BFCM 2026 Readiness Hub: The AI Chat and Support Checklist for Shopify Stores

The short answer. Get AI chat ready for Black Friday and Cyber Monday in nine jobs: set scope and dates, load the holiday facts your agent needs, prepare gift-finding conversations, tune sales conversations, prepare support answers, write handoff rules, test with real holiday questions, set a routine for the peak days, and plan the week after. Each section below is a checklist you can copy into your launch plan and assign to an owner. In 2026, Black Friday is Friday, November 27 and Cyber Monday is Monday, November 30.

This is a preparation method. It makes no promise about sales or ticket volume, because both depend on your traffic, promotions, carriers and catalog.

How to use this checklist

Copy the sections into a shared document. Give each line an owner and a date. Work from the top: the early sections decide what the later ones test. If you only have a few days, do sections 2, 6 and 7, because wrong policy answers and missing handoffs are hard to repair in the middle of a peak.

Peak season can change what shoppers ask. Questions about deals, gifts, delivery dates, cutoffs, stacking and returns may all become more common. An AI agent answers from what you give it. The job before the rush is to make sure what you give it is current, complete and consistent.

1. Set scope, owners and dates

• ☐ Name one owner for AI chat during the peak, and a backup who can be reached out of hours.
• ☐ Decide what the AI covers: pre-purchase questions, order status, returns and exchanges, cancel and modify requests, subscription changes. Write down what it does not cover.
• ☐ Decide which channels are in scope. Store chat is the baseline. If you also use email, social or WhatsApp AI, test those harder, or keep them out of the peak.
• ☐ Write the peak calendar: sale start, early access, shipping cutoffs, carrier holiday schedules, support hours and any store closures.
• ☐ Set a content freeze date, after which only urgent fixes go live. A week before Black Friday is a reasonable default; choose the date that fits your launch.
• ☐ Choose two or three measures before you start, for example the share of conversations answered without a person, how long handoffs wait for a reply, and orders from sessions that included a chat. Write the definition of each next to it.
• ☐ Decide how you will judge the result. Peak week is not comparable to a normal week. If you cannot hold back part of your traffic, record your baseline and say plainly in your report that the comparison is limited.
• ☐ Ask your vendor what happens at a usage or traffic limit and whether you receive a warning before it is reached. Peak traffic can move you across a tier or a cap.

2. Load the holiday facts

The agent is only as accurate as the policies and data behind it. Replace the quiet-season versions.

• ☐ Shipping. Handling time, carrier transit time by region, and the order-by date for each shipping method for delivery before the holidays you serve. Include international orders if you ship them.
• ☐ Returns. The holiday return window (for example, whether a November purchase can be returned in January), gift returns and exchanges, who pays return shipping, and whether store credit applies.
• ☐ Promotions. Only promotions that are live, with start and end dates, eligible products, exclusions, whether codes stack, and whether sale items are included. Delete expired codes and old campaign pages from the sources your agent reads. If an offer is not documented, the agent must not offer it.
• ☐ Stock. What the agent says for out-of-stock items, backorders, pre-orders and waitlists.
• ☐ Gifts. Gift cards, gift wrap, gift messages and shipping to a different address.
• ☐ Support hours. When a person replies, how, and how long a reply takes in practice. Promise only a reply time you can keep.
• ☐ Top sellers. Check titles, variants, sizes, materials and compatibility notes for your top-selling items. Put key facts in text, not only in images.
• ☐ Conflicts. Find places where two sources disagree (a policy page and an old FAQ, for instance) and fix the source. Know which source wins when they differ.

3. Prepare gift-finding conversations

A gift shopper may have only a recipient, a deadline and a budget. A short conversation can help more than a grid of products.

• ☐ Write the questions the agent asks: who the gift is for, the occasion, interests or use, a budget range that fits your price ladder, and the delivery date needed. Keep it to three questions before it shows products.
• ☐ Prepare gift guides as plain lists by recipient and budget band, built from products you choose from your own catalog. A person decides what goes on the list. The agent should not invent categories.
• ☐ Define how the agent answers "will it arrive by" questions. It should confirm a date only when your shipping data supports it, and otherwise give the order-by cutoff.
• ☐ Write the gift-return answer: gift receipts, exchange options and what the recipient needs to bring.
• ☐ Decide which gift questions go to a person: anything about health, safety, allergies or suitability for a particular child or pet. The agent should not give medical, safety or efficacy advice.

4. Tune sales conversations

• ☐ Decide where proactive messages appear and how many a shopper can see in one session. Repeated pop-ups annoy shoppers and add noise. Keep messages out of checkout.
• ☐ Write each proactive message around a question the shopper is likely asking: fit, delivery date, whether a sale price applies, or which of two items to pick.
• ☐ Ban invented urgency. No made-up stock counts, countdown claims or "selling fast" without data behind them.
• ☐ Set discount rules in advance. The agent mentions only published offers, and any request for an exception goes to a person.
• ☐ Add your shipping cutoff to the messages on product and cart pages, so the agent and the page say the same thing.
• ☐ If your tool supports custom proactive sales rules, write them for the peak and review them after the first day. Rep AI supports custom proactive sales skills.

5. Prepare support answers

• ☐ Order status. Confirm your agent reads live order and tracking data, and write a holiday delay message that you can switch on if carriers slow down. Never let the agent promise a delivery date that the data does not support.
• ☐ Cancel and modify. Decide the point after which an order cannot be changed, and make the agent state it plainly.
• ☐ Returns and exchanges. Load the holiday policy. If you use a returns tool, check that the connection works. Rep AI connects to Loop for returns.
• ☐ Subscriptions. If you sell subscriptions, decide what a customer can edit themselves and what needs a person.
• ☐ Delivered but not received, damaged items and missing items. Write the steps and mark them for handoff.
• ☐ Policy questions you answer differently in peak season. Final-sale items, restocking, price matching, holiday exclusions.

6. Write the handoff rules

• ☐ List the topics that go to a person: refund requests outside policy, delivery problems, damaged goods, legal or safety questions, a customer who is upset, and a customer who asks the same question twice.
• ☐ Decide what the person receives: the conversation, the order number and what the customer was already told. Rep AI escalates to your team with the conversation attached, to helpdesks including Gorgias, Zendesk, Kustomer, Freshdesk and Richpanel.
• ☐ Define out-of-hours behavior. State what the agent tells the customer when no one is online, and what reply time you can honor.
• ☐ Assign who reads the handoff queue on each shift, including weekends.
• ☐ Test the full path once: a handoff leaves the chat and appears in the helpdesk with the context a person needs.

7. Test before the rush

• ☐ Build a test set of about 20 real questions from last year's inbox, chat logs and sale-period tickets. Include awkward ones.
• ☐ Add holiday-specific questions: "Will this arrive before Christmas?", "Can I use two codes?", "Does the sale price apply to my cart?", "Can I return a gift without a receipt?", "Can you ship to a different address?"
• ☐ State something false about your policy and check that the agent does not agree.
• ☐ Ask what the agent does when it does not know. It should say so and pass the question on.
• ☐ Ask someone outside the team to try to break it.
• ☐ Test on mobile and on desktop, logged in and logged out.
• ☐ Record who signed off, the date and the version of the knowledge and policies that were live.

8. Run the peak-days routine

• ☐ Read a sample of conversations at set times each day, for example morning, midday and evening. Look for wrong answers, missed handoffs and repeated questions.
• ☐ Keep a wrong-answer log, and fix the source policy within hours, not days. Decide in advance who may edit.
• ☐ Watch the handoff queue: how many conversations are waiting and for how long.
• ☐ Know how to pause a proactive message, or the agent itself, if something goes wrong, and who decides to do it.
• ☐ If carriers slow down, update the delay message and the policy source together.
• ☐ Limit changes during the peak to fixes. New ideas wait until Tuesday.

9. Plan the week after

• ☐ Export the themes: what shoppers asked, what the agent could not answer, and what handoffs were about.
• ☐ Review a sample of handoffs with the team. Was the case routed correctly? Did the person have what they needed?
• ☐ Record results with the date range, the definitions you set in section 1, and the comparison you used.
• ☐ Update policies and pages for the January returns period.
• ☐ Decide what stays on for the rest of the season and what turns off.

Two examples from Rep AI's case studies

Cabana Life: weekend questions. Both examples come from case studies that Rep AI publishes on its own site. Cabana Life sells sun-protective clothing on Shopify. In Rep AI's case study on Cabana Life, dated September 25, 2026, Jennifer Kovac, Director of Ecommerce, describes the problem: “The majority of our sales and revenue happen over the weekend. So when a customer had a question, there was a good chance we’d miss the sale because we wouldn’t be able to respond until Monday. As a customer, I wouldn’t wait around on a website for an answer - I’d move on. We needed a way to respond instantly.”

The case study describes an AI concierge the store named Jess. The store gave it real-time context such as current promotions, upcoming product launches and its free-shipping threshold. Rep AI's case study reports that 89.68% of conversations were resolved by AI with no human needed. The page gives no date range for that figure.

Nutrabio: a welcome message for sale days. Nutrabio is a supplement brand. In Rep AI's case study on Nutrabio, Jonathan Akcay, Digital Marketing Manager, explains why shoppers need guidance: “Because we are a supplement brand where we have over 400 supplements, it can get very confusing for new customers or people that haven’t been to our website in a while.” The page says the store worked with the Rep AI team to create a customized experience, and it describes an intro message for sale events such as Black Friday and Memorial Day that welcomes shoppers, highlights the sale and guides them through product options.

Neither example is a forecast for your store, and the Cabana Life figure is one store's result as Rep AI reported it in its case study.

Where Rep AI fits

Rep AI makes AI agents for Shopify sales and support. The Sales Agent starts conversations from what a shopper is doing on the page. The Support Agent covers order status, returns and exchanges, cancel and modify requests, and subscription edits. The Concierge plan combines sales and support. Core plans are priced by monthly website sessions, and the add-ons carry separate pricing. A simulator is available at /simulator. The 14-day free trial needs no credit card.

Disclosure: Rep AI publishes this checklist and sells a product that it applies to. The two customer examples come from Rep AI's own published case studies. The checklist is written to work with any AI agent.

FAQ

How do I prepare customer support for Black Friday on a Shopify store? Set the scope and owners, load current shipping, returns and promotion policies, write handoff rules, test with real holiday questions, and plan staffing for the peak days. Sections 1, 2, 5, 6 and 7 above cover this.

What should I load into an AI agent before BFCM? Shipping cutoffs and carrier schedules, the holiday return window, live promotions with exact terms, stock handling, gift options and support hours. Remove expired offers and old campaign pages from its sources.

How do I set human handoff rules for peak season? List the topics a person handles, define what the person receives, set out-of-hours behavior and assign who reads the queue on each shift. Then test the whole path from chat to helpdesk before the rush.

Which holiday questions should my AI agent be tested on? Delivery before a holiday, stacking codes, sale-price eligibility, gift returns, shipping to another address, order changes and delayed or missing packages. Add any question that filled your inbox last year.

When is Black Friday in 2026? Friday, November 27. Cyber Monday is Monday, November 30.