
When somebody recommends a place in a chat, screenshot the message rather than typing the name out. The screenshot keeps who said it and what they said about it, which is most of the value, and it takes one second instead of thirty.
Clearing it out, once a quarter, in fifteen minutes
Lists rot. Places close, tastes change, and the thing you saved during a holiday in April has nothing to do with your ordinary Tuesdays. Four times a year, set a timer and go through it with three rules.
- No reason, no place. If you cannot say in four words why it is on the list, check where the name came from for fifteen seconds. Still nothing? It goes.
- Anything you have said no to three times, goes. You have considered it and not chosen it repeatedly. That is an answer. Let it be one.
- Shortlist back down to six. If "go this month" has grown to fifteen, move the bottom nine into "when I am in that area". The shortlist only works while it is short.
Fifteen minutes, four times a year. That is the entire maintenance cost of this system, and it is roughly the same amount of time you spend not deciding on one bad Friday.
Other people's recommendations
Most saved restaurants arrive as somebody else's enthusiasm, and they need handling differently from places you found yourself.
"Write the person down, not just the place"
Recommendations are only as good as the match between their taste and yours. After a while you learn whose suggestions land, and you cannot learn that if the names arrive stripped of who said them.
"Ask the one useful follow-up question"
Not "is it good", they already told you it is good. Ask what to order, or when to go. It turns a name into something you can act on, and people love being asked.
"A recommendation is not an obligation"
You are allowed to delete a place your friend loved. Keeping it out of politeness is how the list fills with things you have no intention of doing, and a list you do not trust is one you stop opening.
Why we are not telling you to install a restaurant app
Search this topic and you will find apps built specifically for saving restaurants: Beli, Eat List, Mapstr, and a good number of Notion templates such as this one. Several are genuinely nice, and if you eat out constantly one of them may be worth it.
They all share one requirement though. Your restaurants now live somewhere your other saved things do not, which means a fourth place to check and a fourth place to keep up to date. The system on this page needs none of that. It is three lists, two answers to sort them and one line of note per place, and it works in a notes app, in your map app, or on paper.
Where we come into it
Restaurant lists have the same problem as everything else you save: you cannot find the right one when you need it. dEssence is for the general version. You save the screenshot of the message, the link, the note about what to order, and later you ask Tracy the way you would say it out loud, along the lines of what was that place near the park Anna mentioned. Being straight about it: it is not a restaurant app, it does not know opening hours, it holds no reviews and it will not book a table. The three-stack system above is worth doing whether or not you ever use us, which is why the sorter asks you to sign up for nothing.
Sharing the list with somebody else
Should the list be shared with my partner?
The shortlist, yes. That is the one you both read on a Friday, and two people adding to one shortlist is how it stays current. Keep the other two lists to yourself if you like, because nobody needs to browse your unsorted pile.
What about a group of friends?
One shared list per group works better than one enormous shared list. The occasions are different, and a place that suits four colleagues at lunch rarely suits the same four at a birthday. Whoever suggests a place writes the note, since only they know what to order.
We disagree about what belongs in the shortlist
Then the shortlist is doing its job. Six places both of you accepted beat fifteen where one of you is quietly vetoing four every time. Argue at the sorting stage, once a quarter, not on the pavement at seven o’clock.
One long list is a museum. Three short ones are a plan for Friday.
Two answers to sort it, one line of note to use it, fifteen minutes four times a year to keep it.
How the sorter decides, and what it does not do
Your answers do all the work. The page has no opinion about any restaurant. It has never heard of them, it does not look anything up, and it cannot tell a great place from a closed one. It sorts the names you typed using the two answers you gave.
The rules, in full. Near you, with a purpose you named: go this month. Anywhere else, with a purpose: when you are in that area. No idea what it is for: into the third stack, wherever it is, because a place with no occasion attached never gets chosen. That third stack is a check-then-decide pile rather than a bin. Places you skip stay unsorted and are listed separately at the end.
You get the last word. Every place in the result has arrows to move it between the stacks. The sorter is there to get through the list quickly, not to overrule you.
Nothing leaves your browser. No account, no upload, no storage, and closing the tab throws it all away. That matters here because a list of where you eat is a fairly good map of your week.
Other tools in this area, if you would rather have a dedicated app: Beli, Eat List, Mapstr, and a Notion-based system.