Ask a bakery what it sells and you get a list with conditions: sourdough until it runs out, the cardamom buns on Saturdays, the quiche only when the oven is free. Ask a website builder the same question and it hands you a template with nine permanent products. Those two things do not fit, and the gap is where most bakery websites die — not from ugliness, from being wrong by mid-morning.
The fix is not a bigger system. It is deciding which parts of the page are permanent, which are daily, and which are paper — and then making the daily part take under a minute to change from a phone, at the counter, with flour on your hands.
What is actually permanent, and what is not
A bakery page has five jobs. Four of them barely change; one changes every day.
| Part of the page | How often it changes | Where it is edited |
|---|---|---|
| Hours, including the closed day | A few times a year, plus holidays | The page |
| Address and a Directions button | Never | Comes from the Maps link |
| Phone, for pre-orders and "is there any left?" | Never | The page |
| Standing list — what you bake in a normal week | Seasonally | The page |
| Today — what is in, what is out, what is new | Every morning | One line, from a phone |
Only the last row needs speed. Everything above it is why a page exists at all: the guest standing outside at 7am wants to know whether you are open, where the door is, and whether it is worth walking to. That part of the page can be built once and left alone for months.
The four fields to change each morning
Speed comes from having the daily edit be the same four actions every time, not from a clever app. In our editor the whole page is one form, and the daily pass is:
- Hide what is gone. Delete the line, or put "sold out" where the price goes. The list is one item and one price per line — no categories to open, no product pages to unpublish.
- Add what is new. One line, typed the way you would say it across the counter: Fig & walnut loaf — $9. Twelve lines is the useful ceiling for a page read on a phone; the printed menu takes the rest.
- Check the hours line. If today is the closed day, the page already says so, because the hours were typed once as a week. That one line — "open now · until 3pm" or "closed today · opens 7am tomorrow" — is the most-consulted line on any shop page.
- Save. The address does not change, so the QR code on the counter does not change, and nobody has to reprint anything.
Why the address must never change. A printed QR code encodes the URL, not the content. If the address stays the same, one print run of counter cards and bag stickers serves every menu change that follows — including the day you change supplier, or the day you stop paying for anything. This is also the argument against "dynamic QR" subscriptions: re-pointable codes exist for pages whose address moves, and that is a problem you can simply not have.
The paper half: a counter card, a board, a bag sticker
The printed pieces are what make the page useful in a bakery, because the guest is standing in front of a glass case, not searching. Three pieces do almost all the work, and each has a size that is set by physics rather than taste:
| Printed piece | Code width | Where it works | Scans from |
|---|---|---|---|
| 70 mm sticker on the till or the case | 40 mm | Someone leaning in, phone already out | ~40 cm |
| Counter card, A6 upright | 62 mm | The queue, one step back | ~60 cm |
| Menu board, A4 | 126 mm | Read from the door or across the room | ~1.2 m |
Those widths are measured in the PDFs we produce, not estimated: the A4 poster places a version-4 code (33 modules, 41 with the four-module quiet zone) at 126 mm on a 210 mm page, and the sticker sheet at 40 mm. The distances follow the industry's 1:10 rule of thumb — the code's width is about a tenth of the distance it scans from — which is a planning heuristic, not part of the QR standard. The full arithmetic, including why "fit to page" silently shrinks a code to 94%, is in the print sizes guide.
Two more pieces are worth the paper in a bakery specifically:
- A pre-order slip. Not a form — a line on the bag: "message us before 6pm for tomorrow's loaves". Bakeries take pre-orders on the phone because a person has to confirm what is actually in the oven; a page that promises a cart promises stock you may not have.
- A closing-time notice. The same text you would post on the door for "sold out of sourdough today, back tomorrow at 7" — short enough to send to the delivery apps and paste into a Business Profile post.
Google: the page is the fast copy, the profile is the slow one
Google's Business Profile has a menu editor for food and drink businesses, and its own Help page says changes can take 24–48 hours to appear on Maps and Search. The same page says Google may transcribe menu data from your business website, with an opt-out if you would rather it did not.
Read those two lines together and the division of labour is obvious:
- The page is where today's list lives, because it changes in a minute and it is the version you control.
- The profile is the slow, authoritative copy — the website field, the menu link, the categories, the photos. Point its menu link at the page and let Google read what it can.
- A Profile post is the same-day channel that does exist: "cardamom buns out at 11 today" is a post, not an edit, and it does not wait on a review queue.
Two rules from the same guidelines are worth repeating because bakeries break them constantly: the business description must not contain links, and the business name must match the sign above the door — so the page address goes in the website field, not in the description.
What this honestly does not do
A static page cannot count stock. It will not tell the eleventh customer that the tenth bought the last croissant; it can only be corrected after the fact. If a bakery needs live availability, pre-order payments, allergen filtering per batch, or a loyalty card, that is a different product — a till with inventory, or an ordering platform — and it comes with a monthly fee and a login for staff at 6am.
What it does do is answer the questions that decide whether someone walks over: what is out today, when you close, where the door is, and how to reserve a loaf by phone. Those answers are static text. Static text is the cheapest thing to serve on the internet, which is why it is strange that it is usually sold by the month.
The alternative, priced with dated numbers
Monthly list prices read from the vendors' own pages on 13–14 November 2026, before tax and before card fees:
| What you buy | Per month | 3 years | What you get for a bakery |
|---|---|---|---|
| Squarespace Basic ($19 billed annually, $25 monthly) | $19 | $684 | Real editor, templates, blog, shop — more than a pastry counter needs |
| Wix Light ($17 on the yearly plan only) | $17 | $612 | Same trade-off, one plan down |
| Linktree Pro ($14 annual, $17 monthly) | $14 | $504 | A bio link list, not a menu with hours and a map |
| MapToSite Card — the page only | $0 | $0 | The daily list, hours, map, call button; one credit line in the footer |
| Turnkey kit — page plus print | once | $49 | The page plus the A4 poster, counter cards, stickers, tent, print menu and edit runbook |
If you want the same arithmetic with your own numbers — your fee, your remaining term, your table (or shelf) count — the three-year calculator does it in the browser and stores nothing you type.
The morning routine, in full
- 6:00 — open the edit link on the phone, hide last night's sold-out lines, add today's special, save. One minute.
- 6:05 — post the same line as a Business Profile post if it is unusual ("fig tart today only").
- 11:00 — when the sourdough runs out, delete the line. The counter card still scans to the same address.
- Sunday — check next week's hours once; if nothing changed, change nothing.
The page and the paper, in one order
Send a Maps link, your bake list and your phone number. Within 24 hours: the live mobile page, the A4 QR poster, the print menu, the counter cards, the table tents, the 70 mm stickers and the rest of the fifteen pieces — $49 once, nothing monthly. A counter with no tables gets the eight-file version at the same price.
Try the page before you decide — thirty seconds, free, and it stays free:
Create a page — free What a café's page should cost
Questions we get asked
How do I show a sold-out item on a bakery menu page?
Delete the line, or write "sold out" where the price goes. The list is one item and one price per line, so hiding a bake is one line edited on a phone — the address, and every printed QR code, stay the same. A live stock count needs a till; a daily truthful list does not.
Should the daily menu be a picture of the chalkboard?
A photo is decoration, not a menu: Google cannot read it reliably, a phone user cannot copy a price out of it, and it cannot fill the profile's menu link. Type the list; photograph the counter separately.
Where does the QR code go?
A 70 mm sticker on the case (40 mm code, ~40 cm), a counter card at 62 mm for the queue, and a menu board at 100–126 mm for people reading from the door. Print at 100%, never "fit to page".
Does the page need online ordering?
No — and adding it changes the job. The page says what is in today, when you open, where you are and how to call. Pre-orders by phone let a person confirm what is actually in the oven.
What about the menu feature on Google Business Profile?
It exists for food and drink businesses, changes can take 24–48 hours, and Google may transcribe menu data from your website. That lag is why the page is the fast copy and a Profile post carries the same-day notice.
How much does it cost?
The page is $0 with no account and no card, and it stays free. Board is $6.99 a month for a six-photo gallery, Sign is $13.99 for a custom domain. The printed kit is a one-off $49 and the files are yours.
Related reading: the millimetres that decide whether a printed code scans, what a café's menu page should cost in 2026, the website and menu links on a Business Profile, and a menu page without a POS contract.
Sources, read on 14 November 2026: support.google.com/business/answer/9455840 — the menu editor is for food and drink businesses, changes to the menu can take 24–48 hours to appear, and Google may transcribe menu data from your business website (with an opt-out). support.google.com/business/answer/3038177 — the business description must not contain links and the business name must match real-world signage. QR measurements (126 mm on an A4 page, 62 mm table tent, 48 mm Wi-Fi card, 40 mm on a 70 mm sticker; version 4, 33 modules, 41 with the four-module margin) were measured in the PDFs this site produces for the showcase; the 0.17 mm module floor is Denso Wave's own FAQ figure, and the 1:10 scan-distance rule is an industry heuristic, not part of ISO/IEC 18004. Subscription prices are list prices from each vendor's own pricing page on 13–14 November 2026 (Squarespace Basic $19 annual / $25 monthly; Wix Light $17 yearly; Linktree Pro $14 annual / $17 monthly), before tax and excluding card processing. The morning routine is a description of the four fields in this product's editor. MapToSite's own prices are on this site.