# Platform Dependency
**Situation E**
***
## The dependency problem
Platform dependency is comfortable. Orders come in. The kitchen is busy. Revenue grows. Someone else handles the marketing, the logistics, and the customer acquisition. All you have to do is cook.
And that is exactly the problem.
When a marketplace accounts for a large share of your revenue, you are not building a business. You are operating a kitchen inside someone else's business. The platform owns the customer. The platform owns the data. The platform controls your visibility — which listings appear first, which restaurants get promoted, which ones get buried. The platform sets the commission rate, and you have no leverage to negotiate because the customers are theirs, not yours.
This is not a criticism of marketplaces. They serve a real function — they aggregate demand and make discovery easy for customers. For many restaurants, they are an important channel and should remain one. The problem is not being on a platform. The problem is being dependent on one.
Dependency means that a decision made in a product meeting you were not invited to can change your revenue overnight. A commission increase. An algorithm change. A new competitor who pays for premium placement. A policy shift on delivery radius. You have no control over any of these, and no recourse when they happen.
The question is not whether to leave the platform. It is whether to build something alongside it that you actually own.
***
## Quantify the dependency
Before you act, understand the scale of the problem. Most restaurants know they rely on a platform. Few know exactly how much.
**Revenue share.** What percentage of your total revenue comes through the marketplace? If it is above 50 percent, you have a significant dependency. If it is above 70 percent, the platform is effectively your business and you are a supplier to it.
**Real margin after commissions.** A $50 order through a marketplace might generate $32 to $38 in revenue after the platform takes its cut. The same $50 order through your own channel might generate $47 to $50. Run the numbers on your actual orders — what is your effective margin on marketplace revenue versus direct revenue? The gap is often larger than restaurateurs realise, because they think in terms of top-line order value rather than what they actually keep.
**Customer visibility.** Of all the customers who have ordered from you through the platform, how many can you name? How many can you contact? How many have you ever communicated with directly? The answer for most restaurants is zero. Those are the platform's customers, not yours. They may have ordered from you 20 times, but if the platform delists you tomorrow, you have no way to reach them.
**Concentration risk scenario.** Ask yourself: what happens if the platform raises commissions by 5 percent? What happens if a competitor in your category pays for promoted placement and your orders drop 20 percent? What happens if the platform changes its algorithm and your listing drops from page one to page three? If any of these scenarios would materially hurt your business, the dependency is real and it needs to be addressed.
***
## Understand who the platform customer actually is
Here is something most restaurants do not consider: the customer who orders from you on a marketplace may be fundamentally different from the customer who would order from you directly.
A marketplace customer is often browsing. They opened the app because they are hungry, not because they specifically wanted your restaurant. They scrolled through options, maybe filtered by cuisine or price, and chose you — possibly because of your rating, your photos, a promotion, or simply because you appeared near the top. Their loyalty is to the platform and the convenience it offers, not to you.
This does not mean they cannot become your customer. But it means you need to think about the transition differently. You are not just asking them to change where they place the order. You are asking them to go from a passive, platform-mediated relationship to an active, direct one. That requires a reason.
Some marketplace customers will never switch. They value the convenience of one app for all restaurants, and the friction of downloading your app or visiting your website is not worth it to them. That is fine. The goal is not to convert every marketplace customer. It is to identify the ones who order from you repeatedly — the ones who have already chosen you multiple times within the platform — and give them a reason to come direct.
***
## Define who the direct channel is for
Not every customer is a direct channel customer. The casual browser who opens GrabFood because they are hungry and picks the first thing that looks good — that customer may never switch, and that is fine.
The direct channel customer is different. They believe in your brand. They have already chosen you multiple times. They tend to order bigger — for a group, for an occasion, in advance. They are not impulse ordering. They are planning.
Two segments matter most:
**Gatherings and hosting.** When someone is hosting friends or family at home, they are not scrolling a marketplace. They want something they trust, something that will impress, and something easy to order for a group. This is a direct channel customer. Make it easy for the host: create bundles built around your signatures, clearly labelled by number of people — "feeds 4," "feeds 8," "feeds 12." Remove the guesswork. The host does not want to assemble an order item by item from a long menu. They want to pick a bundle, add a few extras, and know it will be good.
**Office orders.** Corporate meetings, tea breaks, team lunches, office parties. These orders are larger, more frequent, and often placed in advance. The product format is different — bentos for individual portions, sharing platters for communal settings, snack boxes for tea breaks. These are not the same products you sell on a marketplace. They are designed for a different occasion and a different buyer (often an admin or office manager who reorders regularly once they find something that works).
These two segments share a common trait: the order size justifies the effort of ordering direct. A single person ordering one meal has little reason to leave the convenience of a marketplace app. A host ordering for 10 people or an office manager ordering weekly platters has every reason to go direct — for better pricing, for custom bundling, and for a relationship where you remember their preferences.
***
## Design a distinct offering for direct
This is critical. If your direct channel sells the exact same menu at the same prices as the marketplace, you are asking customers to switch for no reason. The direct channel needs its own identity.
**Product differentiation.** Your marketplace menu is built for individual orders — one person, one meal, delivered fast. Your direct channel menu should include everything the marketplace cannot or does not offer: group bundles, sharing platters, party sets, corporate catering packages, advance-order specials. These products only exist on your direct channel. They give customers a reason to come to you that the marketplace cannot replicate.
**Signature bundles.** Take your two or three signature dishes — the ones from [Lever 4](/docs/guides/how-to-think/the-growth-levers.md) in Chapter 5 that you want every customer to try — and build bundles around them. "The Signatures Bundle for 4" that includes your best dishes, a side, and drinks. Priced attractively, packaged well, and impossible to order on GrabFood. This does two things: it drives direct channel traffic and it ensures the customer experiences your best food.
**Pricing logic.** Your direct channel can and should offer better value than the marketplace. You save 25 to 35 percent in commissions on every direct order. Pass some of that saving to the customer as a price advantage or as more generous portions and bundling. The customer gets a better deal. You keep a better margin. The marketplace cannot match this because their commission structure does not allow it.
The principle: the marketplace is for individual, impulse, convenience orders. The direct channel is for planned, group, and high-value orders. Different products, different occasions, different value proposition. When you make this distinction clear, you are not asking customers to switch — you are offering them something they cannot get elsewhere.
***
## Building the direct channel
The direct channel is the foundation of independence. It is the thing that turns your restaurant from a supplier on someone else's platform into a business that owns its customer relationships.
**What "direct" means:**
Your own online ordering — a website or app where customers can order delivery or pickup directly from you, without a marketplace intermediary. You capture their name, email, phone number, and order history. You control the experience, the pricing, and the communication.
**Why direct matters beyond margin:**
The margin improvement is obvious and important. But the real value of direct is the customer relationship. A direct customer is a known customer. You have their data. You can segment them. You can send them a message when they have not ordered in two weeks. You can tell them about a new dish. You can offer them a loyalty reward that builds habit. You can learn what they order, when they order, and how often. None of this is possible with a marketplace customer.
Over time, a direct customer becomes more valuable than a marketplace customer in every dimension: higher frequency (because you can prompt them), higher lifetime value (because you can retain them), higher margin (because you are not paying commissions), and lower churn risk (because the relationship is with you, not with a platform).
**How to build it:**
This does not require a massive investment. A well-designed online ordering page on your website, integrated with your kitchen workflow, is enough to start. The key requirements are: it must be easy to use (at least as easy as the marketplace), the product offering must be distinct and compelling (group bundles, catering, advance orders), and it must capture customer data at the point of transaction.
Do not wait until it is perfect. Launch it, learn from it, and improve it. The first version does not need to match the marketplace in features. It needs to exist so you can start building the direct customer base alongside the marketplace one.
***
## Migration: find your high-value customers and go after them
The biggest fear restaurants have about building a direct channel is cannibalisation — "if I push customers to order direct, will I lose total order volume?" The fear is understandable but usually misplaced. The goal is not to shut down the marketplace channel. It is to build a direct channel that grows alongside it, and over time, shifts the mix.
But you need to be decisive about how you swing customers over. Consumers are sticky to platforms. Inertia is strong. A passive "we also have a website" message will not move anyone. You need to be specific, targeted, and give them a clear reason.
**Map your high-value customers across channels.**
Start with your marketplace data. Even though you do not own the customer relationship, you can see the orders. Look at basket sizes. Among your marketplace orders, there will be small baskets (individual meals) and large baskets (group orders, higher spend). The large basket customers are your migration targets — they are already spending more, which means they are more likely to be ordering for a group or an occasion, which is exactly what your direct channel is designed for.
**Target the large baskets with a physical touchpoint.** For every marketplace order above a certain basket size, include a flyer. Not a generic insert card. A specific message: "Ordering for a group? Next time, order direct and get our exclusive sharing platters and party sets — plus 10 percent off your first direct order." Explain why the offering is different. Show them what they cannot get on the marketplace. Make the first direct order frictionless with a clear incentive.
**Consolidate your customer data from every source.** Your database is not just your marketplace orders. It is your reservation system, your loyalty programme, your CRM, your enrolment data, your email list. Pull it all together into one view. These are people who already have a relationship with your brand across different touchpoints.
**Use that consolidated audience to run targeted ads.** Upload your customer database into Meta and Google as a custom audience. Run ads to this audience specifically — not generic brand awareness ads, but ads that announce your direct ordering channel and its distinct offering. "Now you can order our signature sharing platters for your next gathering. Only on our direct channel." These are people who already know and like you. The conversion rate on this audience will be significantly higher than any cold targeting.
**Build lookalike audiences from your best customers.** Once you have your custom audience running, create lookalike audiences on Meta and Google — people who resemble your existing high-value customers but have not discovered you yet. This extends your reach beyond your current database while maintaining targeting quality.
The principle: do not broadcast to everyone. Identify your highest-value customers, reach them through every channel you have — physical inserts, email, SMS, paid ads — and give them a specific, compelling reason to try your direct channel. Then let the product and the experience do the retention work.
***
## The healthy channel mix
The goal is not to eliminate marketplace revenue. It is to reach a mix where you are not dependent on any single channel, and where the channels you own represent a meaningful and growing share.
**What a healthy mix looks like:**
There is no universal target because it depends on your restaurant type, your location, and your customer base. But as a directional guide: if your direct channel (own website, own app, own ordering) accounts for a growing share of your online orders month over month, you are moving in the right direction. If the marketplace share is shrinking as a percentage even while the absolute numbers hold or grow, that is healthy — it means total demand is growing and the direct channel is capturing the growth.
**What to track:**
**Direct vs. marketplace order share.** The percentage of total online orders that come through your own channel versus the marketplace. Track this monthly. The trend matters more than the absolute number.
**Direct customer repeat rate.** Are customers who order direct coming back? If your direct channel is capturing one-time customers who never return, the channel exists but the relationship is not forming. Compare the repeat rate of direct customers to marketplace customers — direct should be higher over time, because you have the tools to reactivate them.
**Migration rate.** Of customers who first discovered you on the marketplace, how many have placed at least one order through your direct channel? This tells you whether your migration mechanics (insert cards, incentives, packaging prompts) are working.
**Blended margin.** As your channel mix shifts, your blended margin per order should improve. Track this alongside revenue to ensure that the shift to direct is actually improving economics, not just moving orders from one place to another.
***
## What not to do
**Do not delist from the marketplace in protest.** Unless your direct channel can fully replace the volume, pulling off the platform costs you revenue and customers. The platform is a discovery channel. New customers find you there. The goal is to migrate them, not to cut off the supply.
**Do not get into a price war on the platform.** Competing on discounts and promotions within the marketplace is a race to the bottom. The platform benefits from price competition between restaurants. You do not. If you are going to offer a price advantage, offer it on your direct channel where the savings come from commission reduction, not from your margin.
**Do not ignore the platform experience.** Even as you build direct, your marketplace listing still represents your brand to new customers. Bad photos, an outdated menu, slow preparation times, or poor ratings on the platform will hurt your brand everywhere, not just on that channel. Maintain the marketplace presence as a discovery and acquisition tool. Just do not let it be the only tool.
**Do not assume all marketplace customers will switch.** Some will. Many will not. The goal is to build a growing direct base alongside the marketplace, not to convert every marketplace customer. A realistic expectation is that your most frequent marketplace customers — the ones who have already chosen you multiple times — are the most likely to switch when given a reason. The casual, one-time browser will probably stay on the platform, and that is fine.
***
## The long game
Platform dependency is not solved in a month. It is a gradual shift in how your business operates and where your customers come from.
In the short term, the marketplace is still your biggest online channel. That is okay. What matters is that you have built the direct alternative and it is growing.
In the medium term, the direct channel becomes a meaningful share. Your customer database is growing. Your reactivation mechanics are running. Your repeat rate on direct is higher than on the marketplace. You are spending less on commissions and more on building your own brand.
In the long term, you have a business that uses platforms for what they are good at — discovery and incremental reach — while owning the core customer relationship. A commission change no longer threatens your business. An algorithm shift no longer determines your revenue. You have customers who know you, who order from you directly, and who come back because you have given them a reason to.
That is the difference between renting your revenue and owning it.
***
## How Oddle helps in this situation
**Oddle Shop** — The direct channel itself. Group bundles, catering, sharing platters, advance orders. Further delivery radius. Advance orders expand kitchen production hours.
**Oddle Enrolments** — Captures direct customers, builds the owned database. Front-loaded return rewards for repeat on direct. Compare direct vs. marketplace customer quality.
**Oddle Marketing** — Reaching your consolidated audience. Announcing the direct channel. Reactivation sequences. Turns a database into repeat revenue.
**Oddle Terminal** — Captures dine-in customers through card tokenisation. Consolidate with online data. Upload to Meta/Google for targeted ads.
**Oddle Reserve** — Another touchpoint feeding the consolidated database. Occasion data helps design group and celebration bundles.
---
# Agent Instructions: Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://www.oddle.me/docs/guides/what-to-do/platform-dependency.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.