At Gorgias we recently switched our flask & celery apps from Google Cloud VMs provisioned with Fabric to using docker with kubernetes (k8s). This is a post about our experience doing this.

‍

Note: I'm assuming that you're somewhat familiar with Docker.

‍

Docker structure

The killer feature of Docker for us is that it allows us to make layered binary images of our app. What this means is that you can start with a minimal base image, then make a python image on top of that, then an app image on top of the python one, etc..

‍

Here's the hierarchy of our docker images:

• gorgias/base - we're using phusion/baseimage as a starting base image.
• gorgias/pgbouncer
• gorgias/rabbitmq
• gorgias/nginx - extends gorgias/base and installs NGINX
• gorgias/python - Installs pip, python3.5 - yes, using it in production.
• gorgias/app - This installs all the system dependencies: libpq, libxml, etc.. and then does pip install -r requirements.txt
• gorgias/web - this sets up uWSGI and runs our flask app
• gorgias/worker - Celery worker

‍

Piece of advice: If you used to run your app using supervisord before I would advise to avoid the temptation to do the same with docker, just let your container crash and let k8s handle it.

‍

Now we can run the above images using: docker-compose, docker-swarm, k8s, Mesos, etc...

‍

We chose Kubernetes too

There is an excellent post about the differences between container deployments which also settles for k8s.

‍

I'll also just assume that you already did your homework and you plan to use k8s. But just to put more data out there:

‍

Main reason: We are using Google Cloud already and it provides a ready to use Kubernetes cluster on their cloud.

‍

This is huge as we don't have to manage the k8s cluster and can focus on deploying our apps to production instead.

‍

Let's begin by making a list of what we need to run our app in production:

• Database (Postgres)
• Message queue (RabbitMQ)
• App servers (uWSGI running Flask)
• Web servers (NGINX proxies uWSGI and serves static files)
• Workers (celery)

‍

Why Kubernetes again?

We ran the above in a normal VM environment, why would we need k8s? To understand this, let's dig a bit into what k8s offers:

• A pod is a group of containers (docker, rtk, lxc...) that runs on a Node. It's a group because sometimes you want to run a few containers next to each other. For example we are running uWSGI and NGINX on the same pod (on the same VM and they share the same ip, ports, etc..).
• A Node is a machine (VM or metal) that runs a k8s daemon (minion) that runs the Pods.
• The nodes are managed by the k8s master (which in our case is managed by the container engine from Google).
• Replication Controller or for short rc tells k8s how many pods of a certain type to run. Note that you don't tell k8s where to run them, it's master's job to schedule them. They are also used to do rolling updates, and autoscaling. Pure awesome.
• Services take the exposed ports of your Pods and publishes them (usually to the Public). Now what's cool about a service that it can load-balance the connections to your pods, so you don't need to manage your HAProxy or NGINX. It uses labels to figure out what pods to include in it's pool.
• Labels: The CSS selectors of k8s - use them everywhere!
• ‍

There are more concepts like volumes, claims, secrets, but let's not worry about them for now.

‍

Postgres

We're using Postgres as our main storage and we are not running it using Kubernetes.

‍

Now we are running postgres in k8s (1 hot standby + pghoard), you can ignore the rest of this paragaph.

The reason here is that we wanted to run Postgres using provisioned SSD + high memory instances. We could have created a cluster just for postgres with these types of machines, but it seemed like an overkill.

The philosophy of k8s is that you should design your cluster with the thought that pods/nodes of your cluster are just gonna die randomly. I haven't figured our how to setup Postgres with this constraint in mind. So we're just running it replicated with a hot-standby and doing backups with wall-e for now. If you want to try it with k8s there is a guide here. And make sure you tell us about it.

‍

RabbitMQ

RabbitMQ (used as message broker for Celery) is running on k8s as it's easier (than Postgres) to make a cluster. Not gonna dive into the details. It's using a replication controller to run 3 pods containing rabbitmq instances. This guide helped: https://www.rabbitmq.com/clustering.html

‍

uWSGI & NGINX

As I mentioned before, we're using a replication controller to run 3 pods, each containing uWSGI & NGINX containers duo: gorgias/web & gorgias/nginx. Here's our replication controller web-rc.yaml config:

‍

apiVersion: v1
kind: ReplicationController
metadata:
 name: web
spec:
 replicas: 3 # how many copies of the template below we need to run
 selector:
   app: web
 template:
   metadata:
     labels:
       app: web
   spec:
     containers:
     - name: web
       image: gcr.io/your-project/web:latest # the image that you pushed to Google Container Registry using gcloud docker push
       ports: # these are the exposed ports of your Pods that are later used by the k8s Service
         - containerPort: 3033
           name: "uwsgi"
         - containerPort: 9099
           name: "stats"
     - name: nginx
       image: gcr.io/your-project/nginx:latest
       ports:
         - containerPort: 8000
           name: "http"
         - containerPort: 4430
           name: "https"
       volumeMounts: # this holds our SSL keys to be used with nginx. I haven't found a way to use the http load balancer of google with k8s.  
         - name: "secrets"
           mountPath: "/path/to/secrets"
           readOnly: true
     volumes:
       - name: "secrets"
         secret:
           secretName: "ssl-secret"
And now the web-service.yaml:apiVersion: v1
kind: Service
metadata:
 name: web
spec:
 ports:
 - port: 80
   targetPort: 8000
   name: "http"
   protocol: TCP
 - port: 443
   targetPort: 4430
   name: "https"
   protocol: TCP
 selector:
   app: web
 type: LoadBalancer

‍

That type: LoadBalancer at the end is super important because it tells k8s to request a public IP and route the network to the Pods with the selector=app:web.
If you're doing a rolling-update or just restarting your pods, you don't have to change the service. It will look for pods matching those labels.

‍

Celery

Also a replication controller that runs 4 pods containing a single container: gorgias/worker, but doesn't need a service as it only consumes stuff. Here's our worker-rc.yaml:

‍

apiVersion: v1
kind: ReplicationController
metadata:
 name: worker
spec:
 replicas: 2
 selector:
   app: worker
 template:
   metadata:
     labels:
       app: worker
   spec:
     containers:
     - name: worker
       image: gcr.io/your-project/worker:latest

‍

‍

Some tips

• Installing some python deps take a long time, for stuff like numpy, scipy, etc.. try to install them in your namespace/app container using pip and then do another pip install in the container that extends it, ex: namespace/web, this way you don't have to rebuild all the deps every time you update one package or just update your app.
• Spend some time playing with gcloud and kubectl. This will be the fastest way to learn of google cloud and k8s.
• Base image choice is important. I tried phusion/baseimage and ubuntu/core. Settled for phusion/baseimage because it seems to handle the init part better than ubuntu core. They still feel too heavy. phusion/baseimage is 188MB.

‍

‍

Conclusion

With Kubernetes, docker finally started to make sense to me. It's great because it provides great tools out of the box for doing web app deployment. Replication controllers, Services (with LoadBalancer included), Persistent Volumes, internal DNS. It should have all you need to make a resilient web app fast.

At Gorgias we're building a next generation helpdesk that allows responding 2x faster to common customer requests and having a fast and reliable infrastructure is crucial to achieve our goals.

If you're interested in working with this kind of stuff (especially to improve it): we're hiring!

We've released a new version of the Chrome Extension, with sharing features and a new navigation bar. We hope you'll love it!

‍

Share templates inside the extension

Before, the only way to share templates with your teammates was to login on Gorgias.io.

If you're on the startup plan, when you create a template, you can choose who has access to it: either only you, specific people, or your entire team.

‍

‍

The account management section is now available in the extension, under settings.

‍

New navigation

Tags are now available on the left. It's easier to manage hundreds of templates with them.
You can also navigate through your private & shared templates. Shared templates include templates shared with specific people or with everyone.

‍

‍

We hope you'll enjoy this new version of our Chrome Extension. As usual, your feedback & questions are welcome!

Today, we’re thrilled to announce that we’ve raised a $1.5 million Seed round led by Charles River Ventures and Amplify Partners, to help build our new helpdesk.

‍

We’re incredibly grateful to early users, customers, mentors we’ve met both at and Techstars.

‍

We started the journey with Alex at the beginning of 2015 with our Chrome extension, which helps write email faster using templates. We’ve been pleased all along with customers telling us about how helpful it was, especially for customer support.

‍

While building the extension, we’ve realized that a big inefficiency in support lies in the lack of integration between the helpdesk, the payment system, CRM and other tools support is using. As a result, agents need to do a lot of repetitive work to respond to customer requests, especially when the company is big.

‍

That’s why we’ve decided to build a new kind of helpdesk to enable customer support agents to respond 2x faster to customers. You can find out more and sign up for our private beta here.

‍

When a company has a lot of customers, support becomes repetitive. We want to provide support teams with tools to automate the way they treat simple repetitive requests. This way, they have more time for complex customer issues.

‍

We'll now focus on this helpdesk and on growing the team, oh, and if you'd like to join, we're hiring! We're super excited about this new helpdesk product. If you’re using the extension, don’t worry.

‍

Romain & Alex

We've been busy, but not deaf!

Last few months we got lots of feedback about our extension and found to our delight that most people are satisfied, but still a few recurrent issues came up:

• The HTML/WYSIWYG editor sucks.
• No support for Outlook.com.

‍

We listened and now we're presenting:

• A brand new editor
• Support for outlook.com
• More on the Rich-Text editor

‍

WYSIWYG editors for the web are notoriously buggy and are just difficult to develop.

‍

I have yet to see one that is bug free. There are few venerable editors that do a good job like TinyMCE, FKEditor or CKEditor.. but they are big and all have edge cases that break the intended formatting and add a lot of garbage html.

‍

There are newer good quality editors in town such as Redactor. The one that got my attention and finally landed in Gorgias is this wonderful editor called which is super lightweight, uses modern content-editable (no i-frames) and 'just works' most of the time. That's not to say it's perfect, but it's good enough and I'm satisfied with it's direction in terms of development.

‍

Enjoy it and as always send us bug-reports or feedback on: support@gorgias.com

TL;DR

• Eight metrics define support performance: CSAT, NPS, CES, FCR, FRT, AHT, TTR, and SLA adherence are the core benchmarks to track.
• First response time varies 5.5x across ecommerce verticals — your vertical benchmark matters more than any industry average.
• Benchmarking is a six-step process: Define goals, select metrics, source benchmarks, capture data, analyze gaps, and set targets.
• Benchmarks connect support to revenue: Faster response times and higher resolution rates directly affect repeat purchases and customer retention.
• Brands at 30%+ automation respond 10x faster than brands at zero — with the same or better CSAT.

The benchmarks in this article are drawn from the Gorgias Ecom Lab, a research hub that publishes platform-level behavioral data from thousands of ecommerce brands. Where we cite a specific figure, it comes from that data, not generic industry surveys.

What is customer service benchmarking?

Customer service benchmarking is the practice of comparing your support performance to industry standards or peer data to find gaps and set improvement targets.

It means measuring how your team performs on metrics like response time and satisfaction scores — then checking those numbers against what similar businesses achieve. The goal is not just to measure. It's to create a clear picture of where you stand and what to fix first.

Benchmarking has four core components:

• Baseline measurement: Recording your current performance across all support channels before making any changes.
• Peer comparison: Looking at how brands of similar size and type perform on the same metrics.
• Gap analysis: Pinpointing where your numbers fall short of industry targets.
• Target setting: Using benchmark data to define realistic, time-bound goals for your team.

Why customer service benchmarks matter for ecommerce brands

Benchmarks turn "we need to improve support" into a specific, actionable goal. Instead of a vague directive, you get a clear target: reduce email response time from 48 hours to 24, or lift CSAT from 72 to 82 percent.

Here's what most benchmarking guides won't tell you: the all-industry average is often the least useful number in the room.

Ecom Lab data across 14 ecommerce verticals shows that first response time varies 5.5x at the same $10M GMV band — from 1.6 hours in Hardware to 8.8 hours in Apparel. A brand sitting in the middle of that range looks fast against one peer and slow against another. Without vertical context, the comparison tells you nothing.

For ecommerce brands, support performance also connects directly to revenue. Customers who get fast, accurate answers are more likely to complete a purchase and come back again. That shows up across the whole business — from support team efficiency to operations, finance, and marketing.

Core customer service benchmarks to track

These eight metrics are the foundation of support performance measurement. Each one captures a different dimension of the customer experience, from speed to ease to loyalty.

First response time

First response time (FRT) is the time between a customer sending a message and receiving your team's first reply.

This is the metric customers feel most immediately — and the one with the most variation across ecommerce brands. According to the Ecom Lab report Stop Benchmarking Against the Average, FRT varies 5.5x across 14 ecommerce verticals at the same revenue band. CSAT, by contrast, varies by just 0.2 points across those same verticals. If you want to know whether your operation is ahead of your peers, FRT is the metric that will tell you.

General targets by channel:

• Email: Under 24 hours (excellent: under 12 hours)
• Live chat: Under one minute (excellent: under 30 seconds)
• Social media: Under two hours (excellent: under one hour)
• Phone: Under three minutes hold time

AI automation changes what's achievable here. Brands automating close to zero percent of tickets average 736-minute response times. At 30% automation, that drops to 80 minutes. At 40%, 12 minutes.

The gains don't scale evenly — they accelerate. If your team has deployed any AI, your FRT benchmark should reflect your automation rate, not just your channel.

First contact resolution

First contact resolution (FCR) measures the percentage of tickets resolved in a single interaction, without the customer needing to follow up.

A high FCR means your team has the right information and authority to solve problems on the first attempt. The industry target is 70 to 75 percent across all channels.

For teams running AI, there's a more actionable metric to track alongside FCR: AI Resolution Rate — the share of AI-touched tickets that close end-to-end without any human involvement.

Ecom Lab data shows the median ecommerce brand resolves 45% of AI-touched tickets end-to-end. The top quartile reaches 65%. Every point of improvement removes a 10-hour median wait and a one-in-three abandonment risk from a customer's experience.

Read more: First contact resolution rate: Your guide to understanding the metric

Customer satisfaction score

Customer satisfaction score (CSAT) measures how satisfied a customer was with a specific support interaction, typically through a short post-conversation survey.

The standard benchmark is 80 to 85 percent. Ecommerce brands with proactive, personalized support often reach 85 to 90 percent. For a deeper look at moving that number, see How to Improve CSAT: 8 Fixes That Make a Real Difference.

Two things stand out from Ecom Lab data. First, CSAT is remarkably stable across verticals — it varies by only 0.2 points at the same revenue band, so your category matters far less here than it does for FRT.

Second, there is a modest tradeoff early in AI adoption: brands at 20% automation average 87.9% CSAT, versus 90.3% at zero. This reflects AI encountering more diverse ticket types as coverage expands. Brands that move past 30% automation and properly configure their AI bring CSAT back up while keeping response times fast.

Net promoter score

Net promoter score (NPS) asks customers how likely they are to recommend your brand to someone else, scored on a scale of zero to 10.

It reflects the overall customer experience — not just a single interaction. The ecommerce benchmark is between 30 and 50, with anything above 50 considered strong. See How To Calculate Net Promoter Score for the full methodology.

Customer effort score

Customer effort score (CES) measures how easy it was for a customer to get their issue resolved, typically on a seven-point scale.

Lower effort correlates with higher repeat purchase rates. The industry benchmark is 5.5 or higher.

The single biggest driver of high-effort experiences is the handoff wait. According to the Ecom Lab report The Cheapest Ticket Is the One a Human Never Touches, the median wait between an AI handing off and a human responding is 10 hours. At the 90th percentile, that wait hits 71 hours — three full days. And a third of handed-off tickets never receive a human response at all.

That experience is what drives CES scores down. Reducing handoffs, not just handling them faster, is the most direct path to a better effort score.

Average handle time

Average handle time (AHT) is the total time an agent spends on a live interaction, including hold time and wrap-up work.

It measures efficiency without accounting for quality, so it works best alongside CSAT and FCR.

Targets by channel:

• Phone: 6 to 8 minutes
• Live chat: 5 to 7 minutes

As automation rate rises, AHT on human-handled tickets typically drops — AI absorbs the simple volume and leaves agents with a shorter, more focused queue. Ecom Lab data shows that at 50%+ automation, AI does the equivalent work of 6.3 full-time agents while the human team at that tier averages just 3 people. Those agents handle 29% more tickets per month and spend more time on the complex issues that actually require judgment.

Time to resolution

Time to resolution (TTR) is the total time from when a ticket opens to when it fully closes.

Unlike FRT, which measures only the first reply, TTR captures the entire support interaction. For a closer look at this metric and how to reduce it, see Resolution Time: What It Is and How to Reduce It.

General targets by complexity:

• Simple inquiries: Under 24 hours
• Complex issues: 24 to 72 hours
• Technical problems: 3 to 5 business days

For brands with AI in the mix, Ecom Lab data gives channel-level baselines for tickets that require human involvement. Contact form handoffs resolve in a median of 36 hours with a 42% abandonment rate. Email handoffs resolve in 32 hours and abandon 30% of the time. Chat resolves in 8 hours and abandons 13%, because real-time pressure forces faster responses.

These aren't just benchmarks to optimize — they're the cost of every ticket that doesn't resolve end-to-end.

Service level agreement adherence

A service level agreement (SLA) is a defined commitment to respond to or resolve tickets within a set timeframe.

SLA adherence measures the percentage of tickets where your team meets that commitment. The industry benchmark is 90 to 95 percent compliance. For the tactics that make hitting those commitments repeatable, see SLA Best Practices for Effective Support Ticket Management.

How to benchmark customer service performance

Benchmarking works best as a structured process, not a one-time audit. These six steps take you from identifying what to measure to building a plan for improvement.

Step 1: Define your goals

Start by deciding what you want to improve and why.

Are customers complaining about slow responses? Are agents spending too long on simple tickets? Tying your benchmarking effort to a specific business problem keeps the process focused and the results actionable. Limit your initial scope to three to five metrics — tracking everything at once makes it harder to act on what you find.

Step 2: Choose the right metrics

Pick metrics that match the problem you identified in step one.

If customers are frustrated by how long it takes to get help, FRT and TTR are your starting points. If satisfaction scores are slipping, CSAT and CES will tell you more. Match the metric to the pain point.

If you have any automation running, add AI Resolution Rate to your list. The median brand sits at 45%; the top quartile is at 65%. A gap between your rate and the top quartile almost always comes down to one of four things: limited intent coverage, insufficient action authority (AI can't issue refunds or apply discounts), missing system integrations, or an escalation policy that's routing too much to humans by default.

Step 3: Source your benchmarks

You need external data to compare against. Reliable sources include:

• Industry reports: Published by customer service associations and research firms.
• Helpdesk platforms: Many publish benchmark data from their customer base, including Ecom Lab.
• Peer networks: Non-competing brands in adjacent markets often share performance data.
• Customer surveys: Direct feedback on what customers expect from your support.

One critical caveat: filter by your specific vertical, not just "ecommerce." Ticket volume per 100 orders varies nearly as much as response time. Electronics brands generate about 46 support tickets per 100 orders. Food & Beverages brands generate about 20. If you're in a high-ticket-volume vertical, that's your baseline — not a problem to fix.

Step 4: Measure your current performance

Pull at least three months of data from your helpdesk to establish a reliable baseline.

Shorter windows get skewed by seasonal spikes or one-off events. Make sure you're measuring each metric the same way across all channels so the data is consistent. How to Evaluate the Effectiveness & Impact of Your Customer Service Team is a good companion resource for this step.

Step 5: Identify and diagnose gaps

Compare your numbers to the benchmarks you sourced.

Look for patterns. Are certain channels consistently slower? Do specific ticket types take longer to resolve? The goal is to understand why the gap exists, not just that it does.

Step 6: Set targets and build an action plan

Use your gap analysis to set incremental targets.

If your email FRT is 48 hours and the benchmark is 24, aim for 36 hours first. Assign ownership, set a timeline, and schedule a review date. Benchmarking only drives improvement when it leads to a concrete next step.

What good benchmarking looks like in practice

Benchmarking changes how your team operates day to day. Agents know what "good" looks like and can measure their own progress against it. Managers can identify coaching opportunities using real data rather than observation alone.

For ecommerce brands specifically, the operational benefits compound over time:

• Lower cost per ticket: Efficient workflows reduce the time and resources needed to resolve each issue.
• Higher repeat purchase rates: Customers who get fast, easy support are more likely to buy again.
• Reduced cart abandonment: Quick answers to pre-purchase questions keep shoppers moving toward checkout. See Reduce Cart Abandonment: Proven, Data-Backed Strategies.
• Stronger team accountability: Clear targets give agents and managers a shared definition of success.

The financial picture from the Ecom Lab report Most Brands Are Overpaying for Support is concrete. Even at the lowest automation tier, brands net $73K per year after platform costs.

Nearly 1 in 4 brands (23.5%) reduced their support team after enabling Gorgias AI Agent. Of those, 51% achieved all three outcomes at once: fewer people, same ticket volume, same or more revenue. Brands that reduced by at least one person saw each remaining agent handle 29% more tickets per month while revenue grew 22%.

The adoption gap matters too. Only about 1 in 5 ecommerce brands has deployed AI in customer-facing support today. Brands at near-zero automation average 736-minute response times. Brands at 30%+ automation average 80 minutes.

That's not an incremental improvement. It's a structural shift.

Improve your benchmarks with Gorgias

Benchmarks tell you where to focus. The right tools help you get there.

Gorgias gives ecommerce support teams a unified view of every customer conversation, with built-in reporting that tracks the metrics that matter most. AI Agent resolves routine tickets automatically — across email, chat, and SMS — so your team spends less time on repetitive requests and more time on work that requires a human touch.

Book a demo to see how Gorgias helps ecommerce brands hit and exceed their customer service benchmarks.

{{lead-magnet-2}}

TL;DR:

• AI Agent only answers from sources you connect. Help Center articles, your website, uploaded documents, and Shopify data. Nothing outside of that.
• Every behavior is configurable. What AI Agent says, how it sounds, which topics it handles, and which ones it always passes to a human are all settings you control in plain language.
• You decide what AI Agent never touches. Handover topics and exclusions let you route specific ticket types directly to your human team, every time, regardless of what AI Agent knows.
• Configuration is ongoing, not a one-time setup. A monthly review of flagged tickets and Guidance Opportunities keeps AI Agent calibrated as your policies and products change.

How much of AI Agent can you actually control? The answer is more than most people expect.

AI Agent has several distinct control layers — what it knows, how it speaks, which topics it handles, which ones it passes to your team, and what actions it can take on a customer's behalf. Each layer is configurable in plain language, directly inside your Gorgias settings.

This article walks through each control, what it does, and what good configuration looks like in practice. 

{{lead-magnet-1}}

How to tell AI Agent how to behave

AI Agent allows you to configure four inputs: Guidance, tone of voice and language, handover topics and exclusions, and knowledge sources.

Each one controls a different dimension of AI Agent's behavior. Together, they determine what AI Agent knows, how it communicates, what it will and won't handle, and where it draws its answers from.

Guidance: your policies and instructions, in plain language

Guidance is the highest-priority input in AI Agent's knowledge hierarchy. When information exists in both a Guidance and another source — like a Help Center article or a connected URL — AI Agent always follows the Guidance first.

It's designed for rules and behavioral instructions, not just answers. Good examples of what belongs in a Guidance:

• "Do not offer a discount without a human agent approving it first."
• "If a customer mentions a legal dispute or uses the word 'lawyer,' hand over immediately."
• "Always confirm the order number and email address before processing a return."
• "Do not provide medical advice or interpret how our products may affect specific health conditions."

You can add up to 100 Guidance, and each one can be as specific as your policies require.

Be specific. Vague instructions produce vague behavior. "Handle returns politely" leaves AI Agent to interpret what that means. A more useful Guidance looks like this: 

"For return requests: confirm the order number, check whether the item is within the 30-day return window, and if eligible, send the prepaid return label link. If outside the window, explain the policy and hand over to a human agent."

Nothing left to chance. When writing Guidance, focus on the situation and the desired outcome.

Keep it current with Guidance Opportunities. Over time, AI Agent detects recurring questions it could not confidently answer and surfaces them as suggested new Guidance in your dashboard. Review, edit to match your policies, then approve or dismiss. Nothing is added without your sign-off.

Read more: How to write Guidance with the “when, if, then” framework

Tone of voice: how AI Agent sounds

Tone of voice is a separate setting from Guidance and knowledge. It controls how AI Agent communicates, the register, the warmth, the vocabulary, rather than what it says or what it does.

You have four options:

• Friendly. Warm, approachable, and casual. Empathetic and encouraging.
• Professional. Formal, respectful, and focused. Business-like without being cold.
• Sophisticated. Refined and considered. Complex vocabulary, subtle nuance.
• Custom. Your instructions, in plain language, up to 5,000 characters.

Custom is where brands with specific voice guidelines should spend time. Describe your communication style the same way you'd brief a copywriter: words and phrases that fit your brand, what to avoid, how to handle emotionally charged conversations. Emoji usage and phrases your team always or never uses can all go here.

Two things to keep separate as you configure this:

• Tone vs. Guidance: Tone of voice tells AI Agent how to write. Instructions about specific situations, like "when a customer asks about returns, do X," belong in Guidance, not here. Mixing the two produces inconsistent results.
• Tone vs. Language: AI Agent supports over 80 languages and automatically replies in the language a customer writes in. Language is its own dedicated setting. If you've been using the tone of voice field to control language, switch, as it can conflict with how responses are generated.

Tone usually takes a few iterations to get right. Use the test conversation feature, found under AI Agent > Test, to simulate customer conversations and verify how AI Agent sounds before it goes live.

Learn more: Customize AI Agent's tone of voice

Handover topics and exclusions: what AI Agent should never touch

Handover topics and exclusions give you explicit control over which conversations AI Agent handles and which go straight to your human team. Both live in the Handover and Exclusion section of your AI Agent settings.

Handover topics are subjects AI Agent is always instructed to pass to a human, even if it has relevant information. You add them in plain language. Some examples worth considering for most stores:

• Legal disputes or any mention of legal action
• Medical questions or references to health conditions
• Fraud-related language or chargeback mentions
• Any request that requires a judgment call outside of written policy

Exclusions work differently. Using the "Prevent AI Agent from answering" Rule, you can tell AI Agent to ignore certain tickets entirely, based on a tag, a sender's email address, or specific words in the message. These tickets stay in the queue for your human agents without AI Agent touching them.

You can also toggle whether AI Agent tells the customer it is handing them over, or does so silently. Both options are available in the same section.

Note that AI Agent hands over automatically in some situations regardless of configuration, including when it cannot find a relevant answer, when a response does not pass its internal quality check, and when a customer asks to speak to a human.

Read more: Customize how AI Agent hands over to your team

Knowledge sources: what AI Agent draws from

AI Agent's answers are grounded entirely in the sources you connect. It won't speculate beyond them, and if it cannot find a relevant answer, it hands over instead of guessing.

The sources you can connect are:

• Help Center articles. Your existing documentation and policy pages.
• Your store website. Crawled automatically when you connect it.
• Specific public URLs. Individual pages you want AI Agent to reference.
• Uploaded documents. PDFs, Word docs, or spreadsheets with product or policy information.
• Shopify data. Order history, customer profiles, product catalog, and fulfilment information.

Start by connecting what you already have. Gaps will surface quickly through handovers in the inbox or through Guidance Opportunities flagging unanswered questions. Adding a URL or uploading a document is usually faster than writing a Guidance for every scenario.

Learn more: Onboard AI Agent with knowledge sources

Which channels should AI Agent be active on?

AI Agent runs on email, chat, and SMS. None are on by default. You enable each one manually. Turning a channel off doesn’t affect your configuration.

Email is where most brands start. AI Agent handles incoming tickets, filters spam, and pulls from your knowledge sources and Shopify data to reply with context. Anything it cannot resolve gets handed over with the full conversation attached.

Chat is faster and more transactional. Customers tend to ask shorter, more immediate questions — order status, return eligibility, quick policy checks. AI Agent adapts automatically, writing shorter and more conversational replies on chat than it would on email.

SMS requires a separate add-on subscription. It is the most tone-sensitive channel, so configure your tone of voice carefully and test thoroughly before enabling AI Agent here.

There is no required order for activation. Most brands pick the channel with the highest ticket volume and clearest policies, then expand. Switching a channel off is a simple toggle. Nothing gets deleted.

How to test AI Agent before it goes live

Before enabling AI Agent on any channel, test it first. Gorgias has a built-in test conversation feature that lets you simulate customer interactions without affecting real tickets, reporting, or customers. Go to AI Agent > Test.

Step 1: Start with your hardest tickets. Skip the generic questions. Test the scenarios that made you hesitant in the first place — complex product questions, return edge cases, sensitive topics on your handover list. If AI Agent handles these well, you can activate with confidence.

Step 2: Test each channel separately. AI Agent adapts its response style by channel. Shorter on chat, more detailed on email. A response that reads well on email may feel too long on chat, so configure the channel in the test settings to match the one you are evaluating.

Step 3: Check your handovers. Send a message containing the language or scenario you want escalated. Confirm AI Agent passes it to your team rather than attempting a response. Do this for every topic on your handover and exclusion list.

Step 4: Test your tone. Run several conversations before settling on your tone configuration. Try emotionally charged messages, not just neutral ones. Tone usually takes a few rounds to get right.

Test conversations do not count toward your automated interaction billing.

What should AI Agent escalate to a human?

Some tickets should never be handled by AI. Not because AI Agent cannot generate a response, but because the situation calls for a human regardless.

There are two types of escalation to understand: the ones AI Agent does automatically, and the ones you configure yourself.

Automatic escalations. These are built in and can’t be turned off. AI Agent hands over automatically when it lacks confidence in an answer, when it can’t find relevant content in its knowledge sources, when it detects customer anger or frustration, and when a customer explicitly asks to speak to a human. 

Every response also passes through an internal QA step — a second AI model measures confidence, and if the response does not meet the threshold, it is not sent.

Configured escalations. These are yours to define. Two tools handle this.

Handover topics tell AI Agent to always pass a conversation to a human on a specific subject, even if it has relevant information. Add them in plain language in your AI Agent settings. Good candidates for most stores:

• Legal disputes or any mention of legal action
• Medical questions or references to health conditions
• Fraud-related language or chargeback mentions
• Any complaint requiring a judgment call outside of written policy

Exclusions go further. Using the "Prevent AI Agent from answering" Rule, you can tell AI Agent to ignore certain tickets entirely, based on tags, sender email addresses, or specific words in the message. These tickets never get touched by AI Agent at all.

Route escalations to the right place. When AI Agent hands over, configure which tag, team, or queue the ticket routes to. A return escalation goes to fulfilment. A billing dispute goes to a senior agent. Escalated tickets should never land in a generic inbox.

Learn more: Security and privacy FAQ for Gorgias AI Agent

How to review what AI Agent did and improve it over time

Configuration is not a one-time task. The brands who get the most out of AI Agent check in regularly, flag what is not working, and update its knowledge as policies and products change. Here is a simple review checklist to run through on a monthly basis, or after any major policy or product update.

• Read flagged tickets. Open your inbox and filter for AI Agent conversations labeled "Automated." Check the full thread and which knowledge sources were used. If a response looks off, the source is visible.
• Rate what you see. Mark tickets as good, ok, or bad. For negative ratings, select a reason — wrong answer, wrong tone, should have escalated. This feeds directly into Guidance Opportunities.
• Check the Intents page. Go to AI Agent > Intents to see which topics AI Agent is handling and where handovers are clustering. Recurring handovers on the same topic mean a Guidance is missing.
• Review Guidance Opportunities. Open your dashboard and work through pending suggestions. Edit each one to match your policies, then approve or dismiss. Nothing is added without your sign-off.
• Update your Guidance. Any time a policy changes, a new product launches, or a seasonal shift affects your return or shipping terms, review relevant Guidance and update accordingly.

Learn more: Continuously improve AI Agent with Opportunities (Beta)

Considerations for sensitive and regulated categories: what to configure differently

Most of the configuration covered in this article applies to every store. This section is for brands in categories where the stakes of an incorrect or out-of-scope response are higher. Skim to your vertical and take what applies.

Health and wellness

• Block medical advice. Write a Guidance that explicitly instructs AI Agent not to interpret how products may affect specific health conditions, and to hand over if a customer asks.
• Add a second layer. Add medical questions as a handover topic in addition to your Guidance.
• Upload usage documentation. For products with complex instructions, upload those documents as a knowledge source so AI Agent answers accurately without improvising.

Food and beverage

• Keep allergen content current. Make sure Help Center articles on ingredients and allergens are accurate, up to date, and connected as a knowledge source.
• Set clear boundaries in Guidance. Specify exactly what AI Agent can and cannot say about allergens and ingredients.
• Use exclusions for ambiguous formulations. Where product ingredients change frequently, exclusions are safer than handovers. AI Agent will not touch those tickets at all.

Subscriptions

• Gate sensitive actions with conditions. Only allow cancellations or pauses after a minimum subscription duration.
• Require customer confirmation. Enable confirmation for any action that modifies a subscription before it executes.
• Add billing topics to your handover list. Billing disputes and chargeback mentions should always go to a human.

Now, let's set it up for your store

You now know exactly what you can control, and there is more of it than most brands expect. The next step is seeing it configured for your store specifically.

Book a demo →

{{lead-magnet-2}}

TL;DR:

• Intercom's pricing structure jumps sharply between tiers, making it expensive for growing ecommerce brands
• Gorgias is built for Shopify with native order management, AI automation, and proactive sales tools in one platform
• Budget-friendly options like Crisp and Tidio offer free tiers for brands just getting started
• Enterprise platforms like Zendesk and Salesforce Service Cloud offer scale but require significant setup and investment
• The right choice depends on your platform, ticket volume, and whether you need support, sales, or both

Intercom built its reputation as a customer messaging tool for SaaS companies. As it shifted toward enterprise, many ecommerce brands found themselves paying more for features they didn't need — and missing the ones they did. This guide covers the 15 best Intercom alternatives, evaluated specifically for ecommerce brands on Shopify and beyond.

Why brands switch from Intercom

Intercom is a customer messaging platform that combines live chat, email, and automation for support and sales. It works well for SaaS companies, but ecommerce brands often run into friction fast.

The biggest issue is pricing. The jump from the Essential plan ($74 USD per month) to the Advanced plan ($395 per month) is steep, and many features — like custom bots and product tours — cost extra on top of that. For a growing brand managing hundreds of tickets a week, the total cost adds up quickly.

Beyond price, Intercom was not built with ecommerce in mind. There is no native Shopify integration, no order management inside conversations, and limited automation for common requests like "Where is my order?" That gap pushes many brands to look for tools designed around how online retail actually works.

Best Intercom alternatives at a glance

The 15 best Intercom alternatives for 2026

1. Gorgias

Gorgias is a customer experience (CX) platform built specifically for ecommerce brands. It connects your helpdesk directly to Shopify, so agents can view orders, issue refunds, and update subscriptions without leaving the conversation.

The AI Agent handles up to 60% of incoming tickets automatically — shipping questions, return requests, order status — across email, chat, and SMS. The Shopping Assistant goes further, proactively engaging shoppers on product pages to help them find the right item and complete their purchase.

Gorgias is the strongest Intercom alternative for Shopify brands that want support and sales in one place.

Key features:

• Native Shopify integration with real-time order data
• AI Agent that resolves tickets autonomously
• Shopping Assistant for proactive product recommendations
• 100+ ecommerce app integrations
• Unified inbox for email, chat, SMS, voice, and social

Pricing:

• Starter: $10/month (50 tickets)
• Basic: $60/month (300 tickets)
• Pro: $360/month (2,000 tickets)
• Advanced: $900/month (5,000 tickets)

{{lead-magnet-1}}

2. Zendesk

Zendesk is an enterprise helpdesk platform that handles high ticket volumes across email, chat, voice, and social. Its reporting tools are among the most advanced available, and its marketplace includes hundreds of integrations.

It is a strong fit for large teams with complex workflows. For smaller ecommerce brands, the setup time and per-agent pricing can be a barrier.

Pricing: Suite Team starts at $55/agent/month

3. Freshdesk

Freshdesk is a helpdesk platform with a free tier for up to 10 agents, making it one of the most accessible Intercom alternatives for teams on a tight budget. It includes ticketing, automation, and a knowledge base across its plans.

Its ecommerce integrations are less deep than purpose-built tools, but it covers the basics well for brands managing moderate ticket volume.

Pricing: Free for up to 10 agents; paid plans from $15/agent/month

4. Help Scout

Help Scout is a shared inbox tool that makes email support feel personal and organized. Agents work in a clean interface that looks like a regular email client, which reduces the learning curve significantly.

When comparing Help Scout vs Intercom, Help Scout wins on simplicity. It suits brands that rely heavily on email and want a tool their team can use on day one.

Pricing: Standard plan starts at $25/user/month

5. Kustomer

Kustomer is a CRM-first platform that organizes every customer interaction into a single timeline. Agents see the full history of a shopper's orders, conversations, and behavior in one view.

It is best suited for enterprise brands that need deep personalization and are willing to invest in a more complex setup. Pricing reflects that positioning.

Pricing: Enterprise plan starts at $89/user/month

6. Drift

Drift is a conversational marketing platform built for B2B sales teams. Its chatbots qualify leads, book meetings, and route prospects to sales reps automatically.

When comparing Drift vs Intercom, both target sales use cases — but neither is built for ecommerce. Drift's pricing starts at $2,500/month, making it one of the most expensive options on this list.

Pricing: Premium starts at $2,500/month; advanced plans are custom

7. Tidio

Tidio is a live chat and chatbot tool aimed at small ecommerce businesses. Its free plan covers up to 50 conversations per month, and its paid plans are affordable for brands just scaling up.

The chatbot builder is easy to use without technical knowledge. It lacks the deep integrations and automation power of more advanced platforms, but it is a solid starting point.

Pricing: Free up to 50 conversations; Starter from $29/month

8. Crisp

Crisp is a multichannel messaging platform with a generous free plan that includes two seats. It brings together live chat, email, and social messaging in one shared inbox.

It is a practical, cheaper alternative to Intercom for startups and small teams. Paid plans unlock chatbots and more advanced automation.

Pricing: Free for two seats; Pro from $25/month

9. Zoho Desk

Zoho Desk is the support module within the Zoho software suite. Its biggest advantage is how tightly it connects with Zoho CRM, giving sales and support teams a shared view of every shopper.

If your brand already uses Zoho products, Desk is a natural fit. If you don't, the value of the integration is less compelling.

Pricing: Standard plan starts at $14/agent/month

10. Front

Front is a collaborative inbox tool that brings email, SMS, and social channels into one shared workspace. Teams can assign conversations, leave internal comments, and track response times without switching tools.

Front is strong for collaboration but lighter on traditional helpdesk features like ticket routing and automation rules. It suits teams that manage a high volume of email and need better internal coordination.

Pricing: Starter plan starts at $19/seat/month

11. Gladly

Gladly organizes all customer communication into a single, ongoing conversation thread — no ticket numbers, no channel silos. Agents always see the full picture, regardless of where the shopper reached out.

This model works well for premium brands that prioritize a personal, high-touch experience. The price point reflects that focus.

Pricing: Hero plan starts at $150/agent/month

12. LiveAgent

LiveAgent is an all-in-one helpdesk that includes live chat, email, a call center, and a knowledge base in a single platform. It is one of the most affordable options with a broad feature set.

The interface feels dated compared to newer tools, but the functionality is solid for brands that want everything in one place without a high price tag.

Pricing: Small plan starts at $9/agent/month

13. HubSpot Service Hub

HubSpot Service Hub is the customer service layer of the HubSpot platform. It connects directly to HubSpot CRM, giving support teams full visibility into a shopper's marketing and sales history.

For brands already running on HubSpot, it is a logical extension. For those who aren't, adopting the full suite just for support is a significant commitment.

Pricing: Starter from $20/month for two users

14. Salesforce Service Cloud

Salesforce Service Cloud is one of the most powerful and customizable support platforms available. It handles complex workflows, advanced automation, and deep reporting at enterprise scale.

The tradeoff is complexity. Implementation takes time, requires technical resources, and the cost scales quickly. It is best suited for large brands with dedicated operations teams.

Pricing: Starter from $25/user/month; Enterprise from $165/user/month

15. Groove

Groove is a simple helpdesk for small teams that have outgrown shared Gmail inboxes. It includes a shared inbox, knowledge base, and basic reporting without the overhead of a larger platform.

It is a practical, no-frills option for brands with low ticket volume and straightforward support needs.

Pricing: Standard plan starts at $16/user/month

How to choose an Intercom alternative for ecommerce

The right platform depends on three things: your ecommerce stack, your ticket volume, and what you need the tool to do beyond answering questions.

Start with your platform. If you run on Shopify, you need a tool that connects natively — not through a workaround. Native integration means agents can see order details, edit shipments, and process returns without switching tabs.

Then think about scale. Per-agent pricing works well for small teams but gets expensive fast. Ticket-based pricing, like Gorgias uses, scales more predictably as your volume grows.

Finally, decide whether you need support only or support and sales. Some platforms on this list handle tickets well. Others — like Gorgias — are built to drive revenue through conversations, not just resolve them.

Key questions to ask any vendor:

• Does it integrate natively with your ecommerce platform
• What is included in the base plan versus add-ons
• How long does onboarding typically take
• How does pricing change as your ticket volume grows
• What automation is available without custom development

Next steps to evaluate Intercom alternatives

Most platforms offer a free trial or starter plan. Use it. A week of real usage tells you more than any feature comparison chart.

For Shopify brands that need more than Intercom offers, Gorgias is worth a closer look. You get native order management, 60% automation coverage, and no per-agent pricing. Start a free trial to see it on your actual workflows.

TL;DR:

• AI Agent is priced per resolved interaction, not per seat or per message. You only pay when the AI fully resolves a conversation on its own.
• Most plans are $0.90 per resolved interaction. Starter plans begin at $1. Plans include 90 to 2,500+ automated interactions per month.
• If you go over your plan, overage fees apply per additional interaction. Rates vary by tier and are lower on annual plans.
• Your automation rate emerges from usage over time. Start by estimating your ticket volume and pick an interaction allotment that fits.
• AI Agent runs on email, chat, and SMS, and includes tone of voice customization, Actions, multi-language support, vision, and performance reporting.

If you're wondering what it costs to add AI Agent to your Helpdesk, you're in the right place. This article walks through how pricing works, what counts as a billable interaction, and how to think about the investment before talking to anyone on our team.

The good news: there are no seat fees, no per-message charges, and no token-based billing. You pay for conversations your AI actually resolves. If you've looked into other AI tools for customer support and found the pricing models confusing or hard to predict, Gorgias AI Agent works differently.

What is a billable interaction?

A billable interaction is counted when the AI resolves a customer conversation entirely on its own. The customer asks something, the AI handles it, the conversation closes. That's one interaction.

If the AI can't fully resolve a conversation and hands it to a human agent, that ticket shifts over to your regular Helpdesk plan. It becomes a standard resolved ticket. You're not charged for both.

A few things that don't count as billable interactions:

• Emails that come in but no one replies to
• Spam or filtered messages
• Conversations resolved by a human agent

This matters most for brands coming from seat-based tools. With Gorgias, your whole team can work in the platform. Agent seats are unlimited. Pricing scales with what your AI is actually doing, not with how many people have access.

Understand the difference between seat-based vs. usage-based pricing.

How AI Agent plans work

AI Agent is an add-on to your Gorgias Helpdesk plan. The two are priced separately but work together. Your Helpdesk plan covers all the conversations your human agents resolve. Your AI Agent plan covers the interactions the AI resolves on its own.

When you choose a plan, you select how many automated interactions you want included per month. Depending on your plan, that ranges from 90 to 2,500+ interactions, with custom interaction numbers available for enterprise. You can see the full breakdown on the Gorgias pricing page.

Each resolved conversation costs $0.90 on most plans. Starter plans begin at $1 per resolved conversation. You only pay for fully automated interactions, meaning conversations the AI handles from start to finish without a human stepping in.

Choosing the right plan

The main input is your average monthly ticket volume. From there, you estimate how many of those conversations AI could realistically handle on its own.

Order status updates, return requests, and shipping questions tend to be the highest-volume ticket types AI resolves well. AI Agent actions shows the full range of what it can handle, which makes it easier to estimate your starting number.

Your actual automation rate, meaning the share of total tickets the AI ends up resolving, emerges from usage over time. Most brands start with their most repetitive ticket types and expand from there as they see results.

Related: Which Gorgias plan should you choose?

What happens if you go over your plan

You're charged an overage fee for each additional automated interaction if you exceed your plan's baseline in a given month. The exact rate depends on your plan tier and whether you're on a monthly or annual subscription.

Generally, the higher your plan tier, the lower your overage rate. Annual plans also carry lower overage rates than monthly plans. So if you're regularly going over, upgrading to a higher tier or switching to annual often works out cheaper than paying overage fees month after month.

If you're on a Support + Shopping Assistant plan, the overage rate is $1.50 per interaction across all paid tiers. If you're on a Support-only plan, rates range from $1.00 to $2.00 per interaction on monthly plans, and $0.83 to $1.67 on annual plans, depending on your tier.

For seasonal businesses, forecasting your customer service volume before peak periods is the best way to choose the right plan size and avoid unexpected fees.

How to think about the cost

At $0.90 per resolved interaction on most plans, each AI resolution costs less than a human agent handling the same ticket. Once you know what a human-resolved ticket costs your business, the comparison becomes straightforward.

For brands building an internal case for the investment, how to pitch AI Agent to your boss covers the ROI framing in detail. 

To see what results look like in practice, how 10 brands transformed customer support into revenue has real ecommerce examples.

What's included with AI Agent

AI Agent comes with everything you need to set it up, customize it, and improve it over time:

• Knowledge training — AI Agent learns from your Shopify data, store website, Help Center articles, URLs, documents, and custom guidance. The more content it has, the more accurately it responds.
• Tone of voice — set instructions for how AI Agent sounds, whether that's professional, friendly, or something else, and it stays consistent across every conversation.
• Actions — connect AI Agent to your other tools so it can complete tasks like cancelling an order, processing a return, or modifying a subscription without a human stepping in. See what AI Agent can do.
• Multi-language support — AI Agent detects the language a customer writes in and replies in the same language automatically.
• Vision — AI Agent can read and understand images, so it can handle tickets where customers share photos of damaged items or order issues.
• Performance reporting — track automation rate, CSAT, first-response time, and ticket topics directly in the dashboard.
• Testing — preview how AI Agent responds to real customer questions before going live or after making changes.
• Handover to humans — AI Agent automatically passes conversations to your team when it lacks confidence, detects frustration, or encounters a topic you've marked for human handling.

Learn more: Gorgias AI Agent guardrails: What they are and how to configure them

Curious what AI Agent would automate for your store?

The best way to get a sense of what AI Agent will cost is to look at your own ticket volume and the types of questions your customers ask most. From there, the right plan becomes much clearer.

If you want to talk through the numbers with someone from our team, book a demo and we'll walk through it with you.

If you'd rather keep exploring first, here are a few good next reads:

• AI Agent keeps getting smarter — the latest performance data across the Gorgias customer base
• 10 must-know AI Agent use cases — the ticket types brands are automating today
• How 10 brands transformed customer support into revenue — real ecommerce results
• How to customize AI Agent with 7 brand voice examples — what the experience looks like for customers

{{lead-magnet-2}}