Written by an Adobe Commerce engineer, not a review site
I install live chat on Adobe Commerce, BigCommerce and WooCommerce stores for a living. This is the honest comparison — including the two situations where the tool I'm paid to recommend is the wrong answer.
Gorgias published support benchmarks across more than a thousand ecommerce brands, split by the channel a conversation started in. The spread is not subtle.
Same customer. Same question. Three times the loss — and the only variable is the container you put the conversation in.
To be precise about what that 42% is: it's conversation abandonment, not cart abandonment. Forty-two percent of the people who fill in your contact form give up before you answer them. That's not the same claim as "42% of your carts", and I'm not going to make the bigger one.
What makes it expensive rather than merely annoying is which questions arrive there. On the stores I audit, 70–80% of contact-form submissions are pre-purchase: does this fit a 2019 Tacoma, is it really in stock, can I return it if the sizing is wrong, will it ship to Alaska. Those aren't support tickets. They're sales conversations arriving in a queue with a 36-hour SLA.
Baymard Institute's data on why shoppers abandon carts — excluding the ones who were only browsing — puts 19% on not trusting the site with their card details, 13% on the returns policy, and 12% on not being able to see the total cost up front. Every one of those is a forty-second answer from a human who happens to be present. Grouping them as "problems a conversation solves" is my interpretation, not Baymard's finding — but I don't think it's a stretch.
Most comparison pages try to sell everybody. This one is trying to lose about two-thirds of you, because a recommendation that fits everyone fits nobody.
Keep reading if
Close this tab if
The category has quietly repriced itself. Gorgias charges around $0.90 per AI resolution. Intercom Fin is $0.99. Zendesk is roughly $1.50 committed and $2.00 pay-as-you-go, on a meter that auto-charges overages. Text's own newer plans meter AI at $0.99 too.
Which means your support cost is now a function of your traffic. Black Friday gives you four times the volume and four times the invoice — in the week your margin is thinnest, because you discounted everything, paid for the ads and ate the shipping promo. I have yet to meet a merchant who modelled this before signing.
Flat per-seat is the counter-position: ten thousand chats and a hundred thousand chats cost the same. It loses badly if you run a large team on low volume. It wins decisively if you run a small team on high volume — which describes most ecommerce.
No email required. This is the same calculation I run for clients, and it sometimes tells them to buy the other thing.
Defaults: 45% is the median AI resolution rate Gorgias reports across its merchant base; $0.99 is Intercom Fin's and Text's published rate; $79 is LiveChat's Business tier on annual billing. Change all four to match your reality — and check whether your current contract's meter is capped. Several aren't.
| Tool | Pick it when | Where it hurts |
|---|---|---|
| Gorgias | Shopify, high volume, low AOV, and you want AI that can genuinely take actions on orders — refunds, cancellations, address changes. | Effectively Shopify-or-bust. Your bill scales with traffic and peaks in Q4. |
| Intercom (Fin) | You're a SaaS or subscription business more than a store, and you need product tours and onboarding alongside support. | $0.99 a resolution, and it's built for software companies, not catalogues. |
| Zendesk | You're already on it enterprise-wide and switching costs are real. | The most expensive per-resolution rate in the category, on an uncapped meter. |
| Shopify Inbox / Tawk.to | Under ~$250k in revenue. Free is the correct answer and anyone telling you otherwise is selling. | Limited routing, reporting and integration depth once you have a real team. |
| LiveChat MY RECOMMENDATION |
Adobe Commerce, BigCommerce or WooCommerce; 2–10 full-time agents; AOV high enough that chat is a sales channel; and you want a support bill you can forecast twelve months out. Flat per-seat on legacy plans. 100% uptime across all services June–August 2026. | Three real weaknesses. Per-agent pricing gets expensive past ~8 seats. Its AI suggests replies but can't take actions on an order. And it's a minor player in the Shopify ecosystem — 43 reviews on that app store against Shopify Inbox's 5,500. |
If the middle column of that last row isn't describing you, one of the other four is your answer. I'd rather you got that right than clicked my link.
A chat widget takes thirty seconds to paste in and about ninety minutes to do properly. The gap between those two numbers is where most implementations quietly fail — and it's the part I actually do for a living.
No credit card required · Commissioned link · Then email me for the install
Not sure it's the right fit? Send me your store URL and I'll run a six-point audit of your pre-purchase support path and send back a short Loom. No charge, nothing attached — I look at a lot of stores and the patterns are useful to me. If the audit says you need Gorgias, it'll say Gorgias.
Every figure on this page traces to a named organisation with a published methodology. The live chat industry runs on statistics nobody can source; these are the ones that survive being checked.