Friday, 6:40 p.m. Three tablets on the pass are chirping at once. The expo is reading tickets off DoorDash, Uber Eats and your own ordering page, and the driver for order #4471 is leaning on the counter asking how much longer.
Order #4471 is a woman who has eaten at your place in Spring every other Friday for four years. She ordered through the app because the app was on her phone and your website wasn’t. On a $46 order, the app kept somewhere between $7 and $14 of your money, and it kept her email address too.
That’s the problem this post is about. You probably can’t quit the delivery apps, and we wouldn’t tell you to. They find you new customers, and some of those customers will only ever order through an app. What you can do is stop paying a marketplace commission on regulars who already know you. The job is to make your own ordering channel the easiest way to order for anyone who has ordered before, and then to measure the mix every week like it’s food cost.
Our general restaurant marketing guide covers reviews, social and slow nights. This one stays on ordering.
The commission math, per order, at 2026 rates
Start with what each order actually costs. The published marketplace rates as of this writing:
- DoorDash. Basic is 15% of the order subtotal, Plus is 25% and Premier is 30%. Pickup orders are 6% on every plan. There’s no activation or subscription fee, and payment processing is included in the commission, per DoorDash’s merchant pricing page.
- Uber Eats. Lite went from 15% to 20% in March 2026, Plus stayed at 25% with Uber One member orders moving to 30%, and Premium stayed at 30%, per Restaurant Dive. Pickup went from 6% to 7% when you have validated in-store pricing, and 10% when you don’t. Uber Eats’ merchant pricing page also notes Lite rates are lower in some cities, so check the rate for your address.
The tiers aren’t identical products. Each plan comes with different reach and perks, so read what yours includes on the pricing page. A cheaper tier that brings fewer orders isn’t cheaper, so compare what you paid per order you actually got, not the headline rate.
Here’s what a $40 subtotal costs you by channel. The direct rows use ChowNow’s published processing fee and Uber’s Webshop fee as examples; your own processor will differ a little.
| Channel | Fee rate | Fee on $40 | You keep |
|---|---|---|---|
| DoorDash Premier or Uber Eats Premium, delivery | 30% | $12.00 | $28.00 |
| DoorDash Plus or Uber Eats Plus, delivery | 25% | $10.00 | $30.00 |
| Uber Eats Lite, delivery | 20% | $8.00 | $32.00 |
| DoorDash Basic, delivery | 15% | $6.00 | $34.00 |
| Uber Eats pickup, no validated in-store pricing | 10% | $4.00 | $36.00 |
| DoorDash pickup | 6% | $2.40 | $37.60 |
| Your own site, pickup (2.95% + $0.29 processing) | ~3.7% | $1.47 | $38.53 |
Two things fall out of that table. The biggest saving is moving a regular from app delivery to direct pickup, about $8.50 to $10.50 an order at the common tiers. And app pickup is already fairly cheap, so if a guest insists on the app, nudging them to pickup inside the app still helps.
Direct delivery is a different calculation. You still need a driver. Uber Direct starts at $7.99 per delivery, ChowNow lists Flex Delivery at $7.98 an order, and Toast sells delivery as a flat fee per order. Most restaurants pass some or all of that to the guest as a delivery fee. Even then the direct order is usually cheaper than a 25% or 30% commission on anything above about $30.
Is there a delivery fee cap in Houston?
No. We get this question a lot because Chicago, San Francisco, New York and others capped delivery commissions during the pandemic, and some kept the caps. Houston never did, and Texas went the other way.
In 2021, a bill to cap delivery app fees in Texas at 15% (HB 598) drew plenty of press. Katy restaurateur Alex Au-yeung of Phat Eatery told the Houston Chronicle at the time, “If you give 30 percent away, how can we survive?” The bill died in committee.
What passed instead was SB 911, which created Chapter 114 of the Business and Commerce Code and took effect January 1, 2022, per the Legislature’s bill analysis of SB 911. In plain terms:
- A delivery service can’t charge you a fee, or make you absorb one, unless a written agreement authorizes it and states each fee.
- It can’t use your name or logo in a way that suggests you endorse it.
- If you have no agreement and ask to be removed, it has 10 days to take you down, and it can’t re-add you without written consent.
- Cities and counties can’t adopt rules that affect the terms of those agreements, which effectively rules out a local Houston cap.
So in Houston the fee is whatever you signed. Pull your current agreement, find the tier, and check whether marketing programs or promotions add charges on top. Then decide on the tier with real order data in front of you. We’re marketers, not your lawyer, so anything contractual goes past them.
Choosing a first-party ordering platform
You need an ordering page that’s yours, plugged into your POS, fast on a phone, and able to remember a guest. Most independents and small groups end up with one of these:
| Option | What it is | Published pricing notes | Best fit |
|---|---|---|---|
| Toast Online Ordering | Ordering built into Toast POS, with loyalty and a Google ordering integration | Commission-free ordering and flat delivery fees, per Toast’s online ordering page; pricing by quote | Restaurants already on Toast |
| Square Online | Ordering site tied to Square POS | Free plan with no added monthly fees, paid plans above it, per Square’s online ordering page | Counter service, cafés, food trucks on Square |
| ChowNow | Standalone ordering, branded apps and marketing tools | Entry plan listed at $229 to $249 a month, setup $119 to $499, 2.95% + $0.29 processing, per ChowNow’s pricing page | Independents on a POS without strong native ordering |
| Olo | Enterprise ordering, payments and loyalty for multi-unit brands | Custom contracts; Olo says it serves more than 750 restaurant brands | Franchise and multi-unit groups |
| Online Ordering by DoorDash / Uber Eats Webshop | Commission-free ordering sites run by the apps | DoorDash: 0% commission plus processing; Uber: 2.5% + $0.29 per order | Fast start; you stay inside the app company’s system |
How we’d pick. Native to your POS beats bolted on, because menu changes, 86’d items and prep times sync without someone retyping them at 5 p.m. The checkout should take a returning guest under a minute on a phone. And make sure you own the guest list: name, phone, email and order history, exportable. If you can’t download your customers, you’ve swapped one landlord for another.
Speed matters more than design. We’ve seen ordering pages that took eight seconds to load on LTE in a parking lot on Kuykendahl. Test yours on your own phone, off Wi-Fi, before you spend a dollar sending people to it.
Make Google send orders to you
Plenty of “order online” journeys start on Google, not on an app. Someone searches your name or “pho near me,” sees your Business Profile, and taps Order. Which provider that button opens is a setting most owners have never looked at.
Per Google’s help page on food ordering, in your Business Profile go to Food ordering, pick your own ordering provider (or add your link at the bottom with Add a link), then select Set as preferred and turn on Preferred for pickup and Preferred for delivery. You can also choose Show provider to customers on any third-party provider and switch it off for pickup, delivery or both. Changes usually show up in Search and Maps within a few hours.
Hiding the apps entirely is a judgment call. If 30% of your delivery volume comes through Google and you don’t run your own drivers, hiding DoorDash may just send that guest to the next restaurant. A middle path we like: set your own link as preferred for pickup, keep one app visible for delivery, and revisit after a month of data.
The rest of Google matters too. Put a large “Order pickup” button at the top of your website on mobile, ahead of the hero photo. Make sure the menu on your profile matches the menu on your ordering page. If you run Google Ads with location assets, the “Order” clicks from your profile show up as a local actions conversion, though Google treats it as reporting-only and won’t bid on it, per Google Ads’ local actions documentation. It’s still a free read on how many people try to order from Maps.
When we opened locations for The Toasted Yolk Cafe, location pages and Business Profiles went live before the doors opened, and the case study shows 21 days to No. 1 in the local map pack. A profile that ranks is only half the job. The order button has to point somewhere you keep the margin. More on profiles in our Google Business Profile guide.
Menu pricing parity: what to charge where
A lot of restaurants mark up their app menus 15% to 25% to cover commission. That’s a normal practice and nothing to be embarrassed about. The question is what you show on your own channel.
Our default: your direct ordering page matches in-store prices, and the apps carry the markup. Then the guest who compares sees that ordering direct is cheaper, and you can say so in plain words on the bag, the receipt and the website. “Same menu, lower prices when you order from us.”
Watch one rule. Uber Eats now ties its pickup rate to price parity: 7% with validated in-store pricing and 10% without, per Uber Eats’ merchant pricing page. If your Uber menu is marked up, your Uber pickup orders cost more. Some restaurants keep a separate, unmarked pickup menu for that reason. Your POS or menu management tool should make that a setting rather than a second menu someone forgets to update.
Don’t try to win on discounts alone. A permanent 20%-off code on direct orders teaches guests to wait for a code. Price parity plus a loyalty reward gets you there with less margin given away.
Loyalty and SMS: give the regular a reason
Here’s where direct ordering pays for itself. Your regulars already like the food. They need a reason to open your link instead of the app, and the reason is usually a perk the app can’t match.
The National Restaurant Association’s 2025 off-premises report says more than 60% of takeout and delivery customers say a membership program affects where they order, and nearly 9 in 10 would use limited-time, app-only offers, per the National Restaurant Association’s 2025 off-premises report. The same report puts off-premises at nearly 75% of all restaurant traffic.
A loyalty design that works for most independents:
- Points on direct orders only. Spend $100, get $10 off. Simple enough to explain at the register in one sentence.
- A sign-up reward the first time someone orders direct (a free side, a dessert, a drink).
- A birthday reward, which earns you the birthday field and a reason to text once a year.
- A “we miss you” offer at 30 or 45 days since the last order, sent automatically.
Toast Loyalty, Square Loyalty, ChowNow’s marketing tools and Olo’s loyalty product (Olo bought Spendgo in late 2025) all handle this. Pick the one your ordering platform already talks to.
Text messages move direct orders faster than email, and they need care. Get marketing consent separately from order updates, keep the record, and send fewer messages than you think you should (we like two or three a month at most). Texas brought marketing texts under its telephone solicitation law with SB 140, effective September 1, 2025, with some exemptions, per Morgan Lewis’s summary of SB 140. Our Texas SMS rules post walks through the consent setup.
A sample text that works: “Gino’s: Friday pickup is open till 9. Order direct and your 4th order this month is on us. [link] Reply STOP to opt out.”
Packaging inserts: the bag is your best ad
Every delivery bag is a direct mail piece that lands in the hands of someone who just bought from you. It costs you a few cents to print. Most restaurants put nothing in it.
The NRA report found 90% of consumers would order a wider variety of items if food held its quality in delivery, and more than half would pay more for packaging that protects it. So the bag carries two messages: the food travels well, and next time order from us.
Keep the card small and the ask single. One QR code, one offer, one sentence. Use a unique tracked link (a UTM-tagged URL or a short link for each app) so you can tell whether DoorDash bags or Uber bags convert better. Pickup is the easiest ask, because the guest saves the delivery fee and you save the commission.
Read your marketplace agreement before you start. Platform terms on customer data and in-bag promotions change, and you want to know where the lines are before you print 5,000 cards.
Ads that point to direct ordering
Paid ads for ordering are cheap to run and easy to get wrong. The goal is narrow: catch people who already want your food and send them to your link, not the app.
Google brand search. In accounts we’ve looked at, delivery apps and aggregators sometimes show up on a restaurant’s own name. A small brand campaign on terms like “[name] menu,” “[name] order online,” “[name] delivery” and “[name] Spring TX,” sending clicks to your ordering page, usually costs very little and puts your link above theirs. Add negatives like jobs, hiring, careers, application, nutrition and calories.
Meta to people who know you. Upload your opted-in customer list as a Custom Audience and retarget website visitors, then run a simple offer to them: the direct-order reward, a new item, a rainy-day pickup special. Keep a broad radius campaign for discovery small. The apps already do discovery, and you’re paying them for it.
Sample ad copy for Meta:
- Primary text: “Same tacos. Lower prices when you order direct. Pickup in 15 minutes on Woodlands Parkway.”
- Headline: “Order pickup direct and save”
- Button: Order Now, linked to your ordering page with a UTM tag.
An illustrative starting budget for a single location: $300 to $600 a month on brand search and $500 to $1,500 on Meta retargeting and customer-list ads. It’s an example to adjust, not a benchmark. Track direct orders as the conversion in both platforms, which our conversion tracking guide covers. If you want a team to run it, that’s our paid social and search marketing work.
Measure channel mix like a P&L line
If you only look at total sales, you’ll never see the shift. Put these on one sheet and review it every Monday with the same weight as food and labor.
| Metric | Formula | Why it matters |
|---|---|---|
| Direct share | Direct off-premises orders ÷ all off-premises orders | The number you’re trying to move |
| Effective commission rate | All marketplace fees ÷ marketplace sales | Shows what your tier really costs once promos and fees are in |
| Contribution per order, by channel | Order subtotal − fees − packaging − food cost | Some app orders lose money; this shows which |
| Direct repeat rate | Direct guests with 2+ orders in 60 days ÷ direct guests | Whether loyalty is working |
| Cost per new direct guest | Direct-ordering ad and insert spend ÷ first-time direct guests | Whether your marketing pays back |
Illustrative example: a Katy café does 600 app delivery orders a month at a $38 average on a 25% tier, or $5,700 in commission. Move 120 of those regulars to direct pickup at about $1.41 in processing each, and it saves roughly $970 a month, close to $11,700 a year, before counting the loyalty cost. Your numbers will differ. The direction usually doesn’t.
Don’t judge it in a week. Direct share tends to climb slowly over two or three months as regulars try the link once, get their reward, and switch for good.
The first 90 days
If we were starting your direct ordering push on Monday:
- Days 1 to 30. Pull your delivery agreements and a month of orders by channel. Pick or fix your ordering platform. Set your own link as preferred in Business Profile food ordering. Put an Order pickup button at the top of the mobile site. Set in-store pricing on direct ordering.
- Days 31 to 60. Launch loyalty with a first-direct-order reward. Start bag inserts with tracked QR codes. Turn on SMS opt-in at checkout. Launch the brand search campaign.
- Days 61 to 90. Add Meta customer-list and retargeting ads. Review the channel mix sheet weekly. Revisit your marketplace tier with three months of real data.
None of this means turning off the apps. It means using them to find people and keeping the regulars yourself. If you want a senior strategist to look at your order mix and build a plan around it, get in touch.
Frequently asked questions
How much does DoorDash charge restaurants in 2026?
DoorDash’s published Marketplace plans charge 15% of the order subtotal on Basic, 25% on Plus and 30% on Premier for delivery, and 6% on pickup orders across all plans. There are no activation or subscription fees, and payment processing is included in the commission. Its commission-free Online Ordering product charges a processing fee instead. Check your own signed agreement, since that controls what you pay.
How much does Uber Eats charge restaurants?
As of March 2026, Uber Eats’ published U.S. delivery rates are 20% on Lite, 25% on Plus with Uber One member orders at 30%, and 30% on Premium. Pickup is 7% when you have validated in-store pricing and 10% when you don’t. Lite can be lower in some cities. Its Webshop ordering site charges 2.5% plus $0.29 per order instead of a marketplace commission.
Does Houston cap delivery app fees for restaurants?
No. Houston has no cap, and a 2021 bill to cap delivery fees in Texas at 15% died in committee. Texas instead passed SB 911, effective January 1, 2022, which requires delivery companies to have a written agreement that states each fee, lets restaurants demand removal within 10 days, and stops cities and counties from adopting rules that change those agreements.
How do I get the Order button on Google to use my own online ordering?
In your Google Business Profile, open Food ordering, choose your ordering provider or use Add a link at the bottom to add your ordering page, then select Set as preferred and turn on Preferred for pickup and Preferred for delivery. You can also hide third-party providers for pickup, delivery or both. Changes usually appear on Search and Maps within a few hours.
Should my online ordering prices match my in-store menu?
We usually recommend matching in-store prices on your own ordering page and keeping any markup on the delivery apps. That makes ordering direct visibly cheaper, which is the easiest message to put on bags and receipts. On Uber Eats, pickup orders cost 7% with validated in-store pricing and 10% without it, so parity on pickup menus also lowers your fees.
