Your phone order is commission-free. Most of them ring out anyway.

A phone order costs a restaurant nothing in commission. A delivery-app order costs 15–30% of it before any other fee lands. So the cheapest channel you have is also the one most likely to ring out during a rush, which pushes the caller who would have ordered directly onto the app that takes the biggest cut instead.

How much commission do DoorDash, Uber Eats and Grubhub actually take?

Between 15% and 30% per order, depending on the platform and the visibility tier a restaurant pays for. DoorDash's published tiers run 15% (Basic) to 30% (Premier); Uber Eats' standard marketplace rate is around 30%, with lower rates available for pickup-only listings; Grubhub runs 15–25% plus separate charges for marketing and promoted placement. On top of the headline commission, payment processing typically adds another 2.9–3.5%, so the effective cost on a delivered order commonly lands at 30–40% once everything is counted.

None of this is Hilda's data; it's what Toast, DoorDash and the platforms themselves publish, and it's consistent across independent write-ups of restaurant delivery economics. Check current rates directly with each platform before making a decision; they change tiers and terms.

Why does the commission-free channel get dropped first?

Because a delivery-app order lands on a tablet whether or not anyone is free to take it, and a phone order needs a person to pick up. Under pressure, the order that requires nobody wins by default, not because it's better for the restaurant, but because it's the only one that doesn't need staff time right now.

• An engaged line at peak. One phone, one person, a Friday rush: every caller who hears a busy tone or unanswered ring is a caller who orders somewhere else, often through an app.
• Orders taken wrong under pressure. A rushed phone order means wrong modifiers, missed allergens and remakes that cost twice, which makes staff even more reluctant to pick up next time.
• No one assigned to the phone at all. Marketplace orders get a dedicated tablet by design. Phone orders usually don't get an equivalent; whoever is nearest picks up, or nobody does.

What does a missed phone order actually cost, compared to the same order through an app?

Take a $30 order as a simple illustration, not an average; every kitchen's real ticket size differs. Taken by phone, the restaurant keeps the full $30 minus normal cost of goods. Routed through a delivery app at a 25% commission plus card processing, the same $30 order nets closer to $21–22 before the food is even made. The margin difference doesn't shrink on that order; for many independent restaurants operating on thin single-digit net margins already, a quarter of the ticket gone in commission is close to the entire profit on it. Our ROI calculator works this out against your own missed-call numbers rather than an illustrative one.

How does answering every call keep more orders on the free channel?

The mechanism is simple: nothing rings out. Hilda answers unlimited concurrent calls, so a Friday-night rush never produces a busy tone, reads the venue's live menu so it quotes what's actually available and stops offering items marked out of stock, and pushes completed orders into Toast or Clover as tickets rather than a message someone has to re-type. See the full picture for takeaway and delivery. None of that recovers a caller who has already given up and opened an app; it just means fewer of them get to that point in the first place.

It doesn't take card payments over the phone, and it won't confirm a dish is safe for a severe allergy; both stay with a person, every time. The goal isn't replacing a phone order with something automated; it's making sure the phone order still happens instead of becoming a delivery-app order by default.

If you want to know whether this is costing you money

1. Pull last month's delivery-app statement and note the total commission paid; the number is usually on the summary page, not something you have to calculate.
2. Separately, count how many calls to your main line went unanswered during your three busiest service hours last week. Most POS or phone systems log this even if nobody's looked at it.
3. Multiply unanswered calls by your average phone-order value. That's a rough floor on what's leaking to voicemail, before any of it even reaches an app.
4. Compare that number to the commission line from step one; for a lot of venues, the two are closer than expected.

Not every unanswered call would have been a direct order; some callers just hang up, some ring the place next door regardless. But the two numbers above are usually close enough to be worth five minutes with last month's statements before assuming the fix is a cheaper delivery-app plan rather than a phone that gets picked up.