This post in 30 seconds.

• Gorgias AI Agent bills $0.90 per resolved conversation on annual plans, $1.00 month to month, and each resolution also eats one helpdesk ticket, so you pay twice for the same conversation.
• Above your bundled allotment, overages hit $1.50 per resolution, a 67% jump, which is why the monthly bill moves around so much at volume.
• Written for the Head of CX, COO, or founder at a $10M to $100M Shopify brand already paying for Gorgias and watching the AI line item climb.

The thing nobody tells you about per-resolution pricing is that the bill arrives before the resolution rate does. You turn the AI on, it starts closing tickets, and the invoice reflects every one of them at $0.90 or $1.00 a piece. The resolution rate you were sold ("up to 60%") shows up weeks later, usually lower than the number. Meanwhile the phone line in your footer is still ringing, and a human is still answering every call, and none of that touches the per-resolution model at all.

If you run customer experience at a $10M to $100M Shopify brand, you've watched this line item move month to month and wondered what actually drives it. This is the breakdown: what Gorgias AI Agent charges per resolution in 2025, why you get billed twice, how the cost compounds at volume, and how to forecast it before peak season catches you. If you want us to do the math against your real ticket and call numbers, book a 30-min call and we'll run it live.

How Gorgias AI Agent per-resolution pricing actually works

On May 28, 2025, Gorgias moved AI Agent to an outcome-based model. Instead of a flat AI add-on, you now pay per resolution. The headline number is $0.90 per resolved conversation on an annual plan, and $1.00 if you pay month to month, with starter pricing beginning at $1, according to Gorgias's own pricing breakdown.

So what counts as a resolution? Gorgias defines it narrowly, which works in your favor. A billable interaction is one where the AI handles a customer conversation entirely on its own and that conversation closes, inside a 72-hour window. The customer asks, the AI answers, the ticket closes, you pay.

What doesn't count is just as important. Spam and filtered messages are free. Emails that come in but never get a reply are free. And any conversation a human touches is free on the AI line (you still pay for it as a ticket, more on that next). That last rule matters: if your reps jump in on the hard tickets, those don't add to the AI bill. For the full plan-by-plan view, our Gorgias AI Agent overview walks through what the tool does on each channel.

Here's where accounts split. If you opened your Gorgias account before late May 2025, you may still be on legacy single-fee billing, per Gorgias's own documentation. New accounts are on per-resolution. Check which one you're on before you forecast anything, because the math is completely different.

The part that surprises people: you pay twice per resolution

Every AI resolution also counts as one billable helpdesk ticket. You pay the per-resolution fee, and you consume a ticket from your plan's allotment. Two charges, one conversation. Both eesel's per-resolution breakdown and myAskAI's cost analysis confirm the same structure.

The real unit cost of an AI resolution is the $0.90 fee plus the marginal ticket cost, not the $0.90 alone. On a Pro plan that ticket overage runs roughly $0.36 once you blow past your 2,000-ticket bundle, so a "$0.90" resolution is closer to $1.26 in the months you're over.

Here's the foundation layer you're paying before a single AI fee, on annual billing:

Tier prices and overage rates per eesel and myAskAI, cross-checked against the general Gorgias pricing page. When I forecast this for a real brand, I always model the ticket overage and the AI fee together, because separating them is how teams under-budget by a third.

What it costs at volume (and why the bill is a moving target)

Each plan bundles a set number of automated interactions, from about 90 on the smallest plans up to 2,500 or more on the larger ones. Resolve more than that and you're into overage territory at $1.50 per interaction, a 67% premium over the $0.90 base, per myAskAI. The bundle caps long before a busy month does, so most of your peak-season resolutions get billed at the higher rate.

eesel ran a clean example. A Pro plan during a Black Friday spike: $300 base, plus $72 in ticket overage for 200 extra tickets, plus $500 in AI fees for 500 resolutions, lands at $872 for the month. That's a 2.9x jump off the $300 base, driven entirely by volume you can't fully predict.

Scale it up and it gets steep fast. myAskAI models a brand at 50,000 tickets a month resolving 50% with AI: roughly $37,500 a month, with most of the 25,000 resolutions billed at the $1.50 overage rate, not the $0.90 base, because the bundle ran out at 2,500.

Now the part that quietly inflates every forecast: the resolution rate. Gorgias markets "up to 60%." Its own published case studies land between 24% and 56%, with brands like Shinesty at 54%, Psycho Bunny at 26%, and Obvi at 27%, per myAskAI's review of the case-study data. Forecast the bill at the rate you'll actually hit in the first month (25% to 30% is typical), not the rate on the sales deck. The bill is real before the resolution rate catches up.

How to forecast your Gorgias AI bill before you turn it on

I ran this against a real $30M brand's 6,000-ticket month, then priced the same brand's phone calls per resolution, and the gap between the two is the whole story. Here's the way to model the AI side so you're not surprised in November.

Three levers move the number. Get these three right and your forecast holds:

• Your real monthly ticket volume: not last quarter's average, your peak month. The bill scales with tickets, and the spike is when it hurts.
• Your honest resolution rate: use 25% to 30% for month one, climbing toward 40% over a quarter if the knowledge base is strong. Don't plug in 60%.
• Billing term and overage exposure: annual locks the $0.90 rate, monthly costs $1.00, and everything above your bundle is $1.50. The bigger your spike, the more of your bill lives at that $1.50 line.

Here's an illustrative forecast for a $30M brand on a Pro plan, 6,000 tickets in a busy month, annual billing. Treat every cell as illustrative, not a Gorgias quote:

Add the $300 base and the ticket overage on 4,000 tickets above the 2,000 bundle (roughly $1,440 at $0.36) and the all-in number climbs again. The point isn't the exact figure, it's that the same brand can see a 2.5x swing month to month on a model that bills per outcome. If predictability matters to your CFO, this is the conversation to have before peak, not after. The Gorgias CFO breakdown is a useful companion read here.

The blind spot the per-resolution model never bills you for: the phone

Here's what the per-resolution math quietly leaves out. It only ever credits the AI for resolving a text conversation. Your phone line is not part of the deal.

When a customer calls, that call still creates a ticket (so you pay the helpdesk ticket cost), the AI does nothing with it (so you earn zero automation credit), and a human picks up (so you pay a rep). The most repeatable, most expensive channel you run is the one channel the per-resolution model can't touch. You're paying for the phone three ways and automating it zero.

And the phone is where the volume hides. WISMO ("where's my order") runs 30% to 40% of all tickets, north of 50% at peak, per Salesforce. At a $250 AOV, 12% to 18% of orders generate a phone call, versus 3% at a $40 AOV. Those are the same WISMO calls your AI handles beautifully over chat, arriving on the one channel it's locked out of. If reducing that load is the goal, the playbook on WISMO automation for Shopify covers both channels.

And the cost of not answering is brutal. 85% of callers who can't reach a person never call back, and 62% switch to a competitor, per PCN's 2026 missed-call study. So the phone isn't a side channel. It's where the same questions over and over show up, where your reps burn their day, and where the per-resolution model sends you exactly zero help. If your phone backlog is the real problem, book a 30-min call and we'll map your routing.

What this costs you on the phone vs Ringly

Run the per-resolution logic on the phone channel and the comparison gets simple. A human rep handles a call at a loaded cost of around $2.70, and replacing a single CS rep runs $14,113 against 31.2% industry turnover per Gartner data via Insignia. A BPO contract runs $1.50 to $3.50 a call plus minimums (more on the tradeoffs in our guide to reducing call center costs). Here's how that lines up against AI phone support priced per resolved call:

Ringly.io is AI phone support for Shopify brands. The phone shouldn't be a tax on your support team. Instead of growing headcount every time call volume goes up, the AI takes the routine inbound calls (order status, returns, product questions, abandoned-cart rescue) so your reps get the hard calls back. Across 50+ brands the AI resolves 73% of calls autonomously at roughly $0.42 per resolved call, and the calls that need a person escalate cleanly to Gorgias, Richpanel, Reamaze, or whatever helpdesk you already run.

WashCo, a Shopify brand we launched, recovered $22,664 in its first 7 days on the phone at $0.91 per call. That's the per-resolution comparison that actually moves your P&L: the per-resolution number you can't get on the phone from Gorgias, on the channel that carries your most repeatable volume.

Plans: Grow $349/mo (1,000 minutes), Pro $799/mo (2,500 minutes), Enterprise custom (see full pricing). Live in under an hour, with a 65% resolution guarantee. Ringly doesn't replace Gorgias. It sits in front of the phone line and hands the complex calls back to the helpdesk you're already paying for, the way the AI customer support phone agent for Shopify is built to do. Want to compare the two cost models against your real numbers? Book a 30-min call and we'll do the math live.

Frequently asked questions

How much does Gorgias AI Agent cost per resolution in 2025? $0.90 per resolved conversation on an annual plan, $1.00 month to month, with starter pricing from $1. A resolution is billed only when the AI closes a conversation on its own within 72 hours.

Does Gorgias really bill me twice for one AI resolution? Effectively, yes. Each AI resolution counts as one billable helpdesk ticket and incurs the per-resolution fee, so the true unit cost is the AI fee plus your marginal ticket cost, not the $0.90 alone.

What happens when I exceed my plan's automated interactions? Overages are charged at $1.50 per interaction on Support plus Shopping Assistant plans, a 67% premium over the $0.90 base. Bundles cap around 2,500 interactions, so most peak-season resolutions hit the overage rate.

What resolution rate should I forecast? Use 25% to 30% for the first month and climb toward 40% over a quarter. Gorgias markets "up to 60%," but published case studies land between 24% and 56%, so forecasting at the marketing rate overstates your savings and understates your bill.

Does Gorgias AI Agent automate phone calls? No. The per-resolution model only credits text conversations. Phone calls still create a ticket, still need a human to answer, and earn no automation credit, which is why the phone is the channel this pricing model never touches.

Is annual or monthly billing cheaper for AI Agent? Annual locks the lower $0.90 per-resolution rate and lower overage exposure; monthly costs $1.00. If your volume is steady, annual usually wins. If it's spiky, model the overage either way before committing.

Can I keep Gorgias and add AI phone support? Yes. Ringly sits in front of your phone line and escalates complex calls to Gorgias, so you keep your helpdesk, your number, and your workflows while the routine calls get resolved for roughly $0.42 each.

Talk to us

If you run a $10M to $100M Shopify brand on Gorgias and the AI line item keeps moving while your phone still goes to a human, a 30-min call is the fastest way to see both cost models side by side and what your phone channel is leaving on the table.

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 65%.

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

Book a 30-min call →