Online shoppers increasingly look for more engaging experiences when deciding to purchase or not. AI chatbots help take that experience to the next level by holding a conversation with Shopify shoppers by recommending the right product or reminding them of items left in their cart or getting the support they need post-purchase. 

Shopify store owners should implement AI chatbots in the correct way to help boost their revenue and retain repeat customers. Let’s take a look at several mistakes Shopify stores should avoid when building the best AI chatbot for their shoppers.

shopify chatbot

Not Planning Before Building

AI chatbots are extremely versatile carrying the ability to handle many different roles and perform a wide range of tasks that help augment the sales, marketing, and customer support teams. With a wide range of options comes the risk of trying too many things at once causing errors and disorganization.

When Shopify stores are first building their AI chatbots their first steps should focus on planning or mapping out exactly what they want their AI chatbots to accomplish. Remember, AI chatbots perform their tasks conversationally so the planning phase should concentrate on how the Shopify store-to-shopper conversations normally flow. Shopify stores can look at the conversations they are already having with their current shoppers as inspiration for their AI chatbot architecture such as conversations in-person in their brick-and-mortar store, over a phone call, or even reviewing email exchanges or live chat transcripts.

There are several different techniques Shopify stores use to map out conversations they want to be included inside their AI chatbot. One utilizes a bubble diagram where each bubble represents a conversation with other bubbles surrounding them representing different conversations within a certain topic of conversation. You can draw out bigger bubbles that cover major topics with smaller bubbles that are specific conversations inside of that topic.

Think of a large bubble for “Shipping” with smaller bubbles for “when will my order arrive?”, “what are your shipping rates?”, and “what shipping courier do you use?”. No need to connect any conversations in a flow or pattern because everything is interconnected with an AI chatbot so shoppers can switch topics mid-conversation and the AI chatbot will follow along. Just think of and write out different potential conversations that could arise surrounding different subjects. 

Another planning technique is using a spreadsheet program with different columns for potential shopper questions, Shopify store responses, answers, listeners and keywords, and even URL links or images to present the shopper. This technique is a very structured way to script out specific questions and answers ensuring the AI chatbot will help shoppers correctly and efficiently.

Shopify stores can take the spreadsheet a step further by mapping out conversations that have many layers by outlining each response along with the button options that will accompany them. This is a great way to build the AI chatbot that will help increase accuracy and 

Shopify stores can also combine these techniques where the bubble diagram initially helps map out conversation topics leaving the writing of individual responses and replies for the spreadsheet. Having a spreadsheet where AI chatbot builders can easily copy and paste responses and URLs to create intents helps tremendously with organization and speed.

Not Using eCommerce Features Correctly

One of the main reasons stores implement a Shopify AI chatbot is to increase their revenue. Engaging shoppers in conversations to help them answer questions and find the right product for them is a big reason why AI chatbots sell more products and increase AOV. But another main driver of revenue is eCommerce features such as email-less Abandoned Cart, Order Status, and Page Targeting. Shopify stores should utilize these key eCommerce features with correct timing and messaging.

Abandoned cart reminders should have the feeling of more of a friendly reminder that nudges shoppers to check out the items left in their cart from a previous session. They should not annoy shoppers or bother them while they go about their shopping. This is why the correct timing is key so the Shopify chatbot is not pestering while still trying to complete the sale.

Order Status addresses the #1 Shopify question each store faces: “Where is my order?!?”. This feature should be as available as it can be inside the chatbot. Shopify bots should present to the shopper the option to see their order status in as many places as it can such as on the greeting question and inside FAQ intents that cover shipping topics.

Page Targeting is a great way to engage shoppers depending on where they are in their shopping journey. The greeting question should be more general and cover several topics initially but great Shopify bots target certain website pages to ask specific questions about whatever topic that page covers. For example, if a shopper is on a category page the Shopify bot can ask the shopper a question that helps them narrow down what product they are likely looking for. Alternatively, if a shopper is on a product page the Shopify chatbot can ask a question that covers commonly asked questions about that product to help overcome sales obstacles instantly.

Shopify bots are also able to display certain Up-Sells and Cross-Sells inside the chatbot window. These Up-Sells and Cross-Sells should, again, not be bothersome and annoying when displayed and should be appropriate products that shoppers might be interested in. Similar products in a category are good for this eCommerce feature but it is recommended to keep the Up-Sell and Cross-Sell products at an equal or lesser value to help increase conversions.

Shoppers Not Knowing They are Chatting with an AI Chatbot

Shopify stores want to gain the trust of their shoppers to build a relationship that keeps them coming back. Engaging them in a meaningful conversation is key, so why start out the conversation by tricking them into thinking they are chatting with a live person? This can embarrass them and lead them to not believe other things about what you are saying about your company, its policies, or its values.

Some AI chatbots try to impersonate humans by having an icon with a real person’s headshot or by starting the conversation by introducing themselves with just a name – “Hi, I’m Sophia! How can I help you?”. Naming your AI chatbot is a great way to add a touch of personality but you should always let shoppers know they are chatting with an AI chatbot by following up the name with a line about being an AI chatbot- “Hi, I’m Sophia the AI chatbot…” or “Hi, I’m Sophia the AI Shopping Assistant…”.

The last thing a Shopify store wants is their shoppers chatting with their Shopify chatbot, completing a sale or getting their questions answered, then finding out at the end of the conversation they were actually talking with an AI Shopify chatbot instead of the human they thought they were chatting with. This breaks down the trust you were trying to build throughout the conversation leading shoppers to not return for repeat business or even going as far as telling their friends or family about their negative experience with the Shopify store.

Not Performing AI Training

Shopify chatbots are not a “set-it-and-forget-it” eCommerce tool. They should be constantly learning and expanding their role as a sales assistant and extensions of the customer support team. It is imperative they learn from their mistakes and contain up-to-date information and policies. This is where an NLP Specialist performs “AI Training” by reviewing conversations the chatbot is having with shoppers to see where the chatbot is answering questions correctly. Also fixing or “training” the AI when it does not know the answer or answers a question incorrectly.

Shopify stores need to review the chat transcripts in different conversation topics as well as review the fallbacks. Fallbacks are when the Shopify bot doesn’t know the answer to a shopper question and replies with a generic response – “Sorry, I didn’t understand that. Please try asking in a different way. If you would like to talk to a human you can…”. These fallback incidents are key to knowing where the Shopify bot is lacking and where exactly to fill in those gaps of knowledge.

This AI Training needs to be constant in order for the Shopify chatbot to continue to improve. In the beginning, there will be plenty of things to fix and add such as keywords or phrases not covered during the initial build. But as the Shopify bot gets smarter and smarter the workload of performing the AI Training will start to lessen. This does not mean the AI Training will cease entirely, merely that there will be less to fix as the Shopify bot gets smarter.

Not Using Reporting Data to Gain Valuable Business Insights

Shopper behavior online is something Shopify stores are constantly monitoring and using to try and improve their online shopping experience. One of the main benefits of Shopify chatbots is the fact that they are an extremely useful source of knowledge about what shoppers are talking about and searching for. Shopify bots allow shoppers to let down their guard and ask questions they normally wouldn’t because they think no one is listening to their conversation. They feel anonymous so they are a bit freer about what they ask about and how they ask it.

Shopify stores use these conversations for instant feedback such as when a payment system is down or when a certain coupon code is not working. This allows stores to react quickly to the issue knowing exactly what to fix and where to fix it; saving the store potential lost revenue and lost customers.

There is also great potential in discovering key marketing or sales insights based on what shoppers are asking about. The marketing team can benefit from seeing the results and topics of conversation surrounding the latest campaign or the sales team can learn more about a particular topic to improve the sales representative rebuttal scripts.

The possibilities are limitless regarding the knowledge and insight Shopify stores can gain through their AI chatbot. These conversations can be mined for a wide range of uses from SEO keyword insight all the way to product improvements so keep a keen eye when scrolling through conversations.


Shopify chatbots are essential tools for any Shopify store no matter the industry. They provide valuable insights and help boost revenue when shoppers engage with the Shopify bot. There are plenty of pitfalls to avoid when building your Shopify bot but don’t be paralyzed by the fear of making mistakes – mistakes will happen so be ready to fix those mistakes and make your Shopify chatbot smarter and smarter over time realizing more revenue and more satisfied customers.

Chatbot Features

Features Ochatbot ManyChat
Independent from FB (FB not required) Yes No
Integrate to all eCommerce platforms Yes No
Search product database Yes No
Add to cart from chat window Yes Yes
View product details from chat window Yes Yes
Order status and tracking Yes Yes
Recommend products (cross-sell) Yes No
Abandoned cart Yes Yes
Page targeting Yes No
Zapier integration Yes Yes
SMS integration No Yes
AI training Yes No
AI ability to switch conversation topics in real-time Yes No
AI conversational eCommerce (typed, without buttons) Yes No
NLP (Natural Language Processing) Yes No
Live chat built-in Yes Yes
Integrate with 3rd party live chat platforms Yes No
Unlimited SKU's Yes No
Unlimited conversations Yes No
Unlimited Interactions Yes No
Saved Conversations Yes No
Prebuilt standard intents Yes No

How to Build an AI Chatbot Free Course Signup

Greg Ahern
Follow Me