Wilson Wu
Chief Revenue Officer, Snappy
November 12, 2026·15 min read
A WeChat mini program reaches the customers who already live inside WeChat. A direct branded ordering page reaches everyone else, and it keeps the guest record and the margin. They answer different questions, which is why running both is usually the right call in Richmond, Vancouver and Burnaby, with one menu treated as the source of truth and one person responsible for it.
Three different sales calls land on the same operator here. One vendor sells a mini program that lives inside WeChat. One marketplace offers a Chinese-language app that brings new delivery customers and takes a cut of each one. One vendor sells a branded ordering page of your own.
All three are real channels. None of them is a strategy on its own.
A mini program is an application that runs inside WeChat rather than being installed from an app store. For a restaurant, it means a customer can open WeChat, scan the QR code on your table or find your store inside the app, read the menu, place an order and pay without ever leaving the conversation they were already having.
Several vendors sell this capability to Canadian restaurants. Some are payment companies that added a scan-to-order mini program; some are marketing agencies that build them per restaurant. Some build locally and some do not, which matters for support response time when something breaks on a Saturday night.
None of the ones we could find posts a price publicly. That is normal in this category and it is not a red flag on its own, but it does mean you cannot compare two mini program vendors from their websites. You have to make them quote.
WeChat reaches the regulars who already message you in Chinese: the table QR code, the group chat deciding lunch, the family that saved your store last time. Google reaches the person searching for your cuisine who has never heard of you. To size each group for your own restaurant, start with your order history, not the technology. Pull the last three months of tickets and sort them into three groups: the regulars who already message you in Chinese, the people who found you by searching, and the delivery orders you never see a real name on.
What WeChat reaches well is the customer who already knows you. The QR code on the table, the group chat where six colleagues decide where lunch is coming from, the family that saved your store the last time they ordered. It is a strong repeat and referral channel because the ordering flow sits inside the app your customer already has open.
What it does not reach is the person typing “dim sum near me” on their phone in a car on No. 3 Road. That search lands on Google, and it lands on whatever Google can show: your profile, your hours, your photos, your reviews and your ordering link if you have one. If your only ordering channel lives inside WeChat, that search resolves to a phone number at best, and to a competitor at worst.
There is a second group the WeChat channel misses: the English-speaking neighbour, the office worker ordering for a mixed team, and the tourist. Those orders are not marginal in Metro Vancouver, and they arrive through search, through delivery apps and through your own site. Our guide to promoting a restaurant on Google covers the profile side of that.
The honest framing is not WeChat versus Google. It is which of your customer groups each channel can serve, and whether you are currently serving one of them with nothing.
Which of your regulars does WeChat actually reach?
Bring a stretch of recent tickets and we will sort them with you: WeChat regulars, Google searches, and orders with no name on them.
Count my WeChat regularsBritish Columbia is one of the few places in Canada where part of this question has a legislated answer. The province’s permanent cap limits the fees delivery companies can charge restaurants to no more than 20% of the dollar value of an order, effective January 1, 2023 under the Food Delivery Service Fee Act, which passed November 3, 2022 (Government of British Columbia).
That cap exists because of what came before it. The province states the Act was passed in response to delivery companies charging restaurants fees as high as 30% of an order’s value during the pandemic (Government of British Columbia). On a $40 order, the difference is real money: 40 x 0.30 = $12 in fees before the cap, against 40 x 0.20 = $8 at the cap, a reduction of $4 per order in the province’s own worst case.
Read the next sentence of that release carefully, because it is the one operators forget. The Act also allows delivery companies to offer optional, enhanced services for restaurants to opt into at their discretion (Government of British Columbia).
Marketing placements, promoted listings and premium delivery tiers are opt-ins, and they sit on top of the capped fee. If your effective rate looks higher than 20%, open your statement and find the opt-ins before you call anyone.
Mini program pricing is a different exercise, because there is little public pricing to compare. Ask every vendor for the same five numbers in writing:
The fifth one is where mini program deals surprise people. A channel that holds your funds for a week is a working capital decision, not just a marketing decision. For the broader math on marketplace fees, see our breakdown of how to reduce food delivery commission fees.
An order that prints straight into your kitchen costs zero keystrokes. An order that lands on a separate tablet costs one person standing at that tablet through the whole rush. Every channel you add lands somewhere, in one of those two places.
The worst version of this is familiar to any busy Chinese restaurant in the Lower Mainland: a shelf of tablets by the pass, each one chiming, each one belonging to a different channel, and one staff member whose real job during the dinner rush is re-typing orders into the register. That staffer is the true cost of a channel, and it never appears in the vendor’s quote.
The better version routes each channel into one order flow. Snappy’s online ordering writes into the same POS the counter runs on when you run Snappy POS, orders route to kitchen stations through the kitchen display system, and Snappy’s delivery app integration brings third-party delivery orders directly into the POS and KDS rather than onto their own tablet. One screen, one ticket sequence, one place to look when a customer calls about an order.
For a WeChat mini program, ask the vendor one blunt question: does the order print in my kitchen without anyone re-typing it, and if not, what exactly does my staff do at 7 PM on a Saturday? A vendor with a clean answer will describe a printer or a POS route. A vendor without one will describe a dashboard, which means a person.
Then test it before you sign. Place a real order through the channel during a real rush, with modifiers, with a note, and with an item that is out of stock. What happens to that order is the product. Our guide to a Chinese restaurant POS system goes deeper on the register side of that decision.
On a marketplace, the customer belongs to the marketplace. On a mini program, ownership depends on the vendor’s contract terms. On your own ordering page, the customer is yours by construction. Channels differ on this one thing that compounds over years: after the order, who has the customer.
On a marketplace, the customer belongs to the marketplace. You get a name on a ticket and a delivery address, and the platform gets the ordering history, the email, the push notification channel and the ability to promote the restaurant next door to the same person next Friday. That is the deal, and for new-customer acquisition it can be a fair one. It is only a bad deal when it becomes your only channel.
On a mini program, ownership depends on the vendor’s terms. Ask directly whether the customer list, the ordering history and the follow-up messaging rights belong to you or to the platform, and ask what happens to that list on the day you cancel. Get the answer in the contract, not in the sales call.
On your own ordering page, the customer is yours by construction. That is what makes a direct channel worth building even when it starts small: Snappy’s restaurant marketing services cover the website and Google side, and Snappy’s CRM holds customer profiles and runs segmented campaigns off them, which is only possible because the profile exists in the first place. A restaurant that has run direct ordering for two years owns a list. A restaurant that has run marketplace-only for two years owns a rating.
Payment flow deserves the same scrutiny. Ask which entity is the merchant of record, who handles a refund at 9 PM, who eats a chargeback, and whose name appears on the customer’s card statement. A guest who does not recognize the name on the statement calls their bank, not you.
Here are the four channels side by side on the dimensions that decide the money. Where a number is not published, the cell says so rather than guessing.
| Channel | Who it reaches | Published cost to the restaurant | Where the order lands | Who owns the customer record | Shows up on Google |
|---|---|---|---|---|---|
| WeChat mini program (vendor built) | WeChat users who already know you, scan your in-store QR code, or find you inside the app | Not posted publicly by the vendors we could find. Build fee, monthly fee, per-order fee and processing rate are all quoted case by case | Depends entirely on the vendor. Often a separate dashboard or tablet unless a printer or POS route is built | Usually the vendor's platform. Ask for the answer in the contract | No. The experience lives inside WeChat |
| Chinese-language delivery marketplace | Chinese-speaking customers browsing an app for delivery, including people who have never heard of you | Commission not published by the platforms. Where the platform is a delivery company under the Act, BC caps its fee at no more than 20% of order value, with optional paid services on top (BC) | Typically a tablet or app the marketplace supplies, unless an integration into your POS exists | The marketplace | Their listing may rank. Your own record does not benefit |
| Mainstream delivery apps | Broad, English-first delivery demand across Metro Vancouver | Same BC cap applies to the fee a delivery company charges you (BC); opt-in marketing and premium tiers sit above it | Their own tablet, or straight into the POS and KDS where a delivery app integration exists, as with Snappy | The app | Their listing, not yours |
| Direct branded ordering page (Snappy Online Ordering) | Anyone who searches for you or your cuisine on Google, plus your regulars, in-store QR traffic and phone customers you redirect | Commission-free per order. Plan pricing depends on the rest of your stack, ask for a quote | Straight into the POS and the kitchen display system | You. Profiles land in your CRM | Yes. The order link surfaces on your Google profile |
Three rows in that table are acquisition channels and one is a retention channel, and they are not substitutes. A marketplace, the delivery apps and a mini program all find you customers. A direct page keeps them, and it is the only row where the cost per order does not scale with success.
For a Chinese restaurant in Richmond, Vancouver or Burnaby, the answer is usually both, with a clear hierarchy. Run the acquisition channels your market actually uses, and run a direct page underneath them so that every customer they send you can be kept without paying twice for the same person.
The failure mode of running several channels is never the technology. It is menu drift. Prices go up on the POS and not in the mini program. An item is 86’d on Friday and stays live in two apps until Sunday. Holiday hours are right in one place and wrong in three. Every one of those is a refund, a bad review or an argument at the counter.
The fix is boring and it works. Name one system as the source of truth, which for most operators is the POS menu, because that is what the kitchen and the register already obey. Name one person who is allowed to change a price, in that system first and everywhere else the same day. Then run a five-minute check every week:
Then put the direct link everywhere it is free to put. On the Google profile, where Snappy’s online ordering integration surfaces it. On the QR code on the table. On the printed receipt. In the reply you send when someone messages the restaurant to ask whether you deliver. A direct channel does not grow because it exists, it grows because it is the easiest link to find. Our guide to commission-free online ordering systems covers what to look for in that system, and if you are large enough to want your own app rather than a page, Snappy also builds a branded restaurant app on the same rails.
It is a menu and ordering flow that runs inside WeChat rather than as an app your customer installs. Customers reach it by scanning a QR code in your restaurant or by finding your store inside the app, then view the menu, order and pay without leaving WeChat.
None of the vendors we could find selling into this market posts a price publicly, so every figure you hear will be a custom quote. Ask for the build fee, the monthly fee, any per-order fee, the payment processing rate and the settlement timing in writing before you compare two vendors.
No more than 20% of the dollar value of an order under the permanent cap that took effect January 1, 2023 (Government of British Columbia). The same Act allows delivery companies to offer optional enhanced services you opt into, so check your statement for opt-ins if your effective rate looks higher.
Yes, because they reach different people. WeChat reaches WeChat users, mostly ones who already know you, while a direct page reaches whoever is searching Google for your cuisine right now and keeps the customer record afterward. The two together cover a market like Richmond in a way neither covers alone.
Only if one system is the source of truth and everything else is updated from it, in practice the POS menu. Give one person the authority to change prices there first, then run a weekly check on prices, unavailable items and hours across every channel.
WeChat and Google are not competing answers to one question. They are two different rooms your customers are standing in, and a restaurant in Richmond or Burnaby with a large Chinese-speaking base has good reason to be in both. The person searching “dim sum near me” and the person scanning a QR code at your table are rarely the same person, and you do not have to pick which one you serve.
Judge every channel on four questions: who it reaches, what it actually costs once opt-ins are counted, whether the order lands in your kitchen without a keystroke, and who holds the customer afterward. In British Columbia the fee question has a legislated ceiling of 20% of order value on what a delivery company charges you (Government of British Columbia), which is a ceiling you can plan around, not a rate you should accept on every order you take.
Run the acquisition channels your market uses. Then run a commission-free direct page underneath them, on the POS your kitchen already reads, so the customers those channels bring you can come back through a door you own.
Put a direct ordering link on your Google profile alongside whatever WeChat channel you keep, and let the two do different jobs. Talk to our team.
Book a walkthrough
Where would each channel's order land in your kitchen?
Book a walkthrough and we will go channel by channel: who it reaches, where the ticket lands, and who holds the customer afterward.

Written by
Wilson Wu
Chief Revenue Officer, Snappy
Wilson Wu is Chief Revenue Officer at Snappy, the restaurant technology company, and is based in Boston. His first job was in a restaurant kitchen, and he has spent his career since at the intersection of hospitality and software, working with operators across Canada on the systems their service runs on. He also builds the technology he writes about. His engineering work includes a per-location customer service agent that answers guests with a restaurant's real hours, menu and prices, deployed with independent restaurants, and a document-extraction model that reads supplier invoices and matches invoice numbers against what is owed. He writes here about what changes on a busy service: ordering, payments, loyalty, the phone, and the Canadian rules behind them. He holds an MBA from Duke University and a Master of Science in Computer Science from Georgia Tech.
Connect on LinkedInNo-obligation walkthrough
Which channel reaches which customers?
See which of your regulars WeChat reaches, which Google reaches, and where each order actually lands in your kitchen.
On this page
Get The Latest Restaurant Data, Trends & Tips