If most of your volume comes through Snackpass, the phone is easy to treat as a legacy channel. That is sometimes correct and sometimes an expensive assumption, and the way to tell is not a vendor pitch. It is a week of your own call logs.
Once you know the phone is worth answering, the only question that matters about a voice agent is whether the order lands in Snackpass correctly. An agent that sounds warm and drops a modifier costs you a remake and a customer, which is worse than the call you missed.
First, find out whether you are actually missing calls
Pull the log from your carrier or phone system for one busy week. Count the calls that rang out, went to voicemail, or were abandoned while on hold. Multiply by your average ticket.
That number is the entire business case, and it is often either obviously large or obviously small. A counter doing heavy app volume with eight missed calls a week should not be buying a voice agent. A shop with thirty unanswered calls on a Friday and Saturday is losing real money to a phone nobody has time to pick up. Do this arithmetic before you talk to anyone, because it is the one calculation a vendor cannot do for you honestly.
How the order gets from the call into Snackpass
X1 Voice reaches Snackpass through Deliverect, the middleware layer many restaurants already run to bring outside orders into their POS. The agent reads your live menu with its modifiers and prices, holds the conversation against that data, and writes the finished order back as a ticket.
The round trip is the whole point. A voice model that converses well and hands your staff a transcript to key in during a rush has moved the work, not removed it, which is the argument in why POS integration depth matters more than voice quality. If you have not set up the Deliverect side before, the Deliverect setup guide covers what that connection involves.
Loyalty is the question worth pressing on
If you chose Snackpass, rewards are probably central to how you run. So the phone raises a policy question before it raises a technical one.
Does a caller earn points? Can they redeem one? What does your counter say when someone arrives to pick up a phone order and asks why their reward did not apply? Rewards logic lives in the ordering platform, and how any phone channel touches it depends on your setup, so press a vendor for a demonstration rather than a yes. Then pick a policy simple enough that a nineteen-year-old at the register can explain it in one sentence. The tradeoffs are laid out in voice AI and loyalty programs.
Whatever you choose, say it on the call. A caller who is told up front that phone orders do not earn points is mildly disappointed. A caller who finds out at pickup is annoyed at your staff.
What a real rush looks like
At 12:20 on a weekday your line is out the door and the tablet is chirping. The phone rings four times in six minutes. Today, one of those gets answered by whoever is closest, badly, while a customer waits mid-transaction. Three do not get answered at all.
With an agent on the line, all four are picked up on the first or second ring. The orders write into Snackpass with pickup times. The caller asking whether you still have the special gets the answer you configured, or escalates to the counter as one interruption instead of three.
The staffing gain is real and smaller than the pitch. Calls per shift multiplied by your genuine average handle time is the number of counter minutes you get back, and that is the figure to bring to your own decision.
Decide the handoff before you go live
Complaints, catering inquiries, and anyone who asks twice for a person should reach a person, and the agent should give up gracefully rather than trying one more clarifying question. Write those rules down, then test them by calling in as an unhappy customer and confirming the escalation rings somewhere a human hears it at noon on a Friday. Human handoff and failover covers the rules worth setting and the ones operators forget.
The failure mode: menu drift
Everything matches on day one because you just imported it. Then it stops matching and nothing tells you.
A limited-time item goes in under a name no caller would say out loud. Something sells out at eight every Saturday. A price moves and the phone quote stays where it was. None of these raise an error. They arrive as a customer at your counter waiting for food the kitchen cannot make.
The fix is a rule. Whoever owns the Snackpass menu owns the phone menu, and every menu change earns a test call the same week. If you run daily specials, read real-time 86ing and menu sync and keep the phone menu conservative until you have timed your own sync lag with an actual call.
Payment, and the numbers to watch
Callers pay by card during the call for pickup and delivery, and the payment reconciles with your other tender types, so close does not gain a new manual step. Nobody on your staff hears a card number or writes one on a ticket that sits by the register.
Four numbers matter in month one. Answered-call rate moves first and is the reason you bought this. Order accuracy matters most, measured by pulling twenty phone tickets and checking them against transcripts rather than reading a dashboard. Escalation reasons become your work list once you sort them into handoffs that were correct by design and handoffs that were defects. Average phone ticket tells you whether the agent's prompts are doing anything.
Watch one more, given your mix: whether app volume moved. If phone orders went up and app orders went down by a similar amount, you did not add revenue, you added a channel, and you should judge the spend accordingly.
Setup, and the test that decides it
Connecting takes minutes and being ready takes a day. You link Snackpass through Deliverect, import the menu, set hours and phone-only rules, run test calls, then forward your number. Most single locations are live inside 24 hours, and our team will handle the menu pass. Plans start at $250 per month, and the current list of systems we reach is on the integrations page.
Then run the test. Call the demo line and place the worst order your kitchen sees: no onions, sub the fries, add a drink, apply the promo, change the size halfway through. Look at the ticket. If it is right, everything else is tuning. If the voice read it back perfectly and the ticket is wrong, you found out in five minutes instead of after a season of comps.