Key takeaways
- Online orders remove the most common phone error, mishearing, but add their own: wrong modifiers, outdated menus and missed special requests.
- A phone order ties up a staff member for the whole call, so price that time using real wage data before you decide which channel to push.
- Measure your own error rate by channel for two weeks before changing anything; national surveys do not tell you how your line performs.
- Keep the phone for guests who want it, including many older guests, and make online ordering the easy default for everyone else[1][2].
The short answer
Online orders tend to win on accuracy for simple items and on staff time, because the guest does the typing. Phone orders tend to win on complicated requests, large or group orders and guests who want reassurance from a person. The useful question is not which channel is better in general, but what each one costs and how often each one goes wrong in your restaurant.
Most guests are now comfortable ordering on a screen. 57% of adults have recently used mobile ordering, including 74% of millennials and 65% of Gen Z adults[1], and three-quarters of delivery customers value tech-enabled ordering and payment[3].
Side-by-side: where each channel goes wrong
| Factor | Phone order | Online order (your site or app) |
|---|---|---|
| Who enters the order | Staff member, while listening | Guest, on their own screen |
| Most common error | Misheard item, size or modifier | Wrong modifier choice, missed note, item that should have been 86'd |
| Staff time per order | Whole length of the call, plus any callback | Near zero until the ticket prints |
| Payment | Often paid at pickup, so walkouts and no-shows happen | Usually paid in advance |
| Upsell | Depends on the person answering | Consistent prompts for sides and drinks |
| Special requests | Easy to discuss and clarify | Limited to the note field and modifiers you offer |
| Busy-night risk | Calls go unanswered or rushed | Orders keep arriving unless throttled |
| Guest who prefers it | Older guests, group orderers, people with questions | Younger guests, repeat orderers, people ordering late |
Neither column is free of errors. Online menus fail when an item sold out at 6 pm is still orderable at 8, or when a modifier list is confusing and guests pick the wrong one. Phone orders fail when a noisy room or a rushed host turns “no onions” into onions. Your fixes are different for each.
Measure your own accuracy for two weeks
National surveys tell you what guests prefer. They do not tell you how often your line gets a phone order wrong. You can find out in two weeks with no new software.
Tag every order by channel
Make sure the POS or ticket shows where the order came from: phone, own site, each marketplace app.
Log every error at handoff and after
Keep a sheet at the expo station. Log missing items, wrong items, wrong modifiers and wrong times, plus any complaint or refund that comes in later.
Note the cause
Was it entry (taken down wrong), kitchen (made wrong), packing (left out) or menu (item not available)? This tells you whether the channel or the line is at fault.
Count orders by channel
Divide errors by orders for each channel. An error rate per 100 orders is easy to compare.
Time a sample of phone calls
Pick twenty calls across lunch and dinner and time them from pickup to hang-up.
If phone orders show many entry errors, the fix is a better call script and readback. If online orders show many menu errors, the fix is keeping the online menu and 86 list in sync with the POS.
What a phone order costs in staff time
Phone orders use paid time that online orders do not. To put a number on it, use the wage of whoever answers. The national median hourly wage was $15.00 for fast food and counter workers[4] and $15.00 for hosts and hostesses[5] in May 2025. Your own rate may be higher, especially in states and cities with higher minimum wages.
Example: a fictional counter-service taco shop timed twenty calls and found an average of three and a half minutes per phone order, including a quick callback on about one call in five. At a $15.00 hourly wage[4], each phone order uses about 88 cents of paid time before any error. At 40 phone orders a day, that is about 2.3 hours of staff time spent on the phone every day.
| Average call length | At $15 per hour | At $18 per hour | At $22 per hour |
|---|---|---|---|
| 2 minutes | 50 cents | 60 cents | 73 cents |
| 3.5 minutes | 88 cents | $1.05 | $1.28 |
| 5 minutes | $1.25 | $1.50 | $1.83 |
That is not an argument to stop taking calls. It is a reason to keep calls short, make the online option easy, and make sure the person on the phone is not also running the register during a rush.
Who still calls, and why that matters
The phone is not only for people who cannot use an app. Many callers want to ask a question first: is the fish fresh today, can you make it without dairy, how long is the wait. In reservation research, about two-thirds of diners who had never booked online said they wanted to talk to someone[6]. Similar reasons apply to takeout.
Smartphone ownership also drops with age: 97% of US adults aged 18 to 49 own one, compared with 79% of those 65 and older[2]. A restaurant with many older regulars should keep a strong phone process even as it grows online orders.
Decision criteria: when to push guests online
- Push online when most calls are simple repeat orders from regulars, when phone errors are mostly mishearing, and when the person answering is also needed on the floor.
- Keep the phone strong when many calls are questions before ordering, large or group orders, allergy discussions or older regulars.
- Offer both on the call: answer questions, then offer to text the ordering link if the guest prefers to order on their phone. An SI Employee such as Ava can do that step for calls you miss, sending your official ordering link when the caller agrees to a text.
- Do not move phone guests to a marketplace app by accident. Give them your own ordering link so the order stays direct. The comparison of first-party and third-party online ordering explains why that matters.
Ticket size, payment and no-show risk
Accuracy and staff time are only part of the comparison. Two more differences show up in the numbers once you split reports by channel.
The first is ticket size. Online menus show every side, drink and dessert, and they prompt for add-ons the same way on every order. A person on the phone during a rush rarely has time to suggest anything. Pull average order value by channel from your POS for the same two weeks you track errors. If online tickets run higher, part of that is the prompt, and part is that guests ordering online take their time.
The second is payment. Phone orders are often paid at pickup, which means a share of them are never collected. Online orders are usually paid when placed. Log unclaimed phone orders for the same period. Even a few a week add up to food cost, packaging and line time with no sale behind them.
- If unclaimed phone orders are a problem, take a card on the call for larger orders, or text a payment link before the kitchen fires the order.
- If online tickets are much higher, put the same add-on prompts into the phone script, worded as a question: “Would you like to add a side of rice or a drink?”
- If phone tickets are higher, look at who is calling. Large family and group orders often come by phone because the guest wants to check quantities with someone.
Moving phone regulars online without losing them
Some regulars call every Friday with the same order. They are the easiest guests to move online and the easiest to upset if you do it badly. Treat the change as a convenience you offer, not a rule you enforce.
Offer, do not insist
At the end of the call, say: “Next time, you can order this online in under a minute and pick your time. Would you like me to text you the link?”
Make reordering easy
Check that your ordering site saves past orders or lets guests reorder in one tap once they create an account.
Keep the phone answered
A regular who calls and gets no answer may order somewhere else. Moving guests online should never mean letting the phone ring out.
Watch the results
After a month, compare how many regulars order online, how many still call, and whether their order frequency changed.
A phone order script that prevents most errors
- Greet and get the name and callback number first, so you can call back if the line drops.
- Ask pickup or delivery, and for delivery confirm the address and zone before taking items.
- Enter each item into the POS as the guest says it, not on paper first.
- Ask about allergies once, plainly, and flag any answer on the ticket.
- Read back every item with its modifiers, then the total and the pickup time.
- Close with the name of the restaurant and the time: “See you at 6:40, Maria.”
Pair the script with a quick bag check at expo, because many errors that guests blame on the phone call happen during packing.
Frequently asked questions
Are online orders more accurate than phone orders?
Usually, for the basics, because the guest picks items and modifiers themselves. But online orders can still go wrong through outdated menus, unclear modifier choices or notes the kitchen misses. Track errors by channel to see which is better in your store.
How long does a phone order take?
It depends on the order and the menu. Time a sample of calls from pickup to hang-up for two weeks; that number, multiplied by the wage of the person answering, is your staff cost per phone order.
Should I stop taking phone orders?
Few restaurants should stop completely. Some guests, including many older guests and people ordering for groups, prefer to talk to someone. Make online ordering easier and offer it on the call, but keep a clean phone process.
How do I reduce phone order mistakes?
Use a set order of questions, enter items straight into the POS while the guest talks, read back every modifier and the pickup time, and confirm the callback number before you hang up.
References
National Restaurant Association (2025). From Trend to Transformation: Off-Premises Dining Now Essential for Restaurant Consumers, Operators (opens in a new tab) ↑
Pew Research Center (2026). Americans’ Use of Mobile Devices in 2026 (Fact Sheet) (opens in a new tab) ↑
National Restaurant Association (2025). From Trend to Transformation: Off-Premises Dining Now Essential for Restaurant Consumers, Operators (opens in a new tab) ↑
US Bureau of Labor Statistics (2025). Occupational Employment and Wage Statistics, May 2025 national estimate: Fast food and counter workers median hourly wage (series OEUN000000000000035302308) (opens in a new tab) ↑
US Bureau of Labor Statistics (2025). Occupational Employment and Wage Statistics, May 2025 national estimate: Hosts and hostesses (restaurant, lounge and coffee shop) median hourly wage (series OEUN000000000000035903108) (opens in a new tab) ↑
Cornell University Center for Hospitality Research (Kimes and Kies) (2012). The Role of Multi-Restaurant Reservation Sites in Restaurant Distribution Management (Cornell Hospitality Report, 2012) (opens in a new tab) ↑