TLDR: https://github.com/xarg/pghoard-k8s

‍

This is a small tutorial on how to do incremental backups using pghoard for your PostgreSQL (I assume you’re running everything in Kubernetes). This is intended to help people to get started faster and not waste time finding the right dependencies, etc..

‍

pghoard is a PostgreSQL backup daemon that incrementally backups your files on a object storage (S3, Google Cloud Storage, etc..).
For this tutorial what we’re trying to achieve is to upload our PostgreSQL to S3.

‍

First, let’s create our docker image (we’re using the alpine:3.4 image cause it’s small):

FROM alpine:3.4

ENV REPLICA_USER "replica"
ENV REPLICA_PASSWORD "replica"

RUN apk add --no-cache \
   bash \
   build-base \        
   python3 \
   python3-dev \
   ca-certificates \
   postgresql \
   postgresql-dev \
   libffi-dev \
   snappy-dev
RUN python3 -m ensurepip && \
   rm -r /usr/lib/python*/ensurepip && \
   pip3 install --upgrade pip setuptools && \
   rm -r /root/.cache && \
   pip3 install boto pghoard


COPY pghoard.json /pghoard.json.template
COPY pghoard.sh /

CMD /pghoard.sh

‍

‍

REPLICA_USER and REPLICA_PASSWORD env vars will be replaced later in your Kubernetes conf by whatever your config is in production, I use those values to test locally using docker-compose.

‍

The config pghoard.json which tells where to get your data from and where to upload it and how:

‍

{
   "backup_location": "/data",
   "backup_sites": {
       "default": {
           "active_backup_mode": "pg_receivexlog",
           "basebackup_count": 2,
           "basebackup_interval_hours": 24,
           "nodes": [
               {
                   "host": "YOUR-PG-HOST",
                   "port": 5432,
                   "user": "replica",
                   "password": "replica",
                   "application_name": "pghoard"
               }
           ],
           "object_storage": {
               "aws_access_key_id": "REPLACE",
               "aws_secret_access_key": "REPLACE",
               "bucket_name": "REPLACE",
               "region": "us-east-1",
               "storage_type": "s3"
           },
           "pg_bin_directory": "/usr/bin"
       }
   },
   "http_address": "127.0.0.1",
   "http_port": 16000,
   "log_level": "INFO",
   "syslog": false,
   "syslog_address": "/dev/log",
   "syslog_facility": "local2"
}

‍

Obviously replace the values above with your own. And read pghoard docs for more config explanation.

‍

Note: Make sure you have enough space in your /data; use a Google Persistent Volume if you DB is very big.

‍

Launch script which does 2 things:

1. Replaces our ENV variables with the right username and password for our replication (make sure you have enough connections for your replica user)
2. Launches the pghoard daemon.

‍

‍

#!/usr/bin/env bash

set -e

if [ -n "$TESTING" ]; then
   echo "Not running backup when testing"
   exit 0
fi

cat /pghoard.json.template | sed "s/\"password\": \"replica\"/\"password\": \"${REPLICA_PASSWORD}\"/" | sed "s/\"user\": \"replica\"/\"password\": \"${REPLICA_USER}\"/" > /pghoard.json
pghoard --config /pghoard.json

‍

Once you build and upload your image to gcr.io you’ll need a replication controller to start your pghoard daemon pod:

‍

apiVersion: v1
kind: ReplicationController
metadata:
 name: pghoard
spec:
 replicas: 1
 selector:
   app: pghoard
 template:
   metadata:
     labels:
       app: pghoard
   spec:
       containers:
       - name: pghoard
         env:
           - name: REPLICA_USER
             value: "replicant"
           - name: REPLICA_PASSWORD
             value: "The tortoise lays on its back, its belly baking in the hot sun, beating its legs trying to turn itself over. But it can't. Not with out your help. But you're not helping."
         image: gcr.io/your-project/pghoard:latest

‍

The reason I use a replication controller is because I want the pod to restart if it fails, if a simple pod is used it will stay dead and you’ll not have backups.

‍

Future to do:

• Monitoring (are you backups actually done? if not, do you receive a notification?)
• Stats collection.
• Encryption of backups locally and then uploaded to the cloud (this is supported by pghoard).

Hope it helps, stay safe and sleep well at night.

‍

Again, repo with the above: https://github.com/xarg/pghoard-k8s

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:

• Your support inbox is full of marketing gold. CX insights can sharpen messaging and inspire high-impact campaigns.
• Ticket data unlocks smarter segmentation. Use support interactions to build more relevant, behavior-based audiences.
• Chat campaigns work better with CX insights. Tackle objections in real time and lift conversions with proactive messages.
• Use objection data to reduce drop-offs. Identify common blockers and address them in product pages, ads, and chat prompts.
• Help Center stats guide better content. Turn top-searched questions into FAQs, landing pages, and ad copy.

Today’s best marketing starts with your customers.

According to Forrester’s 2024 research, “Customer-obsessed organizations reported 41% faster revenue growth, 49% faster profit growth, and 51% better customer retention than those at non-customer-obsessed organizations.”

Support teams interact with hundreds or thousands of customers every week, collecting valuable insights in the process. This voice of the customer (VOC) data is a goldmine for marketers, but it too often stays siloed among CX teams.

Ahead, we’ll break down how ecommerce brands can tap into CX insights to drive better marketing.

5 ways to use CX data to improve marketing

CX can play a crucial role in driving growth, but many brands aren’t leveraging it for marketing insights yet.

When connected to marketing, CX becomes a proactive engine that fuels better segmentation, sharper messaging, smarter campaigns, and more personalized content. 

Support functions collect objections, complaints, compliments, and pre-purchase questions. When you capture and apply those insights, your marketing can target the precise roadblocks—and key sales differentiators—customers care about.

Here’s how to turn CX insights into a high-impact marketing strategy, with real examples from brands using Gorgias.

1. Leverage ticket insights to improve messaging
2. Segment customers based on support interactions
3. Launch more targeted chat campaigns
4. Reduce drop-offs and abandoned carts
5. Monitor Help Center and Dashboard stats to craft smarter content

1) Leverage ticket insights to improve messaging 

When you want to sharpen your brand messaging, there’s no better place to look than your support inbox. Your support inbox is a rich resource full of information specific to your brand and your customers. 

Tools like Gorgias Ticket Insights help surface recurring themes, top questions, and friction points across all conversations. By analyzing these patterns, marketers can identify the exact words customers use to describe problems, questions, or product feedback and then reflect that language across ads, landing pages, and emails.

How to implement 

Spikes in tickets around specific topics (sizing, shipping timelines, and materials, for example) are insights marketers can use to update and improve corresponding content. 

This can increase confidence and conversion on key pages. 

By incorporating the same terminology and phrasing customers use in support conversations, brands can also increase resonance across ads, emails, and social media. Messaging that mirrors the customer’s language builds trust and helps audiences feel understood. 

Ask your CX team 💬 What product issues or themes have emerged this quarter?

For example, cordless heating cushion brand Stoov® used Ticket Fields in Gorgias to understand and resolve a ticket spike. By figuring out that some customers were dissatisfied with the battery life of its core product offering, the team was able to add an optional upsell. For €20, shoppers now have the option to purchase a larger battery. 

The results were meaningful: the brand saw 50% of customers opt for this battery, resulting in a 10% increase in average order value (AOV). And while the team saw a significant increase in revenue, they saw no increase in support ticket volume. 

2) Segment customers based on support interactions

Most marketers rely on transactional data—like past purchases or time since last order—to build audience segments. But support data reveals a whole new layer of context: behavior, concerns, sentiment, and urgency.

Tools like Gorgias’s Ticket Insights and Ticket Fields allow CX teams to customize different properties attached to tickets. Agents can fill these out to capture data more accurately. 

Here’s how these types of tools work: tickets come with a mandatory field for return reasons, product feedback, contact reason, etc. Before the agent closes the ticket, they use a dropdown menu to fill out the ticket field. 

How to implement 

Studying support interactions helps answer key questions around why customers are getting in touch. This data can provide marketing teams with a way to build smarter segments for campaigns or personalized journeys.

For example, if one product is getting a large amount of inquiries, marketing teams could segment customers interested in those products and launch pre-sales education campaigns.

Fashion brand Psycho Bunny switched from Zendesk to Gorgias to improve access to reporting tools that surfaced customer patterns and support trends. 

“By cross-referencing our Gorgias data with insights around basket size, product performance, and store performance, we can inform broader business decisions. For example, we can see if a certain store location generated more tickets or how many incoming queries are about a certain product,” says Jean-Aymeri de Magistris, VP IT, Data & Analytics, and PMO at Psycho Bunny.

By integrating insights like these with marketing workflows, teams can build more relevant segments that improve retention and engagement.

Ask your CX team 💬 Which customer segments are most likely to churn or repurchase?

3) Launch more targeted chat campaigns

Chat campaigns are proactive messages that trigger based on real-time behavior and context. You can use CX trends to design campaigns that directly address common objections, answer FAQs, or deliver tailored offers.

How to implement 

Start by reviewing your most common pre-purchase questions with your CX team. Then, create chat prompts that address those concerns exactly where they arise. For example, a sizing guide prompt on product pages or a shipping FAQ in the cart. 

Make sure your message feels helpful and not overly salesy. Conversational AI assistants like AI Agent can also tailor responses in real-time, helping customers get what they need without leaving the page.

Pepper, a size-inclusive bra brand, put this into practice by combining their AI Agent (named Penelope) with targeted chat campaigns to guide shoppers through one of their most common friction points: sizing. Thanks to insights from their support team, Pepper created messaging that helped customers find the right fit instantly. The result was an 18% uplift in average order value. 

“With AI Agent, we’re not just putting information in our customers’ hands; we’re putting bras in their hands. With Penelope on board, we’re turning customer support from a cost center to a revenue generator,” says Gabrielle McWhirter, CX Operations Lead at Pepper. 

Ask your CX team 💬 How are customers reacting to recent promotions or launches?

4) Reduce drop-offs and abandoned carts

When shoppers hesitate at checkout, it’s often because they don’t have the information they need.

Tapping into support conversations allows CX teams to identify common objections. They can then share those insights with marketing to refine product messaging, improve product pages, ads, and marketing campaigns.

How to implement 

Use customer service data to identify the top three objections customers have before converting. These might be concerns about sizing, compatibility, delivery time, or product setup. Then, pair that knowledge with a proactive AI sales tool like Shopping Assistant to offer timely answers that move shoppers closer to purchase.

For example, TUSHY, a modern bidet company, found that many prospective customers were hesitant because they weren’t sure how difficult the installation would be. By using a real-time shopping assistant to address these concerns directly on-site, TUSHY was able to guide shoppers past uncertainty.

Ask your CX team 💬 What are the top three reasons customers contact us before they buy?

5) Monitor Help Center and Dashboard stats to craft smarter content

If you want to know what content your customers actually need, your Help Center holds the answers. Real customer questions are found right in Help Center search queries and article analytics.

By tracking which articles are most viewed, most searched, and most frequently updated, marketers can spot common knowledge gaps and fill them with high-value content.

How to implement 

Start by reviewing your Help Center Statistics to see which articles are performing well, which ones are underutilized, and what terms customers are searching for. 

If an article about “returns policy” is getting a spike in views, that’s your cue to simplify the policy or preempt questions with a dedicated email campaign. Marketing teams could also use this insight to build FAQ-rich landing pages, preempt questions in email flows, or even turn top-performing help content into organic blog posts or performance ad copy.

You can also use Gorgias's Dashboard to spot emerging trends across all your channels. This custom reporting feature lets you choose from various charts that reveal high-level patterns—like the most common contact reasons or sudden spikes in ticket volume—giving marketers early insight into shifting customer sentiment and trending topics across social platforms.

Ask your CX team 💬 Which articles in our Help Center are most searched right now?

Find alignment between CX and marketing teams

When support and marketing teams collaborate, you unlock a cycle of continuous improvement. CX teams surface the insights, marketing turns them into strategy, and both sides drive measurable results.

Here’s how to make it work:

• Set up recurring syncs between CX and marketing teams to review insights from customer service reports.
• Involve support in campaign planning to consider what customer objections might come up. 
• Encourage CX to tag tickets based on themes or behavior that marketing can act on.

Unlock revenue by listening to your customers

We need to reframe CX as a proactive function that drives revenue.

Support teams already have the answers marketers are searching for. You just need the tools to tap into them. Gorgias makes that easy, with flexible reporting features, powerful AI, automated tagging, and integrations that bridge the gap between CX and marketing.

Want to connect your support data to better marketing?

Explore Gorgias’s analytics tools or book a demo to speak to a product expert about how to integrate your support strategy with marketing.

{{lead-magnet-1}}

TL;DR:

• Ecommerce brands not using AI are falling behind, as 77.2% already use it daily to boost efficiency and revenue.
• AI saves time and cuts costs, like Trove Brands saving $23K/month and reducing cancellations by 70%.
• Customers want speed and privacy—AI provides fast, judgment-free answers in sensitive categories.
• AI empowers support teams by handling routine tasks so agents can focus on high-value interactions.

Doing nothing when there’s rapid change happening in an industry is risky business.

Right now, according to our latest report, 2025 Ecommerce Trends, 77.2% of ecommerce professionals are already using AI in their day-to-day work. What happens if you’re part of the 22.8% that isn’t?

Inaction is action—one that’s a quiet drain on revenue, resources, and reputation.

Every minute spent on manual work is a minute your competitors are focusing on higher-value customer interactions, improving CX, testing offers, and scaling campaigns.

And the cost of falling behind is compounding fast. Here’s what you’re losing when you pass on AI.

Time lost = money lost

As support volume grows, so does the cost of inefficiency.

Nearly 80% of CX professionals say AI saves them time. In fact, 83.9% of support leaders using AI in Gorgias say it has made their teams more efficient.

Trove Brands experienced this firsthand:

• They reduced missed cancellations by 70%
• And saved $23,000/month in labor costs by automating repetitive support tasks

If AI can handle 70% of your support tickets, your team finally has the time—and headspace—to focus on the 30% that actually builds trust, drives repeat revenue, and improves the customer experience.

Trust when customers need it most

Hot take: AI isn’t impersonal. Not using it is.

In 2024, nearly one-third of CX leaders worried AI would make interactions feel less human. A year later, that number dropped by half. 

Why? Brands started to see that AI wasn’t hurting the customer experience, it was removing friction from it.

For sensitive or personal products—think wellness supplements, intimate gifts, or anything a shopper might feel awkward asking about—AI creates space for honesty without judgment. And that can change the outcome entirely.

“Too often, a great interaction is diminished when a customer feels reduced to just another transaction,” said Ren Fuller-Wasserman, Senior Director of Customer Experience at TUSHY. “With AI, we let the tech handle the selling—unabashedly, if needed—so our future customers can ask anything, even the questions they might be too shy to bring up with a human. In the end, everyone wins.”

It’s a powerful point, especially for brands where discretion matters. AI removes that barrier. 

You're losing trust if your support experience still makes customers hesitate. For many, that means being able to get an answer without needing to explain themselves first.

Revenue hiding behind unanswered questions

Every unanswered pre-sale question or missed upsell is revenue slipping through your fingers.

Product recommendations alone have the potential to increase revenue by up to 300%, boost conversion rates by 150%, and drive 50% higher AOV. But those results don’t come from hoping customers find what they need. They come from proactively guiding them.

That’s where AI comes in.

With Gorgias AI Agent and automation features, for example, Kirby Allison

• Increased conversions by 23%
• Grew sales from support by 46% in just two months

“Our favorite features are definitely Flows and Article Recommendations. They drive so much automation for us. Shoppers get answers to their questions by themselves—what’s the right size hanger, where is my order, what shoe polish would you recommend, etc,” said Addison Debter, Head of Customer Service.

Flows let Kirby Allison surface up to six commonly asked questions directly in the chat widget. When clicked, each one opens a relevant help article—no agent needed.

Auto responses also allowed the team to handle common inquiries like sizing, shipping, and order tracking before a human ever steps in.

If your support team isn’t set up to handle pre-sale conversations at scale, the cost isn’t just in time. It’s in all the revenue you never realize you’re missing.

A CX team stretched thin

It might sound counterintuitive, but AI gives your team more space to be human.

The myth that AI replaces agents is still floating around in some circles, but the reality inside fast-growing ecommerce teams looks different.

In fact, AI frees up time for your team to focus on what they do best: solving complex problems, building relationships, and creating moments that actually drive loyalty.

SuitShop is a perfect example of this in action. When the team adopted AI Agent, they paired automation with intentional escalation: 

“We’re helping customers feel confident during some of the most important moments in their lives—weddings, proms, job interviews, and everything in between. Naturally, my biggest concern with introducing AI was: ‘Will customers feel like they’re getting the same level of care from AI?’ But learning that AI Agent would pull knowledge from our Help Center articles and Macros, which are already written in our brand voice, made me feel more confident,” said Katy Eriks,
Director of Customer Experience.

AI was able to handle common pre-sale questions like shipping timelines and product availability, while human agents stepped in for customizations, wedding-specific questions, and tailored styling support.

The goal wasn’t to remove the human element. It was to give their agents the time and context to show up more meaningfully.

The longer you wait, the harder it is to catch up

In just one year, AI adoption among Gorgias users jumped from 69.2% in 2024 to 77.2% in 2025.

Excitement is rising, too: 55.3% of ecommerce professionals now rate their interest in AI as 8–10 out of 10, up from 45.6% the year prior.

AI is no longer in its experimental phase. It’s the standard, baked into everyday workflows across ecommerce.

If you’re still on the sidelines, 2026 is going to feel like a catch-up game.

The good news? You don’t have to overhaul everything to get started.

So while we’re on the topic of speed, let’s walk through how to start implementing AI for your brand.

How to get started with AI

You don’t need to automate everything on day one. The best CX teams start small, pick the right entry points, and give AI the same level of care you’d give a new team member. Here’s how to roll out AI in a way that actually works:

1. Vet your options thoughtfully

When searching for a new AI tool to help you manage CX, look for one that:

• Offers strong tone-of-voice control so your AI doesn’t sound like a chatbot from 2012
• Delivers consistently accurate responses, even as inputs and workflows evolve
• Provides real post-sale support to help your team troubleshoot, train, and scale usage

Price matters, but it shouldn’t be your only filter.

Also, AI should make your team feel more capable. If it feels like a bolt-on or requires constant developer help, it’s going to create friction, not solve it.

2. Make someone own it

The most successful AI implementations all have one thing in common: someone owns it.

“One of our CX Managers spent 30–40 hours a week building and refining AI. That ownership was critical,” said Sarah Azzaoui, VP of Customer Experience at Clove, when she was explaining how her team first got started with AI.

What many people don’t realize is that AI isn’t going to be perfect out of the gate. AI takes real time and intention to build out. Assigning a clear point person—or better, a small squad—ensures someone is tracking performance, making optimizations, and flagging edge cases.

3. Involve your CX team from the start

No one knows your customer conversations better than your support team. They see the full range of questions, tone, friction points, and emotional nuance every day.

Bringing them into the AI rollout early helps you:

• Identify which questions are repetitive and low-stakes
• Flag which issues should always be handled by a human
• Set realistic expectations across the org about what AI should handle vs. what it could handle

This step also builds trust. If your agents feel like AI is something being done with them instead of to them, adoption is smoother and the outcomes are better.

4. Start small with the right topics

One of the biggest mistakes brands make with AI is trying to do too much, too soon. AI rollout should feel like a phased launch, not a switch flip.

Start in a test environment if your platform allows for it. Roll out automation in stages—by topic, channel, or ticket type—and QA every step of the way.

We suggest beginning with high-volume, low-complexity tickets like:

• “Where’s my order?”
• Subscription pauses or cancellations
• Returns and exchanges
• Store policies and FAQs

Platforms like Gorgias offer tools like Auto QA that track whether AI responses hit the right tone, offer accurate answers, and resolve issues effectively. Use those tools to catch gaps early and monitor performance over time.

That slow, deliberate rollout pays off in performance. At Psycho Bunny, AI Agent now automates 30% of customer tickets, with custom messaging that reflects their brand tone and processes.

Once you’re ready to scale, you’ll feel more confident that the simple queries are handled correctly while you start to train the AI on more nuanced questions.

For example, Gorgias’s Guidance feature gives AI access to non-public SOPs so it knows how to respond or when to escalate.

“The Guidance feature is so important,” said Tosha Moyer, Senior Customer Experience Manager at Psycho Bunny. “We have a lot of processes that we definitely don’t want described in a customer-facing article, but we want AI Agent to be able to access that information and manage tickets accordingly.”

5. Prep your knowledge base

Even the best AI platform can’t succeed without solid inputs.

Before you roll out, take a hard look at your help docs and macros:

• Are they accurate?
• Are they clear and consistent in tone?
• Are they tagged so AI can understand when to use them?

Think of this step as training your AI. The stronger your internal content library, the more helpful and brand-aligned your AI will be across every channel.

6. Communicate with customers

Whether you disclose AI usage is up to you, but be intentional.

Some brands choose anonymity for a more seamless experience. Others find that transparency builds trust, especially when something goes wrong.

What matters most is that your approach aligns with your brand tone and customer expectations—and that clear escalation paths are in place if a conversation needs a human.

Research shows that 85% of consumers want companies to share their AI assurance practices before rolling out AI-powered experiences. Customers are open to AI. But they expect clarity when it counts.

7. Scale the program over time 

Once you’ve built the foundation, scaling AI across your CX org becomes a lot easier.

“We started with cancellations. Now we’re rolling out warranty claims, retention campaigns, and more,” said the team at Trove Brands.

After proving value with one or two ticket types, look for opportunities to expand:

• Pre-purchase product recommendations
• Exit-intent offers via chat
• Predictive personalization
• Multichannel automation across email, SMS, and live chat

The goal is to implement smarter automation that makes your team more effective and your customers more supported.

The future is human + AI

The best CX teams aren’t choosing between AI and human agents. They’re choosing both and building stronger systems because of it.

“It’s not human agents vs. AI,” said the team at Clove. “Our team helped shape the AI strategy—and that changed everything.”

But ignoring AI? That comes at a cost. And it’s not just inefficiency. It’s:

• Missed sales from unanswered questions
• Slower support that erodes customer trust
• Burnt-out teams stuck in reactive mode
• Lower CSAT from inconsistent experiences
• And eventually, falling behind as the rest of the market moves forward

It’s time to build it into your workflows. Not just as a helper, but as a core part of your team.

Start using Gorgias AI Agent to reduce ticket load, recapture revenue, and deliver the kind of support that actually feels personal.

{{lead-magnet-1}}

TL;DR:

• Actions are tasks automatically performed by AI Agent for customers. From address changes and subscription pauses to order cancellations, Actions can fulfill requests for your customers, even when your human agents are offline.
• Actions connect directly to your ecommerce apps. Currently, Actions have native integrations with Shopify, ShipMonk, ShipHero, ShipStation, Stay AI, Recharge, Loop, Subscriptions by Loop, Skio, Seal Subscriptions, and Wonderment.
• Use pre-built Actions or build your own. There are 12 Action templates available, or you can build Actions using custom HTTP requests.
• Watch out for setup snags. Conflicting Guidance, multiple matching Actions, older orders, or broken logic can block an Action from executing.

Automated responses don’t actually resolve anything. In reality, they increase customer wait time.

What a customer really wants is immediate resolution, whether they’re looking to cancel an order, change a shipping address, or pause a subscription.

So, how do you go beyond automated text responses? AI Agent Actions. 

Below, we’ll go over the 7 most common customer service requests you can resolve with AI Agent Actions, so your team gets time back to strengthen customer relationships, increase revenue, and improve your CX strategy. 

{{lead-magnet-1}}

What are AI Agent Actions?

AI Agent Actions are tasks AI Agent can complete for your customers, such as canceling an order or updating a shipping address. 

Instead of handing it off to a human agent, AI Agent resolves the ticket by connecting to your ecommerce apps and performing the action on its own.

You get maximum control over when and how Actions are executed. Before performing the Action, AI Agent asks customers for confirmation, respecting your processes and maintaining a high level of customer service. Once an Action has been taken, you can even share feedback with your AI Agent to reinforce its behavior or finetune it further.

Pro Tip: Unlike Guidance, which tells AI Agent how to respond in a conversation, Actions determine what happens. It’s the difference between saying “I’ll refund your order” and doing it.

Related: How AI Agent works & gathers data

Top 7 customer requests you should be automating with AI Agent Actions

Ready to resolve requests in seconds? Activate these pre-built Actions in Gorgias to keep your team efficient and your customers happy. 

1. Customer wants to update their shipping address

Action to use: Update shipping address

Supported apps: Shopify, ShipMonk, ShipHero, ShipStation

Incorrect shipping addresses lead to costly re-shipments, delays, and even refunds. Catch errors early to keep customers satisfied and excited about their order.

Why do you need this Action? 

The reality is your agents aren’t available 24/7. Unless you hire a team to cover night and weekend shifts (which is unlikely), requests will be missed. AI Agent fills in that gap, handling time-sensitive issues when your team is off the clock. Missing them isn’t just about poor customer experience—it can also lead to extra costs, like reshipping orders.

2. Customer wants to cancel an order

Action to use: Cancel order 

Supported apps: Shopify, ShipMonk, ShipHero, ShipStation

Perhaps a customer ordered the wrong item, chose the wrong size, used the wrong card, or simply changed their mind. Allow them to quickly cancel their order and receive a refund in one go.

“Actions responds to tickets within about 30 seconds and is available 24/7. Regardless of when a customer places their order, the likelihood of quickly catching and canceling the order has increased by 70% since we started using Actions. It’s an exceptional result."

—Jon Clare, VP of Customer Service at Trove Brands

3. Customer wants to replace/remove an item in their order

Actions to use: 

• Replace item, or 
• Remove item

Supported app: Shopify

It happens—shoppers order the wrong size or color and want to change their order immediately. Regardless of the reason, make their new decision easy to implement. Quick, accessible order updates prevent returns, lost revenue, and, most importantly, customer disappointment.

Here’s what the replace order item setup looks like in Gorgias:

Pro Tip: If you have unique workflows, you can create advanced, multi-step Actions and connect to your tools beyond our default integrations. This option requires some tech know-how (like custom HTTP requests), so feel free to bring in your developers for assistance.

4. Customer wants to skip or pause a shipment

Actions to use:

• Skip next subscription shipment, or
• Pause subscription

Supported apps: Stay AI, Recharge, Subscriptions by Loop, Skio, Seal Subscriptions

Subscriptions shouldn’t be all or nothing. Let customers skip a shipment or pause their subscription, so they can come back when they’re ready. Giving them full control lets them manage their subscription on their own terms, reducing churn rate in the process.

Here’s how AI Agent handles a skip shipment request: 

5. Customer lost or damaged their order in transit

Action to use: Reship order for free

Supported apps: Shopify, ShipMonk

No customer expects a lost or damaged order. Let customers know that you have their backs by reshipping a new order free of charge. Fast resolutions during unexpected events demonstrate your commitment to customer satisfaction.

“An instant response builds confidence. We live in a world with short attention spans, so customers appreciate how quickly we can respond to their inquiries. Customers aren’t worrying unnecessarily for longer than they have to for an address change or order cancellation.”

—Mia Chapa, Sr. Director of Customer Experience at Glamnetic

6. Customer wants to know their return shipping status

Action to use: Send return shipping status 

Supported app: Loop

Customers want to know that their return package is on its way to you, so they can redeem their refund. Easily send them a shipment tracking link to give them that peace of mind.

7. Customer wants to know about order status

Action to use: Get order info 

Supported apps: Shopify, ShipHero, ShipMonk, ShipStation, ShipBob, Wonderment

Based on Gorgias data, order status ranks among customers' top 10 questions for support teams. Reassure your customers with quick updates on their orders, including product details, shipping progress, expected delivery date, and other helpful information.

What to know before turning on Actions

Here are a few helpful setup tips to make sure Actions run without a hitch:

• Guidance can override Actions. If conflicting Guidance exists, it may prevent an Action from triggering, even when all conditions are met. Review your Guidance to avoid overlaps, or write your logic into the Action description instead.
• Any Action that changes data requires shopper confirmation. Actions like canceling orders, updating addresses, or canceling subscriptions mean AI Agent will always ask the shopper to confirm before making a change.
• Currently, only one Action can run per ticket. If multiple Actions qualify, none will run, and the ticket will be handed off. Use conditions carefully to ensure only one Action matches per use case.
• AI Agent can only access the shopper’s last 10 orders. If the customer references an older order, the Action won’t trigger and the ticket will be handed over for manual handling.

AI Agent Actions speak louder than words

If you want…

• Fewer repetitive tickets
• Faster customer support
• Happier customers who get what they need instantly
• More time for your team to strategize
• Lower costs and higher efficiency

AI Agent Actions can get you there.

You’ve now seen how Actions can resolve tickets in a snap—no unnecessary handoffs, canned responses, or long response times.

Book a demo to see AI Agent Actions work in real time and start automating what you shouldn’t be doing manually anymore.

{{lead-magnet-2}}