Pumpkin Patch & Corn Maze Ticketing

The vertical decides the vocabulary, the questions asked at checkout, the inventory model, the door, the payout terms and the theme. This page says "entry" and "visit" because that is what a farm says; a config whose primary call to action is generic ticketing wording fails validation rather than shipping.

The forecast turns on Thursday and the whole weekend turns with it

Autumn agritourism is a weather business with a six-week window and no second attempt at it. A washed-out Saturday in the middle of October is not a slow day, it is a tenth of the trading year, and the decision to close is made at seven in the morning by somebody standing at the edge of a field looking at how far the tractor sinks. Three hundred families have already booked, a good half of them would come anyway in wellies, and the phone starts ringing before the decision has been made.

What bindro does: Nothing in bindro decides that for you — there is no forecast anywhere in the product, and the call is still made at seven in the morning by somebody looking at how far the tractor sinks. Once it is made, closing the day is one control on the day itself: it tells you how many bookings are on it and how much money they are holding, takes it off sale, and applies one choice — refund everybody, give everybody a credit for another visit, or keep their booking for a day you will offer — to every family at once. Tick the box and each of them is emailed after the money has moved, told which of the three happened to theirs. Anyone who has already refunded themselves from their own confirmation link — which they can do until their slot begins — has closed their own booking, so the settlement passes over them and they are not written to twice.

Everybody in the county wants eleven o'clock on the Saturday of half-term

Demand at a patch is not spread across an open day; it is a spike late on a Saturday morning and a ghost town at three on a Tuesday. Sell an undated ticket, or a ticket that is only good for a date, and the entire county turns up inside the same forty minutes: the lane backs up onto the road, one tractor does the work of four, and the families who came in the afternoon walk a patch that has already been picked clean by the people who came at eleven.

What bindro does: Arrival windows are the inventory here rather than a line of small print on the ticket. Every open day is cut into slots, each slot holds its own capacity pool, and a pool that is gone reads "Fully booked" for that time while the next one keeps selling. Capacity moves only under a row lock inside the transaction that confirms the order, with a database constraint standing behind it, so the last place in the eleven o'clock cannot go twice — once to somebody refreshing the page in a car park and once to somebody else at the same instant. The day-of screen carries what is left in each window live while the gate is open.

The gate is in a field and the signal stops at the farmhouse

The one place on a farm where a spinning progress wheel costs real money is the point where the cars meet the gate, and it is reliably the furthest point from anything with a router in it. A scan that takes eleven seconds instead of one becomes a queue on a single-track lane, and a queue on a single-track lane becomes a decision to wave the next twenty cars through on trust — at which point the count for the day is a guess and the maze holds however many people it holds.

What bindro does: The door list is pulled onto the device over wifi before the gate opens, and from then on the scanning happens against what is already on it. A signed manifest of everybody due at that session is validated on the device with no network whatsoever, so a scan is instant in the middle of a field with one bar, and when the device is back within range it replays what it recorded: duplicate uploads count once, and two devices that admitted the same visitor surface in a conflict queue you read afterwards instead of arguing about at the gate. An entry scans as a code, or as a wristband once a band has been bound to that visitor, and a name can be found and admitted by hand when a screen has died in the rain.

A good share of the cars in the lane never booked anything

A pumpkin patch is a day out that a family decides on over breakfast. However much of the field sells in advance, a real proportion of the cars arrive with nothing but a card and an assumption that there will be room. Run that on a borrowed reader and a paper tally and the farm ends the Saturday with two sets of numbers for one day — the platform believes it sold four hundred entries, the reader disagrees, and nobody can say what the maze actually held at half past one.

What bindro does: The gate till sells out of the same pool as the booking page. A card taken in the lane draws down that arrival window exactly as an online sale does and under the same lock, so a walk-up cannot be sold a place that went thirty seconds earlier to somebody sitting at the end of the queue with a phone. Staff can also build an order from the back office — the school that phoned, the party that paid by bank transfer, the family whose card was declined at the gate — and all of it lands in one set of figures rather than two that have to be married up on Sunday night.

The entry price is the smallest number on the day

Gate money is a fraction of what a family spends between arriving in the car park and leaving it. The tractor ride, the pick-your-own bag, the doughnuts, the cider, the parking and the thing in the farm shop the four-year-old will not leave without are most of the takings, and they habitually live in a cash tin, a borrowed card reader and an honesty box on a hay bale. At the end of the season the farm knows what it banked and cannot say which part of the field earned it.

What bindro does: Add-ons sell at the same checkout as the entry and carry stock of their own, so a hayride with sixty seats sells sixty and then stops offering itself. The farm shop and the food stand get tills of their own that record what went out and what it took, and those takings deliberately never enter the platform ledger — bindro never held that money, and crediting your balance for it would be paying you a second time for the same cider. What you are left with is one season of numbers that has the gate, the add-ons and both counters in it.

The rota is a group chat and everybody logs in as the farm

Weekend staff at a patch are students, neighbours and family, the list changes every week of the season, and the schedule lives in a chat thread that four of them have muted. The person put on the gate at eleven on Saturday is frequently working their second ever shift, and the quickest way to get them scanning is to hand over the one login everybody uses — the one that also shows the bank details and every order the farm has ever taken.

What bindro does: Shifts attach to the sessions they cover and appear on the day-of screen beside the gate, so who is on and when is something the farm looks at rather than reconstructs. Being on the rota grants no access at all on its own: what somebody can see is decided by their role in your organisation, which is how a gate scanner is given check-in and nothing else — a name, and whether the code admits. The rule is applied in the handler rather than by hiding a button, so a shared owner login stops being the only practical way to staff a Saturday.

A season of arrival windows, generated rather than typed

Six weekends of half-hour arrivals is several hundred slots, and nobody should be entering them by hand in September. A visit carries the pattern — which days are open, which hours, how long a window runs and how many entries it holds — and the sessions are generated from it, each one holding its own inventory and its own sold count. Adding the half-term week or dropping the Mondays nobody came to is an edit to the pattern rather than an afternoon of typing.

• Each window sells against its own pool, so selling out at eleven leaves midday untouched and still on sale.
• More than one entry type can stand against the same day — a weekend price, a weekday price, under-threes issued as a comp.
• The price is derived on the server at the moment of payment, so a page a visitor left open since Tuesday cannot buy Tuesday's price.
• What is deliberately absent: a ladder of prices that opens and closes on its own schedule. This vertical does not resolve it, so nothing here pretends to.

Add-ons that run out when they have actually run out

A tractor ride, a bag for the pick-your-own, a parking space and a bag of feed for the goats are separate things with separate limits, and treating them as checkboxes on the entry is how a farm ends up with ninety people booked onto a trailer that seats sixty. Here each one is an add-on with its own price and its own stock, bought in the same basket as the entry and counted separately afterwards, so the attach rate is a number you can read rather than a feeling about the season.

• Stock is per add-on and enforced at checkout, not a note in the description.
• Add-ons appear on the same order as the entry, so one refund is one conversation.
• The trailer, the bag and the parking each report their own sales against the visit.
• Weighing a barrow of pick-your-own at the gate stays outside bindro: nothing here prices anything per pound.

A gate that admits with the phone in airplane mode

This is the strongest thing bindro does for a farm and the one place the honest answer has two halves. The model is real and complete: a signed manifest downloaded before the gate opens, scans validated locally against it with no network, first scan wins by timestamp, a replay that uploads the record afterwards and a conflict queue for anything two devices disagreed about. The half bindro does not ship is the handset — there is no scanner app, and the door is an interface your integrator talks to, with the console able to admit visitors one at a time from any browser for a gate small enough to work that way.

• Manifest, local validation and replay are the offline door; the camera is yours or your integrator's.
• Wristbands are real — a band is bound to a visitor through the API or from the visitor list in the console, and only its HMAC is stored.
• Re-entry is allowed and the door performs it: a visitor coming back through the gate that admitted them scans in again as a return, with the time they first arrived, on the device and online alike. The same code at a different gate reads as already used, and names the gate that used it.
• A visitor with a dead phone is found by name and admitted by hand, which is the failure mode a wet October guarantees you.

Two tills whose takings stay out of the platform ledger

The farm shop and the food and drink stand each get a till that records what sold, at the price you set, against the day it sold on. That money is yours the moment it is taken and it never enters the bindro ledger, which is not an omission but the point: a balance credited for cash the platform never held would be a balance that pays you twice for one jar of chutney. The gate till is the third of them and works the other way round, because a card taken at the gate is a real bindro sale drawing on real inventory.

• Shop and food takings are recorded for your numbers and settled by nobody but you.
• The gate till takes cards and draws down the arrival window it sold against.
• Cash recorded as an entry sale is not implemented, for the same reason the other takings stay out of the ledger.
• Everything all three tills recorded exports as CSV with the rest of the season.

Gift cards that spend like money in October

A farm gift card is a real instrument here rather than a code in a spreadsheet: it holds stored value, it is redeemed at checkout like any other tender, and what it pays for is entries and add-ons at whatever they cost on the day it is used. It is the nearest thing this vertical has to an autumn membership, and it is worth being exact about the difference — a gift card is money towards a visit, not a right to come back, because a pass that admits all season is not something this configuration resolves.

• Stored value, redeemed at checkout against entries and add-ons alike.
• Sold in advance, which is autumn money arriving in August.
• Not a season pass: nothing here admits a holder repeatedly without buying entries.

Sales tax shown on the order and held off your payout

You set a rate against the visit, the visitor is charged it on top of the price, and the tax is held on a remittance account of its own rather than mixed into your balance — so the figure you are paid is proceeds rather than a number you still have arithmetic to do on. Read the next sentence twice before you rely on any of it: bindro files nothing with any state, produces no return, and is not a tax service. The rate is a way to show tax on an order and to keep the money you collected out of a payout it was never part of.

• Tax is price-exclusive: charged on top, itemised on the order, visible to the buyer.
• Collected tax is a liability on its own account and is excluded from what settles to you.
• A refund gives the tax back on the same order rather than leaving you short.
• Filing, registering and remitting are entirely yours; nothing on the platform does any of the three.

The booking page, on the farm website you already have

Every visit gets a hosted page that works on a phone held sideways in a farm track with one bar, and the same checkout embeds into the site you have already paid somebody to build. Which one you lead with is a marketing decision rather than a technical one — the inventory, the prices and the arrival windows are identical either way, because both are reading the same visit. A farm with a good site keeps its visitors on it; a farm without one has a page it did not have to commission.

• The hosted page carries the visit, its windows, its add-ons and its opening times.
• The embed puts the same checkout inside your own pages, on your own domain.
• Both are the same inventory: a place sold through one is gone from the other.
• Mobile-first from a small phone upwards, because that is what a car park is.

Comps, codes, and a refund a visitor can do without ringing you

The small controls are the ones a farm reaches for weekly. Under-threes go free as comps rather than as an untracked hand-wave at the gate, the school and the birthday party are quoted with a promo code that carries its own limits, and a visitor who cannot come refunds themselves in full from their confirmation email up to the moment their window begins — which is a phone call the farm does not take on a Friday night. You also choose per visit whether the visitor carries the booking fee or you absorb it, and the fee is the same either way.

• Comps are issued and counted, so a free entry still occupies a place in the window.
• Promo codes carry their own limits, which is how a school visit or a party gets quoted here.
• A group rate that applies itself at twelve entries is a field on the entry type, and the two controls above are still there for a one-off quote.
• Self-service refunds run in full up to the start of the slot; after that the decision is yours alone.

Who pays the booking fee?

The visitor, unless you say otherwise. Every visit carries its own setting — the visitor pays, you absorb it, or you split it — and the fee itself does not change with the choice, only which side of the sale it comes from. On a ten-entry $25.00 visit that is $16.20 either added to what visitors pay or taken out of what you keep.

Because the default is the visitor, a farm can go from signing up to a sold-out visit without paying bindro anything up front, at any point, ever. We are paid out of sales that happened or we are not paid.

The word "event" is load-bearing and worth reading twice. The scheduler groups by visit, admits one only once its LAST session has ended, and pays it once. A visit sold as a run of dates therefore pays after the final date, not after each one — a season's float is not something an operator should discover halfway through the season.

Bindro never pays before delivery, in any vertical. A pre-event advance is a configuration violation platform-wide rather than a policy someone can be talked out of, because paying out on visits that have not happened is exactly how a cancellation becomes visitors with no refund. "Paid" also means the money moved: a payout only reaches its paid state with a real transfer reference attached, enforced by the database rather than by a status field someone can set.

• One payout per visit, after its last session ends, one in flight at a time.
• T+2 for this vertical (low risk tier), no reserve.
• Refunds reverse in a fixed order — your proceeds first, then tax, then our fee — so a refund never leaves you charged for a sale that was undone.
• Sales tax, where it applies, is held as a liability rather than mixed into your balance, so the payout figure is proceeds rather than a number you still have to do subtraction on.

The three numbers beside it are the ones this vertical is configured around, and each has an honest limit worth knowing before you rely on it. Weather sensitivity is a comparison you assemble: the platform records the day and what it took, and records nothing at all about the sky, so joining the two is a column you add in your own spreadsheet from your own memory of the morning. Group booking share is readable because party size is asked of every order. Add-on attach rate falls out of the add-on sales against entries sold on the same visit.

What the day-of screen is genuinely good at is the thing you need at half past ten on a Saturday rather than in November: how many are booked into the next window, how many have actually walked through, and how many places are left to sell in the lane. The season view is the other half — one row per visit, with what was taken, what was refunded and what the platform charged, reconciled against the ledger the payouts are settled from rather than assembled separately for display.

The one arithmetic bindro will not do for you is the one that spans the tills. Shop and food takings are recorded but never enter the ledger, so a true figure for what a weekend day earned is the ticketing rows plus the till rows, added by you. That is a deliberate trade: the alternative is a balance that claims money the platform never touched, which would be wrong in the direction that matters.

What the console shows today, without an integration or a spreadsheet:

• Gross, platform fees and refunds per visit, read from the ledger the payouts reconcile against.
• Entries sold against visitors checked in, with a no-show percentage per session.
• Live remaining capacity while the visit is running, per pool, on the day-of screen.
• This farm's own funnel — view → date selected → checkout → paid → scanned — stage by stage with the drop-off between them, and any step the platform cannot see said so rather than shown as a zero.
• Refunds, transfers, turnout, no-shows, add-on attach and repeat buyers, each printed with the two numbers it was divided from.
• Orders and visitors as CSV, so anything not on the screen is one export away.

Two more that are worth stating plainly. Sensitive answers are classified on the field rather than by convention, and health data is excluded from analytics exports and from AI context absolutely, with no override in any vertical. And a farm's own people see only what their role allows — a door login sees a name and whether the code admits, not what anyone paid — which is checked in the handler rather than by hiding a button.

• Every place on a booking of more than one entry must be named before the gate will admit it. A farm sells "four entries" and bindro makes a roster of four places, emails the booker a link and refuses any place still unnamed — on the offline door list and on a live scan alike. The booking page never asks for those names, because this vertical asks its questions of the order rather than of each visitor, so a family that ignores the roster email arrives with four entries that do not work. Tell your booker to fill it in, or sell single-entry orders, or budget for naming places at the gate on a phone.
• bindro will not tell you the weekend is washed out. Calling a day off is one control — it takes the day off sale and settles every booking on it as a refund, a credit or a kept booking, and emails each family if you ask it to — but nothing in the product reads a forecast, so the decision and its timing are yours. There is no general message tool either: that cancellation mail is the only one bindro sends to a day's visitors, so a change of parking still goes out from your own tool.
• bindro cannot price by weight or by the barrow. A group rate, a per-person add-on and a party price are all real controls now, but pick-your-own sold by the pound is not: nothing reads a scale, and entries and add-ons each have a price you set. Weighing at the gate stays on whatever you use today.
• bindro ships no scanner app. The offline door is an interface — a signed manifest your device downloads over wifi, scans validated on the device with no network, and a replay that uploads them later — and you or an integrator supply the thing that holds a camera. The console can check a visitor in one at a time from any browser, which is enough for a small gate and is not a queue of two hundred cars.
• bindro does not put a wristband on anybody, and does not sell you the bands. What it does is bind one: a band is bound to a visitor from the visitor list in the console or through the API, only its HMAC is stored, and the door then admits that band exactly as it admits a code. A farm not using bands scans the code in the confirmation email and everything else works the same.
• bindro does not sell pick-your-own by weight. Nothing on the platform reads a scale or prices anything per pound: entries and add-ons have a price each, and the till sells items at the price you set them. Weighing the barrow at the gate stays on whatever you use today.
• Money taken at the tills is never paid out to you, because it never left you. The merch till and the food and drink till record what you sold so your numbers are complete, and those takings deliberately never enter the platform ledger — crediting your balance for money bindro never held would pay you for it twice. Cash-recorded entry sales are not implemented at all for the same reason; the gate till takes cards.
• bindro does not remit your sales tax and does not pay it to you. You set a rate per visit, the buyer is charged it on top of the price, and it is held on a remittance account and left out of your payout. Nothing is filed with any state and no return is produced. Treat the rate as a way to show tax on an order, not as a tax service.
• No transfers and no date changes. A visitor who cannot come on Saturday cannot hand their entry to a neighbour or move it to Sunday: this vertical resolves neither ticket transfer nor deferral, and there is no credit note. On a weather-dependent day out that is the request you will get most often, and the only lever is a refund.
• No waiting list. When a slot is gone the page says "Fully booked" and stops — nobody joins a list, nobody is emailed when an entry is refunded, and a refunded entry does not return to sale by itself. If you want to resell it, raise that slot capacity by one.
• No season pass and no autumn membership. A pass that admits all season sits in this vertical optional list, which means the engine refuses it, so "come back any time in October" cannot be sold as one product. A gift card is the nearest real thing — it is stored value, not an entry, and it is redeemed at checkout like any other tender.
• No donation box. A patch raising money for a cause raises it in the entry price: this vertical does not resolve the donation add-on, so there is no "add a donation" control at checkout and nothing issues a receipt for a gift.
• bindro does not pay out mid-season. One visit gets exactly one payout, scheduled only once its last day has ended and released two days later. A whole autumn published as a single visit therefore pays once, in November — publish a visit per weekend if you need the money while the season is still running.
• bindro cannot enforce your refund policy. Every visitor can refund themselves in full from the link in their confirmation email, right up to the moment their slot starts — there is no notice window to configure and no fee to withhold. After the slot has started the decision is yours alone, from the console.
• bindro does not manage your car park, your queue or your access provision. Vehicle registration and access requirements are questions whose answers are stored and shown to you; nothing reserves a space, nothing holds a spot on the tractor and nothing promises a wheelchair-friendly route. Selling parking as an add-on with its own stock is real; directing the cars is yours.

If one of those is the thing you need, say so — the answer is a capability declared, built and gated properly, or a straight no. It is never a feature flag that collects the request and does nothing.

This is the structural decision that costs the most and looks the least important when you are setting up in August. Nothing about it is a policy you can be talked out of afterwards: the scheduler groups by visit, admits one only when its final session has ended, and pays it exactly once. A farm that published the whole autumn as one visit has not done anything wrong — it has simply chosen to be paid at the end of it.

The trade going the other way is real too, and small. Six visits means six sets of arrival windows to publish and six pages to link people at, rather than one page you point every poster and every post at all season. Most farms land on a visit per weekend or a visit per fortnight; a pick-your-own that opens for three weeks straight often stays as one. The cost of getting it wrong is cash flow rather than anything broken, and it is a decision you can make differently next autumn.

Two details that matter once the money starts moving. This vertical sits in the low risk tier, so no reserve is held back from what you are owed. And a refund reverses in a fixed order — your proceeds first, then the tax, then the platform fee — so refunding a family on a bad Saturday never leaves you charged for a sale that was undone.

This is the single most important paragraph on the page for a farm, because it lands on the median sale rather than on an edge case: almost nobody buys one entry to a pumpkin patch. A family of four pays, receives four entries and an email asking them to name the party, and if they do what most people do with that email they arrive on Saturday with four credentials that do not work. Publishing this on the landing page costs us signups. Discovering it in a queue of cars costs you the morning.

Three things work today, and you should pick one before you open rather than at the gate. Tell your bookers plainly, in the confirmation and on the page, that the roster link has to be filled in — most will, once asked directly. Or sell single-entry orders, so every entry names its own visitor at the point of sale and there is no roster to chase. Or plan for naming places at the gate from a phone, which is a minute of somebody's time per family and is workable for a small gate and unworkable for a busy one.

The gate guide walks the whole loop, including what a refused place looks like on the device so that the first time you see it is not on opening weekend. Rehearse it with a real order: buy several entries, leave the roster deliberately unfilled, watch the scan refuse, then fill the roster from a phone and watch the same entry admit. Ten minutes in a quiet field in September is the cheapest version of this lesson available.

The hour before opening is the one that decides how the day goes. The device that will be scanning downloads its manifest while it is still within reach of wifi, which is the step most likely to be forgotten and the only one that cannot be done later. The rota for the session is on the day-of screen, so whoever is running the gate can see who else is meant to be there. If you are selling to walk-ups, the gate till wants a card reader and a signal of its own, because taking money is the one thing that genuinely needs a network.

Through the morning the pattern is the same every twenty minutes: a car arrives, a code or a wristband scans, the device says yes, no or already-used, and nobody waits for anything to load. Places that were never named on a multi-entry booking refuse here, which is why they are worth chasing on Thursday. Walk-ups buy from the same window everybody else bought from, so the number in the field stays the number on the screen.

Afterwards the device replays what it recorded as soon as it is back within range, duplicates count once, and anything two devices disagreed about waits in a conflict queue for somebody to look at on Sunday with a cup of tea. The visit ends up with one set of figures: sold, scanned, refunded, taken at the gate, taken in the shop and taken at the food stand.

• Before opening: download the manifest on wifi, check the rota, check the gate till has signal.
• During: scan codes and wristbands offline, watch remaining capacity in the next window, sell walk-ups from the same pool.
• Awkward cases: find a visitor by name and admit by hand; name a roster place from a phone; build an order in the back office for the school that phoned.
• After: replay the scans, read the conflict queue, and export the day if you need it in a spreadsheet.

Square is the incumbent most farms actually have, and it is not a mistake — it is a better till than bindro is and always will be, with hardware you can buy on a Tuesday. What it does not have is an arrival window with its own inventory, or a gate that admits a named credential with no signal. Eventbrite brings an audience that bindro does not have and takes a fee that reflects it. Roller is built for attractions and is strong where the venue has a building and a network. FarmSpread is closest to this trade of the four.

The comparison pages are dated, and a stale one fails our own build rather than quietly ageing on the site — which is the obligation we think naming a competitor in public actually carries. They are also written to be read by somebody who is going to stay where they are: if the honest answer for your farm is that you should keep the reader you have for the shop and use bindro only for the gate, that is a reasonable outcome and several of the pages say so directly.

The one thing worth checking against whatever you are using now is not a feature at all. It is what happens to the money: when it lands, what is held back, what the export gives you if you leave, and whether a place in the field can be sold twice. Those four answers are on this page in the sections above, and they are the ones we would want compared.

The build guide covers what a six-weekend season really needs: how many slots, how long a window should be, whether the autumn is one visit or several, and what belongs in the entry price versus what belongs in an add-on. The pricing guide takes the quiet weekday problem seriously, because a patch that only works on Saturdays is a patch running at a third of its capacity, and it is honest about what cannot be published here: a pass that admits all season.

The weather guide is the one to read before you need it. Closing a day is one control that settles every booking on it, but it makes you choose between refunding, crediting and rearranging, and it needs somebody with the authority to move money — decisions worth making in September rather than at seven on a wet Saturday. Nothing predicts the weather for you, which is the part no software fixes.

Then rehearse the gate. Every failure worth finding is a failure of your particular field — where the wifi actually reaches, how the lane queues, what a refused credential looks like in the rain — and none of them show up from an office chair in August. The FAQ answers the rest, in the vocabulary a farm uses, and the plain-text summary is there for anyone who would rather hand the whole thing to a machine and ask it questions.