Mike HeapMike is an experienced Product Manager who focuses on all the “non-development” areas of My AskAI, from finance and customer success to product design, copywriting, testing and more.
Table of Contents
- Gladly at a glance
- What is Gladly?
- How easy is it to set up Gladly?
- What channels does Gladly work in?
- What are the limitations of Gladly?
- What knowledge sources can I train Gladly on?
- What features does Gladly have?
- How do I improve Gladly's responses?
- What resolution rate can I expect from Gladly?
- What AI model does Gladly use?
- What languages does Gladly work in?
- How secure is Gladly?
- Who is using Gladly?
- How much does Gladly cost?
- Gladly Team (seats plus AI usage)
- The Shopify plan (per AI resolution)
- What it costs at 2,000 conversations a month
- How do I model Gladly's cost at my own volume?
- Does Gladly have a free trial?
- Is Gladly worth it?
- What are the Pros and Cons of Gladly?
- Pros
- Cons
- Rating
- FAQs
Created time
Sep 9, 2026 01:53 AM
Title length (<60)
Author
Last optimised
Ecomm?
Tags
Image
Publish date
Aug 17, 2026
Video
Slug
gladly-complete-guide-2026
Featured
Type
Article
Ready to Publish
Ready to Publish
Two price lists. On Shopify: $1.50 a resolution, $0.25 an assist, $120 a seat. Off Shopify, Gladly publishes no pricing on its own site. And the unit Gladly bills its AI on is the conversation it could not finish alone.
Gladly AI is the automation layer inside Gladly's own ticketless helpdesk: it answers questions, processes returns and order changes, and hands over when it cannot finish. The only rates it publishes sit on its Shopify listing, where a resolution is billed at $1.50, an assist at $0.25 and a seat at $120 a month; the enterprise platform is quoted only through a demo. Adopting it means replacing the helpdesk your team works in today, and most buyers settle that question long before they argue about price.
Someone upstairs has asked you to price Gladly.
Maybe your CFO saw the seat cost on a renewal sheet. Maybe your CEO spotted Nordstrom and Ulta Beauty on the logo wall and asked why you are not there too.
Either way, you are the one who has to come back with a number. Gladly has taken its pricing page down, and the only rates it publishes sit on a Shopify app listing that most buyers never see.
By the end of this you will know which of Gladly's two pricing models you are being offered. You will also know what each works out at on a mid-sized support load, and whether the switch is one your team can stomach.
Gladly at a glance
Fact | Detail |
What it is | Ticketless consumer-brand platform with native voice |
What buyers pay | ~$180/seat/mo buyer-reported on enterprise, pay-as-you-go on Shopify |
G2 rating | 4.7/5 from 1,112 reviews |
Best for | Retail and DTC brands with voice volume |
Not a fit for | Teams keeping the helpdesk they run today |
Biggest limitation | Adopting it means replacing your helpdesk |
Our verdict | 7/10 |
My AskAI is an AI agent that installs inside the helpdesk you have now, so nothing about your stack has to change and there is no migration to schedule.
What is Gladly?
TL;DR: A ticketless customer service platform that keeps one lifelong conversation per customer across voice, chat, email, SMS and social, with an AI that resolves inside it.
Gladly throws away the ticket. Every message a customer sends threads into one continuing conversation, and Gladly's own product documentation sets it out plainly. An agent opening it sees last week's phone call and this morning's SMS in one place (no ticket number, no case history to reassemble).
The buyer it is built for is a mid-market or enterprise consumer brand with real phone volume, the kind where support drives repeat purchases. Buyers report about $180 a seat each month before any AI usage. The median customer on Vendr's marketplace spends $24,150 a year (that figure is the whole contract).
The trade-off is structural. Gladly replaces the helpdesk you have, so buying it means moving everyone off what they work in today.
In support-team terms it is three things. There is the AI, which answers and acts on its own, and there is Gladly Team, the workspace your agents live in. Behind both sits the Help Center, where the Answers that feed them are written (by hand, by someone on your team).
The names moved, and Gladly's own docs still carry both sets. The AI used to be called Sidekick and is now just Gladly (the dashboard documentation is still titled for Sidekick).
The agent workspace used to be Hero and is now Gladly Team. Buyers still say Hero on procurement calls, so expect to hear it.
The company was founded in San Francisco, and its about page names co-founder and CEO Joseph Ansanelli as the person still running it. Its homepage leads on "Commerce AI for every customer moment".
G2: Gladly scores 4.7/5 from 1,112 reviews on G2.
Gladly is a platform you move onto. Ours is a layer you drop into the helpdesk you already chose, and we work inside Zendesk, Intercom, Freshdesk, Freshchat, Gorgias and HubSpot.
How easy is it to set up Gladly?
TL;DR: Standing Gladly up is a helpdesk migration. Your CX team can own the no-code Guides, but channels, routing and the Answers knowledge base all have to exist before the AI has anything to work from.
Gladly documents the order in its implementation guide. You connect the channel entry points first, then you set up routing.
Next comes the Help Center and the Answers behind it. Only then do you write the Guides the AI follows.
Nothing the AI says works until the three steps before it are finished, so the timeline is however long that rebuild takes.
Three of those four steps are things your current helpdesk already does for you (channels, routing and a knowledge base). You are rebuilding them.
The parts aimed at business users really are aimed at business users. Guides are written in plain language and Journeys are mapped visually, so nobody needs a developer to change what the AI says.
Actions and the App Platform run on REST APIs, and that lands with whoever owns
developer.gladly.com in your org (the no-code promise stops at what the AI says).The Shopify listing is the one self-serve route in. Its free plan gives you "100 AI interactions to test out" and "1 team member seat", which is enough to see the product move.
Reviewers at mid-market size consistently describe the rollout as heavier than they expected. Treat it as a project with a named owner and a start date.
There is nothing to migrate with ours. My AskAI installs into your existing helpdesk in about ten minutes, either as an approved app from that helpdesk's marketplace or as an API integration. Prefer the app: it has been through the marketplace review, the helpdesk knows it is there, it is more stable, and it is far less likely to be pulled.
What channels does Gladly work in?
TL;DR: Voice, chat, email, SMS, Facebook and Instagram, all stitched into one customer timeline.
Gladly covers voice, chat, email, SMS, Facebook and Instagram, and its product page frames all of them as a single timeline.
WhatsApp is supported too. The help center carries a full category for it, with articles titled "Set Up and Configure WhatsApp", "Respond to and Initiate WhatsApp Messages" and "Disconnect WhatsApp". The marketing channel list leaves it out (so the docs are the better source here).
Gladly built voice as a real channel, and its AI takes real actions during the call itself. If a third of your volume arrives by phone, that difference is what puts Gladly on your list.
Gladly is a helpdesk, so it has no helpdesk integrations.
Its app directory is built around commerce and operations: e-commerce, subscriptions, loyalty, returns, shipping, order management, customer data platforms, and workforce tools like Assembled, MaestroQA, Medallia and Playvox. There is no helpdesk category in it and no helpdesk connector.
We answer in whatever channels your helpdesk routes to us, social messages included, because My AskAI works inside Zendesk, Intercom, Freshdesk, Freshchat, Gorgias and HubSpot.
What are the limitations of Gladly?
TL;DR: Three things decide deals here: what a seat costs, the fact that you have to replace your helpdesk to get any of it, and a knowledge base your team writes by hand.
Mid-market reviewers consistently name the per-seat cost, and the absence of a flexible plan, as the sticking point.
The neutral number sits on Vendr. The median Gladly buyer pays $24,150 a year, in a range from $14,323 to $78,521, against a buyer-reported list of roughly $180 per user per month. That is a real budget line before a single AI conversation is billed (the AI usage is quoted on top of it).
You write the knowledge yourself. Gladly's authoring documentation is clear: Answers are written by a person and used as written. Nothing ingests your past conversations to seed the knowledge base.
Somebody on your team owns that library permanently, and what the AI can answer is limited to what they have found time to write.
The agent workspace runs in three languages, and Gladly's own language documentation lists English, Spanish and German.
That is separate from what your customers can write in, which is much broader. Right-to-left scripts are unsupported apart from Hebrew (Arabic and Aramaic are the examples the docs name). If your agents work in Paris or São Paulo, they work in an English interface.
The integration set is built for retail. The app directory carries subscriptions, returns, shipping and order management (the same directory that carries no helpdesk connector).
That is excellent if you sell things. A B2B software company has fewer of those systems to connect.
The knowledge point is where we differ most. My AskAI trains on your help center and connected content. It can also draft starter knowledge from your historic tickets, so a team with thin documentation still has something for the AI to answer from on day one.
What knowledge sources can I train Gladly on?
TL;DR: Two documented sources, which are the Answers you write in Gladly's own Help Center and the websites you point it at, plus live business data through Actions.
Gladly's Knowledge Sources documentation sets out two, under the headings "Use Gladly Answers as a source" and "Add and manage websites".
The first is Public Answers. These are the articles in your Gladly Help Center, and they do double duty: customers read them on the site, and the AI answers from them. The authoring guide is strict about how they should be written (one topic each, key action first).
The second is websites. You add a domain, Gladly indexes it, and the AI answers from what it finds. The docs cover the refresh cadence and how to block specific pages from being read.
Live data is a separate path (knowledge answers the question, Actions go and look something up). Order status, account details and anything else that changes hourly comes through Actions and the commerce integrations such as Shopify, Klaviyo and your own APIs. That is what lets the AI process a return.
On our side, knowledge comes from your help center, your website, uploaded documents and cloud sources like Google Drive, Notion, Confluence and SharePoint, plus training on your historic tickets.
What features does Gladly have?
TL;DR: Three layers. An AI that resolves and takes real actions, an agent workspace built on one customer timeline, and a dashboard that tells you which Guides the AI is actually handling.
Start with the autonomous layer. Guides are plain-language instructions for how the AI should handle a situation, Journeys map those situations out visually, and Answers supply the facts.
On the Shopify side Gladly describes the same layer as "AI Resolution: AI fully resolves orders, returns, and WISMO without an agent". That is a fair summary of what it is aiming at (WISMO being the volume every consumer brand wants gone first).
Then there are actions on real systems. Gladly's Intelligent Action can process a return, update an order, issue a refund or change a booking, by reaching into the commerce systems in its app directory.
It reaches as far as your connected systems do, so a process that still lives in a spreadsheet (nothing wired up yet) stays with your team.
Last is the assist layer. When the AI cannot finish, it hands over with the context already gathered. Gladly's own line for it is "AI Assist: AI gathers context and hands off to your team, reducing handle time".
On the Shopify plan an assist and a resolution are billed as two different things, at two different prices.
We do the action half through Tasks and Tools, written as natural-language procedures with no decision trees in sight. Each action is your call: the AI can complete it outright, or draft it and wait for an agent to approve. Those actions are charged on usage on top of your credit rate.
How do I improve Gladly's responses?
TL;DR: Write better Answers, refine the Guides, read the dashboard, repeat. The loop is content curation, and somebody on your team owns it permanently.
Gladly is unusually direct about how to write for its AI. The authoring rules ask for one topic per Answer, the key action first, present tense, and no two Answers that contradict each other (that last one is the rule most knowledge bases break first).
Follow them and the AI answers from what you wrote.
The measurement side lives in the Gladly AI dashboard. It breaks performance down by contact reason (Gladly's term for a ticket topic).
You can see which Guides the AI is handling and which ones it keeps passing to a person. Improvement Opportunities surfaces the second group for you.
The loop is somebody reading the dashboard every fortnight, spotting the reason code with a poor completion share, and writing the Answer that fixes it. Budget it as part of somebody's week.
We try to take that job off you. My AskAI compares its own draft to what your agent actually sent after a handover, then writes the missing knowledge itself, ready for someone to approve. If your documentation is thin to begin with, we train on your historic tickets, the last 5,000 by default and more on request, so the AI has something to answer from on day one. You can also batch-test changes against sample questions before any customer sees them.
What resolution rate can I expect from Gladly?
TL;DR: Gladly counts a resolution only when its AI finished the job without handing off, and it bills on the conversations that were handed off. The definition underneath the number decides what the number means.
The definition is the most useful thing Gladly publishes about its AI. Its dashboard documentation says a Resolution is an "Assisted Conversations in which Gladly AI completed the steps specified in the Thread without handing off to an Agent." The rate is "Calculated as Resolutions/ Resolutions + Assisted (Handed-off) Conversations."
The same page adds the sentence a buyer needs: "Assisted Conversations is the unit of measurement for Gladly AI billing."
So the conversations that pull your resolution rate down are the conversations you are charged for. Model the invoice on handovers.
That view comes from Gladly's own demo account, so the 31% on it is a demo figure rather than a Gladly benchmark. The resolution rates named later for The Black Tux and Ollie (further down, under "Who is using Gladly?") are figures from Gladly's published customer stories, a separate source from this screen.
Gladly also flags its own inconsistency: "Note that other Gladly AI performance reports may calculate the Resolution Rate using a different denominator."
Our own AI resolution rate benchmark study covers 195 rated deployments across 38 vendors. The median lands at 70%.
Three caveats travel with that figure. It is an aggregate across the field, so no single deployment sits behind it. It is directional only, because vendors count differently.
The sample is also self-selected, so it skews toward deployments people were willing to rate. Gladly's own caveat makes the point for us: the definition moves the number more than the capability does.
We count a conversation as resolved when the AI handled it without escalating to a human, and we make escalation easy, so the number stays conservative. When a number looks wrong, your team can ask Echo, the operator agent inside our dashboard. It will say why the AI answered as it did on any conversation, which knowledge source it used, and why a CSAT score landed where it did. Our published benchmark study is the place to check any vendor's claim, including ours.
What AI model does Gladly use?
TL;DR: Gladly names four providers, which are Microsoft Azure, OpenAI, Anthropic and Deepgram for voice, and does not say which model runs which task.
Gladly names its AI providers and stops short of naming models. Its published security material lists Microsoft Azure, OpenAI and Anthropic on the language side, with Deepgram handling speech for the voice channel.
What it does not publish is the mapping (which provider answers a chat, which one summarizes a call, or what happens if one of them is unavailable).
For most support teams that is fine. If your security review asks vendors to name the model behind a customer-facing reply, expect that question to go to a sales call.
On data handling, Gladly states that customer messages processed by its AI are not used to train models. It adds that sub-processors hold data only briefly for abuse monitoring, and that personal information is redacted automatically (redaction and retention are two separate questions, worth asking apart).
Those are the vendor's own claims. Its Trust Center is the place to ask for the underlying evidence.
We use a mix of models, mostly OpenAI, with Anthropic and Google models where they do a specific job better. Smaller fast models handle routing and summarizing, and a larger one writes the reply the customer reads.
What languages does Gladly work in?
TL;DR: Most ISO 639-1 language codes for customer messaging with real-time translation, an agent workspace that runs in English, Spanish and German, and no right-to-left support except Hebrew.
Buyers conflate two questions here. What your customers can write in is one. What your agents see is another.
On the customer side Gladly is broad. Its language documentation states that "Gladly supports most ISO 639-1 standard language codes in all Channels" (the two-letter code list, which covers most languages you could name). Translation runs inbound and outbound, so an agent can reply in their own language.
On the agent side the same page says that "The Gladly Team interface is offered in English, Spanish, and German." A toggle in the user profile switches between them (per user, so a mixed team can each pick their own).
On scripts it adds that "Right-to-left languages like Arabic and Aramaic are currently not supported except for Hebrew."
Surface | What is supported |
Customer messaging: chat, email, SMS, social | Most ISO 639-1 codes, with inbound and outbound translation |
Agent workspace | English, Spanish and German |
Right-to-left scripts | Not supported, except Hebrew |
My AskAI detects the language of each message and replies in it across 95 languages. On Intercom, Live Translation can also bring the conversation back into your agent's language on handover and translate their reply forward, and that one is charged on usage on top of your credit rate.
How secure is Gladly?
TL;DR: Gladly's Trust Center describes the platform as SOC 2 certified in its own words and keeps the evidence document behind a login, so the rest of the detail arrives through a sales conversation.
Gladly's Trust Center is a Vanta-hosted portal. Its own summary line reads: "our SOC 2 certified platform and transparent Trust Center ensure your brand and customer data are always protected".
The report itself sits behind a request form. That is normal for a platform at this size, so budget a week for it if procurement wants the report before signature.
Beyond that line, Gladly states PCI-DSS handling for card data and GDPR coverage through data processing agreements. It also states AES-256 encryption for data at rest and TLS 1.2 for data in transit.
Take those as vendor claims and put them in the diligence pack to verify.
I ask the same three things of every vendor at this price point: for the SOC 2 report early, for the list of sub-processors, and for the name of whoever can approve a redline.
We hold SOC 2 Type II and are GDPR compliant, with AES-256 encryption at rest and TLS in transit, and our live trust report is public. Customer data is never used to train models.
Who is using Gladly?
TL;DR: Consumer brands where support drives repeat revenue, heavily weighted to retail, DTC, travel and hospitality.
Three Gladly customers have published numbers, and each one measures something different.
The Black Tux rents formalwear, and its story puts AI resolution at 68%. Ollie sells pet food on subscription, and the Ollie story reports a 60% overall AI resolution rate.
ProClip USA makes vehicle mounts, and its story measures something else: average response time came down from 13 hours to under one. That is a speed number (which does not line up against the other two).
Around those three sits the customer list itself, which names Tory Burch, UGG, Nordstrom, Ulta Beauty, Crate & Barrel, Rothy's, TUMI, Condé Nast, Breeze Airways, Tecovas, PacSun, HOKA and gruns. Almost every name sells a considered physical product to a consumer who will buy again. Breeze Airways is the exception, an airline where a service call decides whether someone books again.
The Shopify listing is the down-market route for smaller merchants. The pricing splits in two along with it. MaryRuth Organics is one of them:
"We've recently deployed Gladly AI across Chat, SMS, Email and Voice and we are very impressed with our resolution rates as well as the ease of AI management within the Gladly toolset." MaryRuth Organics, reviewing Gladly on the Shopify App Store
More than 200 ecommerce and SaaS businesses run customer service on My AskAI, and our agents have resolved over a million tickets. We saw Edel Optics reach 79% AI resolution after connecting live customer data. The named stories, with the numbers behind them, are on our reviews and case studies hub.
How much does Gladly cost?
TL;DR: Two completely different models. A seat-priced enterprise platform with no public numbers, and a public per-resolution plan on Shopify. Which critique applies depends entirely on which one you are being sold.
Gladly publishes no pricing page. The old URL redirects to the homepage, and the only route to a quote from Gladly is a demo.
Factor that into your timeline. You cannot build a business case before you have taken a call. On a vendor with no public price I get the number in writing first, then model around it.
Gladly Team (seats plus AI usage)
Gladly quotes the enterprise platform only through a demo. The one figure available on a neutral source comes from a buyer at a company of 201 to 1,000 employees:
"For a new purchase, Gladly was willing to offer Hero AI Support users for $155/user/month (normally $180)." Buyer report on Vendr
The same page shows what a contract looks like. The median Gladly customer pays $24,150 a year, roughly $2,013 a month, with the range running from $14,323 to $78,521.
Gladly contracts annually, so that is a committed number you cannot switch off month to month. Ours is billed monthly, so the comparison to make is a monthly rate against a year of committed spend.
Gladly bills its AI on Assisted Conversations, and it publishes no rate for them.
When you take the demo, ask for the AI line as a figure separate from the seats. The seat price is only the floor of that invoice.
The Shopify plan (per AI resolution)
Gladly's Shopify listing publishes rates, and the model there is completely different: "$1.50 per AI Resolution", "$0.25 per AI Assist" and "$120/mo per team member".
You also set a ceiling: "Set your monthly limit and pay as you go. Everything in the free trial + unlimited interactions up to your spending cap." That cap is a spend control (set it before you switch the AI on).
The per-resolution critique applies to this plan. A resolution costs six times an assist, and the cap you set is what keeps the total in hand.
On the enterprise platform the seats dominate and the AI line is negotiated. Arguing about "Gladly's pricing" without saying which of the two you mean produces a case that is false about one of them. I have priced both.
What it costs at 2,000 conversations a month
Assumption | Value |
Conversations a month | 2,000 |
AI resolution rate | 50% |
Support agents | 8 |
Two of the four options below charge a flat seat fee only; the other two also bill for what gets used.
Option | Plan priced | Monthly cost | What the figure covers |
Gladly Team | Hero AI Support, 8 seats at $180 | $1,440 | Seats at the buyer-reported list on Vendr, assuming all 8 agents need one. Gladly publishes no AI rate |
Gladly Team, negotiated | Hero AI Support, 8 seats at $155 | $1,240 | As above, at the buyer-reported discount on Vendr |
Gladly on Shopify | Pay as you go | $2,710 | 1,000 resolutions at $1.50, 1,000 assists at $0.25, and 8 seats at $120. Assumes the 50/50 split above, and a spending cap raised above the listing's $250 Starter cap |
My AskAI | Pro | $319 | $199 base with 1,000 credits, plus 1,000 more at $0.12. Pro includes 5 team seats, and only the agents who actually use My AskAI need one; if more than five do, Scale at $499 lifts the limit. Either way it sits on top of the helpdesk seats you keep paying |
The Gladly rows are the cost of a whole platform, so they replace what you pay your current helpdesk. The My AskAI row is an AI layer sitting on top of a helpdesk bill you already pay. The two rows are not a like-for-like swap.
On Shopify the AI lines are most of the bill. Push the resolution rate from 50% to 75% and that row climbs to $3,335. Our row does not move at all, because we charge per ticket, resolved or not, at $0.12 a credit on Pro.
How do I model Gladly's cost at my own volume?
Both models have to be priced separately, and each one needs the same inputs from you. Paste this into whichever AI assistant you use, with your own numbers in the brackets (it cannot judge answer quality, so still test on your own tickets before you sign anything).
You are helping me price Gladly for a customer support team.
My inputs:
- Customer conversations a month: [number]
- Share the AI is expected to finish without a human: [percentage]
- Support agents who need a seat: [number]
- Helpdesk we run today, and what it costs us a month: [name and figure]
- Which route we are being sold: [Shopify app listing / enterprise platform]
Do this:
1. Price the Shopify route. Resolutions bill at $1.50 each, assists at $0.25
each, seats at $120 each a month. Treat every conversation the AI does not
finish on its own as an assist.
2. Price the enterprise route on the per-seat figure I give you from the quote.
If I have not given you one, write "unverified, ask the vendor" and leave
that line unpriced.
3. Show both totals again if the AI finishes a quarter more of the
conversations than I said, so I can see which way each bill moves.
4. Add any AI usage line I was quoted on the enterprise route. Mark it
"unverified, ask the vendor" unless I supplied a figure.
5. Subtract what we stop paying our current helpdesk, because Gladly replaces
it rather than sitting on top of it.
Output one table, a row per option, a column for monthly cost and a column for
what the figure covers. Then give me three sentences: which option is cheapest
at my volume, what would change that answer, and which figures you could not
verify.Our pricing is public and you can model it before you talk to anyone. The rate does not climb as the AI gets better. There is a 30-day free trial too, with all features unlocked, unlimited tickets and no card required, so you can prove the resolution rate on your own tickets first.
Does Gladly have a free trial?
TL;DR: Yes on the Shopify plan, where Gladly's own words are 30 days free. On the enterprise platform Gladly publishes no pricing page, and every route in is a demo.
On Shopify, Gladly states the terms itself, in a reply to a merchant review: "It's completely free for 30 days and you only pay when Gladly resolves a customer issue after that." The free plan card alongside it offers "100 AI interactions to test out" on "1 team member seat".
A hundred interactions buys you a demonstration and not much of a pilot. I would want a fortnight of real volume behind a resolution rate before I believed it.
A store doing 60 conversations a day burns through it before lunch on day two. After that you are choosing between the $250 monthly Starter cap and turning it off. Plan what you want to learn from those hundred conversations before you switch it on.
That 30-day offer is available on the Shopify plan only. On the enterprise platform Gladly publishes no pricing page, and every call to action on it points at a demo. Budget several weeks between first contact and a number you can take to finance.
Ours runs 30 days with every feature unlocked, unlimited tickets and no card, so you can prove the resolution rate on your own tickets before you pay for anything.
Is Gladly worth it?
TL;DR: Worth it for a mid-market or enterprise consumer brand with real voice volume that is ready to replace its helpdesk. Not worth it if you like the helpdesk you have.
Choose Gladly if:
- Support drives repeat purchases for your consumer brand
- You have real phone volume and want AI that acts on the call itself
- Your systems are built for commerce: Shopify, Klaviyo, subscriptions, returns, order management
- You are ready to replace your helpdesk and the budget covers roughly $180 a seat
Don't choose Gladly if:
- You like the helpdesk you already run, because there is no way to put Gladly on top of one
- You need a number before you take a sales call
- You want the AI to learn from the tickets you already have, without a hand-written Answer library
- Your agents need the workspace in a language other than English, Spanish or German
I run an AI support company, so I have a stake in this answer.
If you are a growing apparel or DTC brand, your customers phone you, and the helpdesk you run has been irritating you for two years, Gladly is a good answer. I want you buying the thing that fits.
The buyer it does not suit is the one asking a narrower question: how do I add AI to my customer support without replacing my helpdesk? That question is the one we built My AskAI to answer.
Both are legitimate projects of very different sizes. Teams routinely underestimate how big a helpdesk replacement is.
If you want to keep Zendesk, Intercom, Freshdesk, Freshchat, Gorgias or HubSpot and add an AI agent inside it, that is exactly what My AskAI does. Gladly sits in our Kustomer alternatives roundup too, scored against the same criteria as its closest peer.
What are the Pros and Cons of Gladly?
Pros
- One lifelong customer conversation: voice, chat, email, SMS and social thread into a single timeline, so an agent opening a message sees last week's phone call beside this morning's SMS.
- Native voice AI that acts on the call: voice is built as its own channel here, and the AI takes real actions during a call.
- Real actions on real systems: returns, order edits and refunds run through connected commerce data, so the AI finishes the request itself.
Cons
- It replaces your helpdesk: the app directory carries no helpdesk category and no helpdesk connector (it is built around Shopify, Klaviyo, returns and order management), so there is no way to run Gladly alongside the one you own.
- Gladly publishes no enterprise pricing: the pricing page now redirects to the homepage, and the only route to a quote is a sales call.
- The knowledge base is written by hand: Gladly's own docs say Answers are human-written and used as written, so nothing is learned from the tickets you already have.
- The agent workspace runs in three languages: English, Spanish and German only, with no right-to-left support apart from Hebrew (customers can still write in most ISO 639-1 codes).
Rating
The rating covers Gladly's AI capability, setup, channels, knowledge sources, improvement loop, resolution reporting, model providers, languages, security, published customers, pricing and free trial. Evidence comes from Gladly's own product, implementation and language documentation, its Shopify app listing, its Trust Center, its published customer stories, buyer-reported contract data on Vendr, and G2. Where Gladly publishes no figure, the gap is stated as a gap rather than estimated.
Gladly
- Brand: Gladly
- Rating: 7/10
- In a sentence: A best-in-class relationship-first platform for consumer brands, priced and structured so that only a certain size of team can buy it.
Seven of ten, for two things nobody else in this category does as well: one lifelong customer timeline where every peer gives you a threaded case list, and voice AI that takes real actions on a live call. Two things keep it off an eight: what a seat costs, and the fact that adopting it means replacing the helpdesk your team already knows.
For the head-to-head with the cost math at your own volume, the closest published comparison we have is My AskAI vs Dixa Mim AI. Dixa poses the same question Gladly does about replacing the helpdesk you run.
FAQs
What does Gladly AI do?
Gladly AI answers and resolves customer conversations across voice, chat, email, SMS and social, inside Gladly's own platform. On the commerce side it describes its scope as "AI Resolution: AI fully resolves orders, returns, and WISMO without an agent". When it cannot finish a request it gathers the context and hands the conversation to an agent, and Gladly's product documentation covers how that timeline works.
Who owns Gladly?
Gladly is a private company headquartered in San Francisco, led by co-founder and CEO Joseph Ansanelli. It is not a subsidiary of a larger platform (no parent company's roadmap sets its priorities). Its about page carries the leadership and company detail.
Who is the CEO of Gladly AI?
Joseph Ansanelli, who co-founded the company and still runs it. He is named on Gladly's about page alongside the rest of the leadership team.
What is Gladly Sidekick?
Sidekick was the old name for Gladly's AI, which is now simply called Gladly (the docs have not all caught up, so you will meet both names in the same help center). The agent workspace went through the same change, from Hero to Gladly Team. Buyers still see "Hero" on quotes, so a Hero line item is not an old product.
Is Gladly a CRM or a helpdesk?
Neither in the usual sense, though it does the job of a helpdesk. Gladly organizes everything around one continuing customer conversation, with no tickets and no cases (no ticket IDs either, so check that against how your team reports today). That is why its own documentation calls it ticketless. Customers do not always use the vendor's category:
> "We just converted to Gladly and it was a very easy process to get the CRM set up and running."
> Merchant review on the Shopify App Store
Does Gladly integrate with Zendesk or Intercom?
No. Gladly is a helpdesk in its own right, and its app directory carries no helpdesk category and no helpdesk connector. Ask a general AI assistant and it may well list Zendesk among Gladly's integrations, which is worth checking against that directory before it reaches a board slide. Adopting Gladly means replacing your current helpdesk. We only ever run inside a helpdesk you already have.
How much does Gladly cost per seat?
On the enterprise platform, roughly $180 per user per month at buyer-reported list, with one buyer on Vendr reporting a negotiated $155. On the Shopify plan a seat is "$120/mo per team member", with AI usage billed separately on top.
Plan | Seat cost | AI usage |
Gladly Team (enterprise) | ~$180/user/mo buyer-reported list, ~$155 negotiated | Not published |
Gladly on Shopify | $120/mo per team member | $1.50 per AI Resolution, $0.25 per AI Assist |
Does Gladly have a free trial?
Yes on the Shopify plan. Gladly's own words are "It's completely free for 30 days and you only pay when Gladly resolves a customer issue after that." The free tier is "100 AI interactions to test out" (enough for a demo). That offer is on the Shopify plan only, and the enterprise route in is a demo.
Does Gladly support WhatsApp?
Yes. Gladly's help center carries a full WhatsApp category, with articles titled "Set Up and Configure WhatsApp", "Respond to and Initiate WhatsApp Messages" and "Disconnect WhatsApp". The marketing channel list does not enumerate it (the marketing pages and the docs disagree here, and the docs are the ones to trust), so check the help center if WhatsApp matters to you.
How many languages does Gladly support?
Two answers, depending on who you mean. For customers, Gladly "supports most ISO 639-1 standard language codes in all Channels", with translation in both directions. For agents, "The Gladly Team interface is offered in English, Spanish, and German." Gladly's language documentation also states that right-to-left scripts are unsupported apart from Hebrew.
What AI model does Gladly use?
Gladly names four providers and no models: Microsoft Azure, OpenAI, Anthropic, and Deepgram for the voice channel. It does not publish which one handles which task, so if your security review needs that mapping, raise it on the call. We publish our own mix, mostly OpenAI with Anthropic and Google models where they do a specific job better, which is a fair thing to ask of any vendor on your shortlist.
How do I add AI to my customer support without replacing my helpdesk?
Install an AI agent into the helpdesk you already run. That is the question we built My AskAI to answer: it works inside Zendesk, Intercom, Freshdesk, Freshchat, Gorgias and HubSpot, either as a marketplace app or an API integration. Take the marketplace app where there is one, because it has been through review and the helpdesk knows it is there. Gladly is itself a helpdesk.
