The short version: If some of your customers phone or text and the rest order online, the capability that decides everything is whether you can enter an order yourself, on their behalf, into the same system. Shopify calls this a draft order. Square lets you create one from the dashboard or take it on its point-of-sale. Bakesy is built around exactly this, since custom-order enquiries arrive as conversations rather than as checkouts. Platforms without it force you to keep a parallel list, which is the whole problem you were trying to solve. Ask for it by name: "can I create an order on behalf of a customer?"
All figures came from each company's own pricing pages in July 2026.
Because a split system is worse than either system alone.
If online orders live in a storefront and phone orders live in a notebook, then on Saturday morning you have two lists. Your stock count is wrong on both, because neither knows about the other. Your pick list is incomplete. And the thing you bought software to stop doing, reconciling by hand, you are still doing, only now with a subscription attached.
One system with manual entry means:
The capability sounds minor and it is the difference between a platform that replaces your process and one that sits alongside it.
Based on what each publishes:
Shopify has draft orders: you build the order in the admin, and can send the customer an invoice link or mark it paid. This is a long-standing, well-documented feature.
Square lets you create orders from the dashboard and take payment in person through its point-of-sale app with a reader, which for a market vendor is the same problem solved from the other end.
Bakesy is built around it. It is a custom-order management tool rather than a storefront, so the entire flow assumes a conversation that becomes a quote that becomes an order. Our Bakesy pricing breakdown covers what that costs, including the fact that its card processing is optional.
Cheddar Up takes payments through forms and supports a point-of-sale with tap-to-pay on a phone, so an in-person or phoned order can be captured in the same place as an online one.
Big Cartel includes selling in person and taking payments from your phone on its $144-a-year Platinum plan, which covers the counter and market cases.
For anything not on that list, ask directly rather than assuming. It is a specific, answerable question, support will know the answer immediately, and the reply tells you whether the platform was designed for a business that talks to its customers.
Four, in descending order of how well they work.
Order on the customer's behalf through the normal checkout. You place it as if you were them, using their name and a note. This works, it takes two or three minutes, and payment is the awkward part: you either take card details over the phone, which you should not do casually, or send them a link and chase it.
Send a payment link and add the order manually. Two systems again, and the order only exists once they pay.
Keep a separate list for phone orders. Honest, and it reintroduces the reconciliation you were trying to remove.
Refuse phone orders entirely. Cleanest, and it costs you customers. Some people will not order online, and a few genuinely cannot.
The middle two are where most vendors end up, and they are the reason people conclude that their platform "does not really work" when the real gap is one missing feature.
Worth knowing, because it changes how much this matters and it is usually not who people assume.
That fourth group deserves particular thought. An online-only ordering process excludes some people, and while a small food business is not the subject of most formal accessibility requirements, the Department of Justice's ADA guidance is worth being aware of and the underlying point stands regardless of what applies to you: a phone number is the simplest accessibility feature there is.
The practical read is that phone orders are typically 10% to 20% of a local food business's volume and skew towards larger, more profitable orders. That is not a segment to design out.
Generally no, and this is worth being firm about.
Writing a card number on a notepad while someone reads it out creates a record you then have to destroy, and handling card data by hand carries obligations most small vendors have never looked at. It is also the kind of thing that is completely fine right up until it is not.
Better options, in order:
Option one is the standard answer and the reason draft orders matter: the order is captured, your stock is right, and the payment chases itself.
The obligations here are real even at small scale. Any business that accepts card payments is responsible for how it handles that data, and the FTC's privacy and data security guidance for businesses sets out what that involves. The short version for a home food business: do not create a record of a card number in the first place, and every option above is designed so you never have to.
Worth putting a number on, because the feature gap sounds small and the consequence is not.
Take a vendor with 40 orders a week, 15% of them by phone, so six phoned orders and thirty-four online.
Reconciliation time. Merging two lists, checking for duplicates, and rebuilding a combined pick list takes twenty to thirty minutes a week. That is roughly 20 hours a year.
Oversells. With two stock counts, selling the same last four croissants twice is a matter of when rather than whether. Call it once a month: a refund, an apology, and a customer who is slightly less likely to come back.
Missed orders. A phoned order that never made it onto the list is the worst version, because the customer arrives and there is nothing there.
None of that is catastrophic and all of it is avoidable by one feature. Which is why it is worth asking about before you compare anything else: create a manual order in a trial and the answer takes three minutes to establish.
Increasingly the larger share, and the same principle applies with one addition.
Text orders arrive in fragments. "Two sourdough for Saturday" then, an hour later, "actually make it three, and do you have brownies?" Transcribing that accurately is where errors come from, and it is worse than a phone call because the record looks reliable.
What helps:
Our guides to taking pre-orders and handling pre-orders alongside in-person sales cover the process, and our piece on getting market regulars ordering online between markets covers moving some of that volume across without losing anyone.
You do not need to eliminate phone orders, only to stop them being the default.
Point six is the one people resist. A business with 15% phone orders is not failing at digital adoption; it is serving the customers who wanted to talk to someone, and those customers frequently spend the most. Our guide to getting repeat customers covers why that relationship is worth protecting.
Six things, and the first is the whole question.
Step two is the one that quietly fails. Some platforms let you record an order without it affecting inventory, which means your stock is wrong in exactly the situation where you most needed it to be right.
Step five is worth doing properly too. A platform that records manual orders but excludes them from sales reports will quietly understate your best month, which matters when you are deciding whether a market is worth keeping or what to bake more of. Place one manual order and one customer order in the same day, then open the report and check both appear. If the totals do not match what you know you sold, you have found a problem worth knowing about before you build a year of records on it.
If your setup is pickup at several places with orders arriving both ways, Homegrown is $10 a month billed annually with 0% commission and 2.9% plus $0.30 processing published up front, and it handles pickup at each place you sell with its own schedule and cutoff, local delivery with a radius and a route, and sales tax calculated, filed, and remitted in all 50 states. The honest bounds: there is no point-of-sale for taking card payments at a stall, no national shipping, and no drop windows. On manual order entry specifically, do not take my word for it or anyone's: create an order on behalf of a customer during the trial and see for yourself, because that is the one test that decides whether a platform replaces your process or sits beside it. There is a 7-day free trial, and that test takes three minutes.
Every platform below shows the same four commercial facts, because a table that lists one platform's transaction fee and not another's is not a comparison. "Not published" means exactly that: the company does not state it publicly.
| Platform | Entering an order yourself | Subscription (annual) | Free trial | Platform fee | Card processing |
|---|---|---|---|---|---|
| Homegrown | Manual order entry that reduces stock | $10/mo billed annually | 7-day free trial | $0 platform fee (0% commission) | 2.9% + $0.30 processing |
| Shopify | Draft orders from the admin | $29/mo Basic | 3-day trial, then $1/mo for 3 | 2% platform fee if not on Shopify Payments | from 2.9% + $0.30 processing |
| Square Online | Build an order from the dashboard | Free tier; paid from $29/mo per location | 30-day trial on paid plans | $0 platform fee | 3.3% + $0.30 free tier, 2.9% + $0.30 paid |
| Bakesy | Built around orders you enter for a customer | $9.99/mo Standard (monthly only) | 30-day free trial | $0 platform fee | 3.9% + $0.30 processing, optional |
| Etsy | No, orders must come through the marketplace | No subscription | n/a | $0.20 listing + 6.5% commission | 3.0% + $0.25 processing |
| Big Cartel | No native manual order entry | Platinum $12/mo ($144/yr) | 7-day free trial | $0 platform fee | Your own provider, so 2.9% + $0.30 typical |
Shopify has draft orders, Square lets you create orders from the dashboard and take them on its point-of-sale, Bakesy is built around custom-order entry, and Big Cartel includes in-person selling from your phone on Platinum.
You should not write card details down. Take the order, then send a payment link, or take payment at collection with a reader. That keeps the order in your system without you handling card data by hand.
For most local food businesses, roughly 10% to 20%, and they skew towards larger and more complicated orders. That is not a segment worth designing out of your process.
Because without it you keep two lists. Your stock count is wrong on both, your pick list is incomplete, and you are still reconciling by hand, which is the work the software was supposed to remove.
Make the online route genuinely easier, mention it when they call, and text them the link afterwards so it is one tap next time. Do not remove the phone option; that converts customers into ex-customers.
Yes, and they are slightly worse, because a written thread looks like a reliable record while arriving in fragments. Confirm the whole order back as one message and enter it immediately.
Create an order on behalf of a customer, and then check three things: that it reduces stock, that it appears on the pick list with online orders, and that reports treat both as one set of numbers.
If orders reach you two ways, the feature that matters is not the storefront design or the fee structure. It is whether you can put an order into the system yourself, so that one stock count, one pick list, and one set of numbers cover everything.
Shopify, Square, Bakesy, Cheddar Up, and Big Cartel all address this in some form. For anything else, ask by name: *can I create an order on behalf of a customer, and does it affect inventory?*
And do not try to eliminate phone orders. They are typically 10% to 20% of volume, they skew large, and the customers placing them are often the ones who have been buying from you longest. The goal is one system that captures both, not one channel that excludes some of your best customers.
