Maple Review: Price, Fit, and Limits
Maple answers reservations through OpenTable and takes basic orders at a quote price. What it does well, what a quote hides, and which shop should not buy it.
On this page
The phone rings at 6:40 on a Thursday, while the host is walking two parties to the back. The caller wants Friday at 7:30 for four, then adds a margherita for pickup on the way in. One call, two systems, and only one of them is the reason the shop is shopping.
Maple is built for the first half of that call. It is a reservation-aware restaurant phone agent: it answers, works the table request through OpenTable, and takes a basic order. The vendor publishes a 92% resolution rate for standard call types and coverage across 1,000+ locations. Both are its own figures, not an audit.
The question this review keeps coming back to is not whether 92% is a good number. It is which of those two requests pays your rent.
TL;DR
- Maple is a reservation-aware phone agent with OpenTable at its center, a published 92% resolution rate, and coverage across 1,000+ locations, all vendor figures.
- It sells on a quote, so the first job is one written monthly number at your call volume, with setup, per-location, and term lines included.
- Order handling is described as basic and native in-call payment is not advertised, which leaves a card order with a second step somewhere else.
- For a room where callers want tables, that pairing is well matched. For a takeout counter, it is the wrong half of the product.
- One test call with two modifiers, a substitution, and a card request tells you more than any published percentage.
What Maple is, and the calls it was built for
Maple is a phone agent aimed at independent restaurants, sold on a quote, with the OpenTable integration as its lead feature (Maple's site). It answers every call, resolves standard call types, and handles orders at a basic level.
The calls it is built for are the ones a host stand fields all night: hours, parking, party size, dress code, a table at eight. Those are the questions that interrupt a service, and the ones a reservation-aware agent is good at absorbing.
Two numbers deserve plain treatment. The 92% resolution rate is the vendor's, and resolution normally means the call did not need a person. The 1,000+ locations is a coverage claim, also the vendor's. Neither number tells you whether a large half-mushroom half-pepperoni arrived in the kitchen as written, which is the number a takeout shop actually lives on.
What Maple does better than an order-first product
The table. In a 60-seat room on a Friday, the reservation is the revenue. A caller asking for 7:30 gets an answer and, per the vendor, a record on the book the host already reads. This is its home ground.
Publishing resolution instead of accuracy. Accuracy figures in this category are usually measured against a vendor's own corpus. Resolution at least names the thing an operator cares about: how often the phone needed a human to finish. That is a more honest headline than a percentage nobody can define.
The installed base. Coverage across 1,000+ locations, if it holds, means the vendor has met your edge case: the party of eleven that calls twice, the regular who wants the banquette, the caller asking about bus parking.
The shape of the buyer. Independents with a reservation book and moderate takeout are the right customer, and the feature set agrees.
Where Maple runs out of road
The price is a conversation. Maple is quote-based, checked October 2026. You cannot put it beside $199 or $99 until a written monthly figure exists at your call volume. A quote is not a fault on its own, but it moves the comparison work onto you, and the number that arrives is rarely the number in the first email.
Orders are the second job. Basic order handling covers a bar pie and a salad. It is thin for a pie that opens two modifier groups per half, swaps a side, and requires a choice of dressing. Ask what the kitchen ticket looks like before you believe the call is closed.
Payment is not native. Secure in-call payment is not advertised on the public pages. A card order therefore finishes somewhere else, at the counter or through a link, and every extra step is a place a hungry caller hangs up.
Integration depth is the unknown. OpenTable is documented, and OpenTable keeps its own list of voice AI partners for its reservation products. How far an order travels toward the kitchen system, and whether it prints as a ticket or lands as a message, is the question a demo has to answer.
What the quote has to contain
A quote is only useful itemised. Take this table into the call and fill every cell, or you are comparing a four-line invoice to a one-line one.
| Line | What to ask for | Why it changes the answer |
|---|---|---|
| Base | Monthly subscription per location | The number a published plan gives you for free |
| Usage | Per-minute or per-call rate, and the cap | A three-minute call and a six-minute call cost the same only on a flat plan |
| Setup | Onboarding, menu build, number porting | One-time fees hide a cheap monthly price |
| Term | Commitment length, and the exit | A 12-month term makes a monthly number a yearly one |
| Payment | Where a card order finishes | Decides whether a prepaid pickup is possible at all |
| Support | Who answers at 8 on a Saturday | The line you are replacing is the one your staff answers |
Then run the volume. A single shop taking 300 calls a month at three minutes is about 900 connected minutes.
Published plans in this category sit at $99 for 750 minutes, $199 flat, $299 for 5,000 minutes, and $450–$600 per location, all as published in October 2026. A quote landing at $250 a month is not expensive or cheap in the abstract. It is a price per answered call, and that is the number to compare.
The cost of doing nothing is larger than any of them. Industry estimates put missed restaurant calls at roughly 150 a month, at $35 to $85 each, which is $5,250 to $12,750 of ticket value walking away. The missed-call arithmetic walks through it with your numbers rather than the average.
Pick the call that pays the rent
Which of the two calls at the top of this page is your revenue?
If it is the table, stop shopping and buy the reservation tool. Ask Maple for the monthly number, confirm the OpenTable path on your own account, and spend the remaining diligence on how the party-of-eleven call gets handled at 8:50 on a Saturday.
If it is the order, the resolution rate is not measuring your problem. Industry estimates put around 60% of missed restaurant calls into someone trying to order or book, and a decent share of those are orders. An agent that resolves a question beautifully and hands the order to a second system has moved the work, not removed it.
If it is both, pick on the revenue split, not the feature list. A restaurant where takeout is 70% of sales should not buy a product whose center of gravity is the book, and a 90-seat dining room that sells twelve pickup orders a night should not buy an order line and call it a reservation system. The comparison pages are the faster way to see what else exists before you commit.
Five questions for the Maple demo
- "Book me for Friday at 7:30 and show me the record." Then look at the book the host uses, not the vendor's dashboard.
- "Take this order with two modifiers and a substitution, then show me the kitchen ticket." The ticket is the product; the transcript is the demo.
- "The caller wants to pay now. What happens?" Listen for the number of steps between the card and the kitchen.
- "Define resolution, and show me a call that failed." A vendor who cannot produce a failure has not measured anything.
- "What is the all-in monthly number at 400 calls, in writing, and what is the term?" Ask for it in an email you can forward.
Buy Maple if, skip it if
Buy it if the reservation book is the business, OpenTable is already the system of record, and takeout is a side dish. A room where the phone is a host stand with a headset attached is exactly the buyer in the vendor's own description, and the product fits that shape.
Do not buy it for a counter where the caller says "two large, one half mushroom" and the table is never mentioned. Do not buy Maple if a card has to be charged before the caller hangs up, because that path is not advertised. And do not buy it if a written quote cannot be produced before the trial ends, because a quote-based price with no volume anchor is a number you will renegotiate from behind.
A reservation-led room that still wants a second opinion should read the Maple alternatives before the second demo. The shortlist is short, and the comparison is decided by the same question this review is built on.
Where Kovo fits, disclosed
We sell Kovo, so weigh this section with that in mind. Kovo answers one store's phone, takes the order the way the caller says it, reads it back, and prints the ticket to Toast, Square, Clover, Olo, Epson Server Direct Print or Star CloudPRNT. It charges a card the shop already holds while the caller is still on the line, links a first-time caller with the line held open, and answers a website chat on the same engine. It also books tables in its own book: availability from your hours and holiday exceptions, one seat per slot, a desk board, and a confirmation text. Starter is $99 a month for 750 connected minutes, Unlimited is $299 for 5,000, and the trial runs 14 days without a card.
The boundary sits where Maple's strength is. Kovo does not write into OpenTable or Resy. A shop on either one gets its reservation page read out and texted instead, and the partner cards say pending, because OpenTable is partner-gated and Resy publishes no booking API. So a record on the host's OpenTable screen is still a reason to buy Maple. The order limit is the shop's own — defaults of 10 items in total and 8 of one item, plus an optional value limit, and zero switches any of them off — and above one the call rings your handset if you asked for that, otherwise the agent says the order is too large and keeps taking what fits.
FAQ
How much does Maple cost?
Maple is sold on a quote, so there is no published monthly figure to compare as of October 2026. That makes one step non-negotiable: get the all-in monthly number at your own call volume in writing, including any per-location line, setup fee, and contract term. Then hold it next to the published plans in the category, which run from $99 to $600 a month.
Does Maple take a card during the call?
Native in-call payment is not advertised on Maple's public pages, which means a card order probably finishes somewhere else, either at the counter or through a link. Ask on the demo and watch where the money lands. If a prepaid pickup order is part of your Friday, that single answer decides whether the phone is an order line or a receptionist, and it is worth more of the call than the 92% resolution rate.
What does a 92% resolution rate actually measure?
It is a vendor figure for standard call types, and it is not independently audited. Resolution usually means the call finished without a person, not that the ticket was correct. Two vendors can both publish 92% and disagree completely about whether a missing modifier counts against it, so ask for the definition and for a sample of calls that failed.
Can Maple put an order into my POS?
The OpenTable path is the documented one, and order handling is described by the vendor as basic rather than as a full order-closure product. Before signing, ask which kitchen system receives a ticket, what a modifier looks like when it arrives, and whether the ticket prints or waits for a person to re-key it. A 92% resolution rate says nothing about any of that, and the menu sync guide lists the checks worth running.
Is Maple a good fit for a takeout-heavy restaurant?
Usually not. Maple is built around the reservation book, so a shop whose calls are mostly pizza, sandwiches, and pickup is buying the half of the product it uses least. A shop taking 300 order calls a month should test a ticket-first product against Maple on the same menu, then compare what each one prints to the kitchen.
Final Thoughts
Maple's decision comes down to a single sentence: buy it if the caller asking for 7:30 on Friday is the caller who pays you, and buy something narrower if the caller ordering two pizzas is.
Run the quote table with a real monthly figure, then take one test call with the order your kitchen gets wrong most often and ask to see the ticket. Take a test call with your own menu, hear that order read back before the quote lands, and hold the plans next to the number you were given.