Selling at markets and online at the same time

It's 11:40 on a Sunday, someone's holding the last speckled mug, and your phone has just buzzed with an order for the last speckled mug. Two people want it. One of them is standing in front of you with a £20 note and a fiver.

Whatever you do next, somebody's day gets slightly worse.

Selling at markets and online at once is not a marketing problem. It's an inventory problem wearing a marketing hat, and it has exactly one honest solution: decide before you leave the house which units are sellable online and which are coming with you, and then physically separate them. Not mentally. Physically, into different boxes. Everything else in this article is either a way of doing that, or a way of cleaning up when you didn't.

The number that causes all of it

You have one of something. You have two shopfronts. Nothing reconciles them in real time, because your stall has no internet connection to your website and your website has no idea you just took £22 in cash.

That's it. That's the whole problem, and it gets worse in direct proportion to how many one-of-a-kind items you make. A candle maker with 40 identical jars in a scent can absorb this all day. A potter with fourteen unique glazes cannot.

Which means the first question isn't "how do I sync my stock", it's "how unique is my stock". Sellers with deep, repeatable inventory can more or less ignore this article. Sellers with one-offs need a rule before Saturday.

If the online half of this is still new and you're mostly a market trader with an Instagram account, start with the architecture instead. Discovery happens in one place, the decision happens on a page with prices on it, and payment happens on whatever rail people already use. That's laid out in selling on social media without a website, and it's worth reading before you build a stock system around a shop you haven't set up yet.

Three ways to split the stock, and when each one is right

There are only three, and picking the wrong one is what causes the mug problem.

Freeze the online shop while you're at the market. Turn the products off on Saturday night, turn them back on Sunday evening with whatever's left. Simple, foolproof, and it costs you every online order that would have happened during the weekend. Best for sellers whose online sales are small and whose stall takings are most of the month.

Reserve a box. Physically separate stock into "market" and "online" before you pack the car. The online box stays at home. The market box comes with you, and if it sells out at 1pm you sell out at 1pm. Best for one-of-a-kind makers, and by a distance the most reliable option.

Sell online only what you have depth in. List the four things you can make twenty of, and keep the one-offs exclusively for the stall. Best if you've got a mix, and it has a pleasant side effect: the stall becomes the only place to get the interesting pieces, which is a real reason to come and find you.

Your stock	Pick this	What it costs you
Deep, repeatable, 20+ of each	Nothing. Sell both, top up Monday	A rare oversell you can fix by making another
Mixed, some depth, some one-offs	Sell online only the deep lines	The one-offs never reach your online audience
Nearly all unique	Reserve a physical box	Half your stock is idle on any given day
Unique and you need the market takings	Freeze online for the weekend	Every online order Saturday and Sunday

The one nobody chooses and everybody should is the physical box. It feels crude. It works because it's the only method that doesn't rely on you remembering something while three people are queuing.

If your products have sizes and colours, this gets harder fast, and the variant side of it is worth reading separately in how to handle sizes and variants.

The stall QR code is not a checkout

Sellers put a QR code on the stall expecting people to scan it and buy. Almost nobody does that, and it's not because the code is bad. It's because the person is standing in front of the actual object. They're not going to look at a photograph of it on a phone.

The QR at a market is for the people who don't buy.

That's the reframe that makes it work. Roughly, the people who walk past your stall fall into three groups: the ones who buy, the ones who were never going to, and the ones who liked it, picked it up, put it down and said "I'll have a think". That third group is who the code is for. They're the ones who walk away with a full mental note and an empty phone.

So the sign next to the code shouldn't say "shop online". It should say what that person needs:

• "Not sure? Scan and it's saved" for the browsers.
• "Sold out? Scan for the next batch" once you've cleared a line, which is the single highest-scanning moment of the day.
• "Scan for the ones that didn't fit in the van" if you genuinely have more at home.

The sold-out sign is the good one. A hand-written card saying SOLD OUT with a code under it, sat on an empty patch of table, gets scanned more than anything else on the stall, because the person has just been told they can't have the thing, which is exactly when they want it most.

Size, placement and why most printed codes never get used at all is its own subject, covered in QR codes that actually get scanned.

Put the code where the disappointment is. An empty space on the table with a price and a code beats a laminated sign on the corner every single time.

What actually happens to the cash tin

The reconciliation nobody warns you about isn't stock, it's takings.

At the end of a market day you have cash, card reader payments that settle in a day or two, a couple of bank transfers from people who had no cash, and, separately, whatever came in online while you were serving. Four streams, four different arrival times, and one of them, the card reader, deposits a net figure after its own fee, so the number in your bank doesn't match the number on your till roll.

Count the cash before you leave the pitch. Not at home. On the table, into an envelope, with the date and the market written on it. You will not remember on Wednesday whether the £340 was Chatsworth Road or the school fair, and the two have different pitch costs, so your profit per market becomes a guess.

The card reader is a separate thing to your online payments and should stay that way. A SumUp or Zettle reader is for a person standing in front of you. Sailo's card checkout is for a person on their phone somewhere else, and it needs a Stripe account cleared for charges, with 1–3% of the goods taken on each card order. Running your stall through an online checkout would be slower and more expensive than the reader you already have in your apron. Don't.

If the tax side of running both is nagging at you, that's a real question with a real answer and it's in what to do about tax on online sales. The wider UK picture, including registration and what buyers here expect, is in selling online in the UK.

Fran's Sunday in Hackney

Fran makes ceramics in a shared studio in east London. Mugs at £22, small bowls at £18, and a handful of larger pieces between £45 and £70. She does a market pitch most Sundays that costs her £35 for the day, and she posts online orders on Tuesdays and Fridays.

A good Sunday is around £480 on the stall. Her online shop does maybe £250 a week, so the market is the bigger number, but the online orders are the ones that arrive while she's asleep.

She oversold twice in her first two months. Both times it was a mug. Both times she made another and posted it late with an apology, and both times the buyer was fine about it, which she says is exactly why she didn't fix it for so long.

What she does now, in the order she does it:

1. Thursday. She counts everything and decides the split. Deep lines, meaning anything she has six or more of, stay listed online. Everything else comes off the site and goes in the market crate.
2. Saturday night. She prints the week's online orders on one sheet of paper. Item, name, and which shelf it's on.
3. Sunday morning. The printed sheet goes in her apron. Anything on that sheet is not for sale, no matter who's holding it.
4. Sunday evening. Whatever came back from the market gets counted and the numbers go back up online that night, not Monday.
5. Monday. Pack and post, then update stock again if she made anything new.

That printed sheet is the whole system. Everything else is admin around it.

She timed the Sunday-evening relist once and it was eleven minutes for about thirty items, which she describes as the cheapest eleven minutes of her week. Her takings didn't change much. What changed was that she stopped apologising to strangers, and she stopped being nervous about listing her good pieces online in case they sold twice.

The Monday job nobody plans for

Every seller doing both ends up with a weekly job they didn't budget for, and it's not packing. It's counting.

You have to physically count what came back from the market and reconcile it against what you left with. Ten minutes, every week, forever. Skip it once and your online stock numbers become fiction, and fiction is what causes the oversell three weeks later, long after you've forgotten which market caused it.

Some things that make the count faster:

• Count out, not just back. Photograph the crate before you load it. It takes four seconds and it settles every "did I bring the blue one" argument you'll have with yourself.
• One crate per product line. Mixed crates double your counting time.
• Write the count on the crate lid in chalk pen. Not on your phone. Your hands are full and your phone is in your back pocket under an apron.
• Do it before you drive home, if you can face it. The count you do on the pitch is accurate. The count you do on Monday includes the two you gave your mum.

Pricing the same thing in two places

Charge the same price. Both places, same number.

There's a temptation to price the market slightly cheaper because there's no postage and no fee, and it backfires within a month, because the person who bought a mug at the stall for £18 will see it online for £22 and feel something. Not outrage. Just a small, quiet reduction in trust, and you'll never hear about it.

The place to be different is delivery, not price. £22 collected at the stall and £22 plus £4.20 posted is a difference the buyer can see the logic of instantly. It also gives you a genuinely useful line for the stall: "same price online, postage on top, or come find me next Sunday and there's none."

The one exception is a market-only price on something you'd otherwise carry home. End of day, an hour before pack-up, discounting to clear is normal and everyone understands it. Just don't put it in a photo.

What Sailo won't do about your Sunday

There's no point-of-sale mode. Sailo doesn't run a till, doesn't talk to a card reader, and there's nowhere to type in a sale that happened in person.

That last part is the one that matters for this article, so be clear about what it means. Sailo can track stock per product and per variant, and when someone orders through your page, the unit comes off immediately, before payment, which is what stops two online buyers getting the same mug. But a mug you sell for cash at a market doesn't come off anything. There's no order for it. You either edit the stock number by hand on Sunday night, or you take the market unit off the site before you leave, which is why the physical box beats every clever alternative.

There's no app either. Everything happens in a browser, on your phone, on whatever signal the market has. Which on the third Sunday of the month, at a busy pitch, is a real consideration.

What to do before your next market

Do this on Thursday, not Saturday night.

1. Count everything and write two numbers per line: how many exist, how many are coming with you.
2. Take the market units off your online shop, or set the online stock to the number that's staying home.
3. Print your unfulfilled online orders on one sheet and put it in the crate.
4. Make a SOLD OUT card now, with your link on it, laminated if you're feeling organised. You'll use it.
5. Sunday night, before anything else, put the returned stock back online.

Then look at one number after four markets: how many times did you have to say "sorry, that one's actually gone". If it's zero, the box is working. If it isn't, you're still splitting the stock in your head, and your head is busy on a Sunday.