The short version: Export before you cancel, not after, because access usually ends with the subscription. You need three separate files, not one: the customer list, the order history, and the product catalog. The field almost nobody exports is email consent status, and without it you cannot honestly tell who agreed to be marketed to, which means you either stop mailing people or start guessing. Do the export today rather than on the day you leave, because discovering that something does not export is a fact you want while you still have options.
Because a cancelled account is usually a closed door.
Most platforms end your access when the subscription ends. Some keep data for a grace period, some do not, and the ones that do rarely advertise how long. So the sequence that fails is: decide to leave, cancel, then try to export.
The sequence that works:
Step two is the one people skip. A downloaded file is not a verified file, and finding out that your customer export contains 40 rows when you have 400 customers is much better news on a Tuesday than on your last day of access.
They are separate exports on most platforms, and missing one is easy.
The customer list. Names, email addresses, phone numbers, and ideally the date they first ordered. This is the asset. Everything else can be rebuilt.
The order history. What each person bought, when, and for how much. This is what lets you segment later, and it is also part of your business records. Our guide to segmenting an email list by order history covers why that matters.
The product catalog. Names, descriptions, prices, options, and the image files. Re-typing forty products is an evening; re-photographing them is a weekend.
A fourth if you can get it: your reviews or testimonials, which do not move with you and are worth copying somewhere before they vanish.
Email consent status, and it is the one with real consequences.
Marketing email requires permission, and permission is a property of each individual contact rather than of your list as a whole. If your export gives you 400 addresses with no indication of who opted in, who unsubscribed, and who merely bought something once, then you have a list you cannot honestly use.
What to look for in the export:
If those columns are missing, ask support for them specifically before you leave. If they genuinely do not exist, the safe path is to treat the list as unconsented and re-permission it: send one message from the old system asking people to confirm, then start clean.
That is painful and it is far better than importing 400 addresses into a new tool and mailing people who unsubscribed from you last year. Our guide to building a customer email list covers doing it properly from the start.
Ask rather than assume, because this varies more than you would expect and it is rarely on the pricing page.
StandScout lists inventory export on its $9.99 Starter tier and data export and reports on Pro at $29.99. So the export capability is itself tiered, which is worth knowing before you cancel down to a cheaper plan and then try to leave.
Local Food Marketplace does not publish its export terms. For a hub with hundreds of members and years of order history, "what exports, and in what format" is one of the first questions to put on a demo call.
GrazeCart and CSAware likewise publish nothing on this. CSAware does emphasise that your data is downloadable at any time, which is a good sign and still worth confirming in writing.
The general storefronts, Shopify, Square, Squarespace, Wix, and Big Cartel, all offer CSV exports of customers, orders, and products as standard. That is the advantage of a large platform: leaving is a solved problem.
LocallyGrown.net and Big Cartel have a structural advantage: because you connect your own payment processor, your transaction history lives in your own Stripe or provider account and does not move when you do.
Some things do not come with you, and it is worth knowing which so you can capture them manually.
That last one is worth acting on regardless of whether you are switching. Owning your own domain costs about $15 a year and means your address survives any platform change. A vendor on a platform subdomain has been building traffic to an address they do not control.
The National Archives' records management guidance is aimed at institutions rather than bakeries, but its underlying principle applies: decide what is a record, keep it somewhere you control, and do not rely on a vendor to preserve it for you.
Five checks, and they take ten minutes.
Point four catches people constantly. An order export that quietly covers only recent activity looks complete until you go looking for a customer who last ordered in March.
Longer than you think, and for reasons beyond marketing.
Your order history is a business record. It supports your tax filings, it evidences what a customer bought if a dispute arises, and it is the only thing that can answer a question about a sale from two years ago. The IRS's recordkeeping guidance sets out what you are expected to retain and for how long, and the practical answer for most small food businesses is several years, not several months.
That matters here because a platform's retention policy is not your retention obligation. If a system keeps two years of orders and your obligation runs longer, the gap is yours to close, and the way to close it is a periodic export sitting in a folder you control.
Three things worth doing:
None of that is exciting and all of it takes about five minutes a quarter. It is also the difference between a platform problem being an inconvenience and being the end of your customer relationships. If you are trialling somewhere new, export from the trial before you commit rather than after, since a platform that makes leaving straightforward is telling you it expects to keep you on the merits.
Keep them somewhere that is not a platform.
That last point is the useful habit. An export is a snapshot, and a snapshot from the day you left is more useful than nothing but less useful than a recent one. Doing this once a quarter takes five minutes and means a platform going down, changing terms, or locking your account is an inconvenience rather than a disaster.
This applies whichever platform you are on, including one you have no intention of leaving. Our guide to building an email list for a farm stand and our piece on collecting a customer list with QR codes at a market both produce data that should live in a file you control, not only in whatever tool captured it.
It should be one of your questions, and it rarely is.
When evaluating anything, ask:
Question five is the one to get in writing. Question six is one you can fix yourself for $15 a year and should, today, regardless of anything else on this page.
A platform that answers all six clearly is telling you something about how it expects to keep you: by being worth staying on rather than by making leaving difficult.
Sequencing matters more than speed here, and getting it wrong is how people lose a fortnight of orders.
Step six is the one people discover too late. Every bag you printed, every post you made, and every text you sent contains the old link. If you own the domain, a redirect keeps all of it working. If the old address was a platform subdomain, those links break on the day you cancel and there is nothing you can do about it, which is the practical argument for buying your own domain long before you ever plan to move.
If you are evaluating where to move to, 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: it does not ship nationally, it is not a website builder, and there is no point-of-sale. On the question this article is about, apply the same standard you would apply to anyone: ask what exports before you commit, and test it during the trial rather than taking a claim on trust. You can import a handful of real customers and products during a trial and then immediately try exporting them again, which is a five-minute test that tells you more than any feature list.
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 | What the customer export gives you | Subscription (annual) | Free trial | Platform fee | Card processing |
|---|---|---|---|---|---|
| Shopify | Customer export includes marketing consent status | $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 | Customer export available, consent status not guaranteed | 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 |
| Etsy | Order data yes, marketing contact restricted by policy | No subscription | n/a | $0.20 listing + 6.5% commission | 3.0% + $0.25 processing |
| Big Cartel | Order and customer export available | Platinum $12/mo ($144/yr) | 7-day free trial | $0 platform fee | Your own provider, so 2.9% + $0.30 typical |
| StandScout | Data export sits on the $29.99 tier | Free to $59.99/mo (monthly only) | No card required on free tier | No checkout, so no platform fee | No checkout, so no processing |
| Homegrown | Export your customers and orders | $10/mo billed annually | 7-day free trial | $0 platform fee (0% commission) | 2.9% + $0.30 processing |
Before you cancel, and ideally once a quarter regardless. Most platforms end access with the subscription, so the sequence of cancel-then-export fails and there is often no way back.
Three: the customer list with contact details, the order history with what each person bought and when, and the product catalog including the actual image files rather than just filenames.
It records who agreed to be marketed to and who unsubscribed. Without it you cannot honestly tell who you may email, and importing an unmarked list risks contacting people who opted out.
Ask support for it specifically. If it genuinely does not exist, treat the list as unconsented: send one message from the old system asking people to confirm, then start clean on the new one.
Count rows against the dashboard total, open it in a spreadsheet and check the columns parsed correctly, spot-check five known customers, confirm the date range covers everything, and verify images actually downloaded.
Reviews, saved payment methods, search ranking on your old URL, and sometimes photographs. Copy your best reviews into a document manually, since they are usually tied to the platform.
Yes, and it costs about $15 a year. A domain you own means your address survives any platform change, while traffic built to a platform subdomain belongs to the platform rather than to you.
Export before you cancel. Access usually ends with the subscription, and a cancelled account is rarely a recoverable one.
You need three files: customers, orders, and products with their images. Then verify them, because a downloaded file is not a verified file and an order export that quietly covers only 90 days looks complete right up until you need March.
The field that matters most is email consent status. Without it you have addresses rather than a mailing list, and the honest fix is to re-permission from the old platform while you still can. And whatever you decide about platforms, buy your own domain today. It is $15 a year and it is the difference between an audience you own and traffic you are renting.
