This post in 30 seconds.

• Five inputs decide your real phone support cost, and most brands get the loaded rep cost wrong by half. We walk all five, plus the side of the ledger nobody calculates: the revenue you lose on the calls you drop.
• When I added up the true loaded phone cost for a real $30M Shopify brand, it came out to roughly 2.4x what their finance line said.
• Built for founders, COOs, and Heads of CX at $10M-$100M Shopify brands running a paid helpdesk and a visible phone number.

Most brands quote their phone support cost at about half the real number. They look at rep salaries, maybe the helpdesk seat fee, divide by tickets, and call it a day. Then the CFO asks "what's the support headcount line next quarter" and there isn't a clean answer, because the number they've been carrying around was never the real one.

A study from Tymely found one ecommerce company that pegged its cost per ticket at under $1. Once they added training, turnover, management overhead, and software, the real number was over $3. Phone is worse than the blended average, not better. So this is a calculator you can actually run with your own numbers, both sides of the ledger: what the calls you answer cost, and what the calls you drop cost you in lost orders.

If you run customer experience at a Shopify brand doing $10M-$100M, you already know the after-hours queue and the same questions over and over. The goal here is to put a real per-call number in your hand, the kind you can say out loud on a call without flinching. If you'd rather we just pull your last 7 days and do it with you, book a 30-min call and we'll run the math live.

How I built this calculator

I'm Ruben, co-founder of Ringly.io. We run AI phone support for 50+ Shopify brands, which means I look at real support cost data every week, not list-price math off a vendor page.

The numbers below come from two places. The per-call costs are pulled from live Ringly customer billing across those 50+ brands. The human-rep costs lean on the loaded-cost baselines everyone in this space uses: roughly $4,000/month all-in for a US rep, plus the $14,113 it takes to replace one when they quit (industry turnover runs 31.2%).

What made me build this was one afternoon with a spreadsheet. I added up the true loaded phone-support cost for a real $30M Shopify brand we work with, and it came out to roughly 2.4x what their finance line said. Not because anyone lied. The finance line counts salaries and software, and the real cost counts everything that keeps the phone answered.

A phone support cost calculator that's worth running has to include all five of these:

• Loaded cost per rep, not the salary line. Salary, benefits, payroll tax, training ramp, and the cost of replacing the ones who quit.
• Calls per day, the real number, not your guess. Most operators guess low by 2-3x.
• Minutes per call, because phone is the expensive channel and the math is sensitive to it.
• After-hours and weekend coverage, or the lack of it, which is where most of the leak is.
• Missed-call revenue loss, the half nobody calculates: what the dropped calls cost you in orders that went to a competitor.

A cost calculator that only counts the calls you answer is missing the bigger number. Run all five and your real cost usually lands somewhere between 1.5x and 3x the figure you've been quoting.

The 5 inputs that actually decide your cost

Get these five right and the formula does the rest. Get the loaded cost wrong, which most brands do, and every downstream number is off by half. Let's walk them in order.

1. Loaded cost per rep (not the salary line)

The number finance gives you is usually the salary. The number you should use is the loaded cost: salary, benefits, payroll tax, software seat, plus the parts that don't show up on a paycheck.

A US in-house rep runs about $4,000/month loaded. Offshore lands closer to $2,000. But the loaded figure also has to carry attrition. With turnover at 31.2% and 69-73% of new reps quitting within their first year, you're constantly paying to rehire and retrain. It takes 6-8 months for a rep to get fully productive, and the Department of Labor puts the cost of one wrong hire as high as $240,000.

So when you build your loaded cost per rep, start at $4K and add a slice for the rehire-and-ramp cycle. Most brands that do this honestly land at $4,500-$5,500 per rep per month.

2. Calls per day (and what you're guessing wrong)

Operators guess "300 to 500 calls a day." The real number across the brands we onboard is usually 800 to 2,000. People underestimate their own volume because voicemails and after-hours calls never make it into the ticket count.

Volume also scales with order value. At a $250 average order value, 12-18% of orders generate a phone call, versus around 3% at a $40 AOV. Higher-ticket buyers want to talk to a person before and after they buy. If your AOV is high, your call volume is higher than you think, and so is your cost.

3. Minutes per call

Phone is the expensive channel. The blended ecommerce cost per contact sits around $2.70 to $5.60 across all channels, but phone-only runs $8 to $17 per call. Average handle time for ecommerce phone is 3 to 6 minutes, and longer calls cost more, obviously.

There's a trap here worth naming: cost per contact is not cost per resolution. If a question takes 2.3 touches to resolve on average, your real cost per resolved issue is 2.3x your cost per contact. Phone's saving grace is that it usually resolves in one call, which is why the per-call number looks scary but the per-resolution number is more competitive than email.

4. After-hours and weekend coverage

This input is where the leak starts. Businesses answer only 37.8% of inbound calls; another 37.8% go to voicemail and 24.3% get nothing at all. On hold, 60% of callers hang up within 60 seconds.

If you don't staff nights and weekends, you're not saving money. You're moving the cost from your payroll line to your revenue line, where it's harder to see. Staffing a night shift doesn't pencil out either, because the volume isn't steady enough to keep a rep busy. That's the squeeze: too expensive to staff, too costly to ignore.

In the calculator, treat after-hours two ways. If you do staff it, the loaded cost of those weekend and evening reps goes on the cost side, and they're usually idle 50-60% of the shift, so the per-call cost on those hours is brutal. If you don't staff it, the dropped calls go on the revenue side under input five. Either way the after-hours block costs you. The only question is which line of the P&L it shows up on.

5. The missed-call revenue loss (the half nobody calculates)

This is the input every other calculator skips. The calls you drop are not free. 85% of callers who hit voicemail never call back, and 62% switch to a competitor. 80% of voicemail-routed callers hang up without leaving a message.

So multiply your dropped calls per month by the share of them that would have placed or saved an order, times your AOV. That number belongs in your cost calculation as surely as rep salaries do. It's just sitting on the revenue side of the P&L instead of the cost side, which is exactly why it goes uncounted.

The phone support cost formula

Your true phone support cost is two numbers added together: the cost of the calls you answer, and the revenue you lose on the calls you drop. Most calculators only give you the first.

Side one, the cost of answered calls:

Answered-call cost = (reps × loaded cost per rep) + software + management allocation

Then divide by calls handled to get your real cost per call.

Side two, the revenue lost on dropped calls:

Dropped-call revenue loss = dropped calls/month × % that would have ordered or stayed × average order value

Add the two and you have the real picture. A brand can have a "reasonable" cost per answered call and still be hemorrhaging on the second line because half its calls never get picked up. That second number is usually the one that changes the decision.

The three cost numbers operators mix up

Cost per contact, cost per resolution, and cost per order are three different numbers, and quoting the wrong one is how brands talk themselves into the wrong staffing plan. They sound interchangeable on a call. They aren't.

Cost per contact is the simplest: total support cost divided by the number of interactions. It's the number most calculators stop at, and it flatters phone because a single call looks cheap next to four back-and-forth emails.

Cost per resolution is the honest one. If it takes 2.3 touches on average to actually close an issue, your cost per resolution is 2.3x your cost per contact. Email looks cheap at $8 a contact until you count that it takes three or four messages to resolve, which pushes the real number to $24 to $60 per resolved issue. Phone usually closes in one call, so its cost per resolution is more competitive than its scary-looking cost per contact suggests.

Cost per order is the one your CFO actually cares about, because it ties support spend to revenue. Divide your total monthly support cost by orders shipped and you get a number you can defend in a budget meeting. At a $250 AOV, a few dollars of support cost per order is fine. At a $40 AOV, the same support cost can quietly eat your margin.

Use cost per resolution when you're comparing channels, and cost per order when you're defending the line to finance. Don't let a low cost-per-contact number convince you phone is cheaper than it is, and don't let a high one convince you to drop the channel that high-AOV buyers actually want.

A worked example: a $30M Shopify brand

Take a real-shaped example. A $30M Shopify brand running 6 reps, $250 AOV, roughly 1,200 calls a month, with the phone going to voicemail after 6pm and on weekends.

First, the loaded-cost buildup, finance line versus reality:

That's the 2.4x in action. The finance line says $19,500. The real loaded cost is closer to $31,500, which on 1,200 answered calls works out to about $26 per answered call once you carry everything.

Now side two. Say 30% of inbound goes to voicemail after-hours, so roughly 360 calls a month never get answered. If even 15% of those would have placed or saved an order at a $250 AOV, that's 54 lost orders, about $13,500 a month walking out the door. Annualize it and the dropped-call line alone is $162,000.

For comparison, WashCo, a Shopify brand we launched, runs at $0.91 per call. Same job, different cost structure, and the after-hours calls get answered instead of dropped. You can read the full ecommerce cost-per-contact benchmarks if you want to sanity-check your own number against the field.

Fill in your own numbers

Here's the calculator. Copy this table, drop in your numbers, and you'll have your real two-sided cost in about ten minutes.

Then run the two formulas:

Cost per answered call = (reps × loaded cost + software + management) ÷ calls answered

Monthly revenue loss = dropped calls × % that would've ordered × AOV

Read your number like this. If your true loaded cost is north of $15,000/month and more than 40% of your calls are the same repeatable questions (order status, returns, the same five things), the math says automate the routine and keep your humans on the calls that actually need them. If you want a second opinion on the inputs, book a 30-min call and we'll pull your last 7 days and fill the table in with you.

While you're sizing spend, the AOV calculator and ROAS calculator feed the same inputs, and the WISMO calls breakdown shows you which calls are the cheap ones to take off your team first.

What this costs you today vs with Ringly

Once you have your real number, the comparison gets simple. Take that same 6-rep team:

That's roughly 70% of repeatable calls (order status, returns, product questions) routed to the AI. The other 30%, the genuinely complex calls, still go to your team, who now have time to actually solve them. And the after-hours calls get answered instead of dropped, so the revenue-loss line shrinks too.

Per call, the gap is just as clear:

Ringly.io is AI phone support for Shopify brands. Instead of growing your support headcount every time call volume goes up, the AI handles inbound calls 24/7: order status, returns, product questions, abandoned cart rescue. Across 50+ brands, the AI resolves 73% of calls autonomously at roughly $0.42 per resolved call. Calls that need a human escalate cleanly to Gorgias, Richpanel, Reamaze, or whatever helpdesk you already run. Plans start at $349/mo with a 65% resolution guarantee.

"My customers also feel like it's a normal person. They feel like they can communicate if they have questions."
Claudia Droge, TechCraft Studio

If you'd rather not run the table yourself, book a 30-min call and we'll compare your real per-call number to ours, live. For the human-only path, our breakdown of in-house vs outsourced support and the BPO-for-Shopify guide cover the trade-offs, and 24/7 ecommerce phone support digs into the after-hours math.

Frequently asked questions

How do I calculate my phone support cost per call?

Add up everything: loaded rep cost (salary, benefits, tax, training, attrition), helpdesk and phone software, and a slice of management time. Then divide by the number of calls actually handled, not tickets. Most brands land at $8 to $17 per call once they count honestly.

What's a good cost per contact for ecommerce phone support?

The blended ecommerce average across all channels is $2.70 to $5.60 per contact. Phone runs higher, $8 to $17 per call, because it's the live, high-intent channel. A "good" phone number is one where the calls resolve in a single touch, which keeps your cost per resolution competitive even when cost per contact looks high.

Why is my real customer support cost higher than I think?

Because finance counts salaries and software, and the real cost counts everything that keeps the phone answered. One study found a brand's true cost was 3x its estimate once training, turnover, and overhead went in. Add the revenue you lose on dropped calls and the gap gets wider.

How much does an answering service cost vs in-house reps?

Per-call AI answering runs roughly $1 to $5 per call, versus $5 to $12 for live human operators. In-house reps cost about $4,000/month loaded each. The cheapest path depends on volume: low volume favors per-call pricing, steady high volume favors a fixed monthly tool.

Does missed-call revenue loss really count as a support cost?

Yes. 62% of callers who can't reach you go to a competitor, and 85% who hit voicemail never call back. That lost revenue is a direct result of your support coverage, so it belongs in the calculation even though it sits on the revenue line.

How much does Ringly cost?

Grow is $349/month (1,000 minutes), Pro is $799/month (2,500 minutes), and Enterprise is custom for larger brands. Every plan carries a 65% resolution guarantee: if the AI resolves under 65% of your calls in 90 days, we refund the last 3 months.

How fast can I see my real numbers?

On a 30-min call we pull your last 7 days of calls and fill the calculator in with you. No homework on your end. You leave the call with a real cost per call and a real revenue-loss number, whether or not you ever use Ringly.

Talk to us

If you run a $10M-$100M Shopify brand and you can't say your real per-call number out loud, a 30-min call is the fastest way to get it. We pull your last 7 days, fill in both sides of the calculator, and show you what the after-hours queue is actually costing you.

The 3-layer guarantee.

1. Live in 14 days or it's free until launched.
2. 65% resolution in 90 days or we refund the last 3 months of subscription fees.
3. We keep working free until we hit it.

Ruben (Ringly co-founder) takes these calls personally.

Book a 30-min call →