The Black Friday checklist that starts as a new customer

To check a store before Black Friday, test it as a brand new customer: from an email address and phone number that have never bought from you, sign up through the pop up, check the welcome email and its code arrive and the code works at checkout, opt in to texts, leave a basket and a checkout, and wait for each reminder. Run the 12 tests before your code freeze, again after any theme, app or flow change, and on the morning the sale goes live.

On Tuesday 22 September a test shopper put a £40 gift set in a UK skincare store’s basket and left. A second started checkout with a £21 deodorant and left before paying. 48 hours later, neither had heard a thing (the September store check). Nothing in a dashboard shows that. A dashboard shows what you set up and what you sent; only a customer sees what arrived.

Black Friday is Friday 27 November and Cyber Monday is Monday 30 November. Most checklists for the week start inside your dashboards: flows switched on, segments built, templates approved. This one starts where your customer does, with a new email address, a phone and a basket.

Every click costs more that week, so every break in the path costs more too. And the breaks that cost most are quiet. A welcome email that never sends doesn’t raise an error. It sends nothing, and the flow report shows a zero that nobody opens until the month is over.

It isn’t rare. When Joy Joya signed up to 362 ecommerce brands with fresh email addresses and phone numbers in July 2026, 158 never sent a welcome email, 53 never sent the text they promised, and 20 sent an offer that didn’t match the pop up. And these are the emails that earn: automated flows were 5.3% of email sends in Klaviyo’s 2026 benchmarks and nearly 41% of email revenue.

The way to find these breaks is old and cheap: become a customer your store has never met, and walk it the way they will. Below are the 12 tests in the order a shopper meets them, what a pass looks like, and where to look when one fails. Each takes minutes. The waiting is the long part, so start them together.

Before you start: be a customer your store has never seen

Your own login is the worst test account you have. Your store knows you, your email platform already has a profile for you, and your discounts and flows treat you as staff or as a past buyer. Set up a stranger instead.

• A new email address that has never subscribed, ordered or been used by your team. Use a Gmail address, so you also see which tab each email lands in.
• A phone number that can receive texts and has never opted in to yours.
• A private browser window on your phone, not logged in to your store or its admin. Test on the phone first, then on a laptop.
• A sheet with 4 columns: the step, the time you did it, what arrived and when, and a screenshot.
• A note to whoever reports your numbers, with the address, so your tests stay out of them.

Every test, in the order a shopper meets it

1. Sign up through the pop up, on your phone

Wait for the pop up the way a visitor would, enter the new address and submit. A pass: the form submits, the thank you message shows the offer it promised, and if you use double opt in, the confirmation email arrives and its link works. If it fails, check the form still feeds the right list, how it looks on a small screen, and what changed on the site since it last worked.

2. The welcome email arrives, with the code it promised

Write down the minutes from submit to inbox, and compare the code and its terms with what the pop up showed. Note the tab too. Promotions is a tab in the inbox, not spam, and 1 new address shows 1 inbox’s view, so record what you saw rather than a verdict. If nothing arrives, check the flow starts from the list the form feeds, whether double opt in is holding the new profile back, and whether the address already existed as a profile, which can keep it out of a welcome flow.

3. The welcome code works at checkout, on Black Friday prices

Put a sale item in the basket and add the welcome code, on your phone. A pass: the code applies, and it either stacks with the sale or is refused, exactly as you planned, in every market and currency you sell to. Check it hasn’t reached a usage limit or an end date set months ago. If your discounts ran on Shopify Scripts, those stopped running on 30 June 2026 and now need Shopify Functions or an app, so test every custom discount again.

4. The text opt in sends a confirmation, and STOP works

Opt in with the new number. A pass: a confirmation text arrives and names your store, and replying STOP removes the number with no texts after it. Note when each text lands in the shopper’s time zone. The rules on consent and quiet hours differ by country and by US state, so check yours with your text platform or a lawyer. This checklist isn’t legal advice.

5. Browse and leave, if you run a browse email

View 2 products and leave without adding either. A pass: the browse email arrives after the delay you set and names what you looked at. Then check the flows talk to each other: a shopper who has left a basket should get the basket email, not a browse email that asks if they were just looking.

6. Leave a basket

Add 1 product and leave. A pass: the reminder arrives after your flow’s delay, with the product’s name, photo and price filled in, and its link brings the basket back. Watch for blanks: an empty product table reads as a broken store. If nothing arrives, check the event that starts the flow is still being recorded, the flow’s filters and Smart Sending. Theme changes are a common cause.

7. Start checkout, and stop before paying

Enter the new address at checkout, tick the marketing box if you show one, and leave before payment. A pass: 1 checkout reminder arrives. Check which tool sends it, Shopify’s own abandoned checkout email if it’s on or your email platform’s flow, so a shopper gets 1 reminder, not 2. Then decide on purpose what happens in Black Friday week: a shopper who opened your morning campaign can have the afternoon reminder skipped by Smart Sending, unless you switch it off on that email.

8. The Black Friday price is the same from the ad to checkout

Open your Black Friday ad and follow it to the landing page, the product page, the basket and checkout. A pass: the same price, offer and dates at every step, on your phone, with the sale price shown against the old one wherever you meant it to be. A banner that promises 30% and a checkout that takes 20% costs the sale and the shopper’s trust.

9. Every live ad lands on a page that works

Open the link of every live ad on your phone, not a sample. A pass: the page loads, the product is in stock and the offer matches the ad. Ads can keep running after a product sells out until your feed or catalogue catches up. If you run Google’s AI Max, Final URL expansion is on by default and can send clicks to other pages on your site. The landing pages report shows which ones Google chose, and URL exclusions narrow them. An ad landing check does this every morning.

10. Place a test order

Shopify gives 3 ways to place one: its test payment gateway, Shopify Payments test mode, or a real order you cancel and refund straight away, which also tests your live payment provider and may cost a fee. A pass: the order confirmation arrives, the thank you page loads with the right order, and the shipping email arrives when you mark it fulfilled. Turn test mode off afterwards. While it’s on, customers can’t place real orders.

11. Ask your chat what customers will ask this month

Ask: what’s the last day to order for delivery before Christmas, how long do I have to return a sale item, and does my welcome code work on sale prices? A pass: every answer matches this season’s policy pages. A bot set up last year can still be quoting last year’s dates. If your chat is an AI, it should say so: the EU AI Act’s transparency duties for chatbots have applied since 2 August 2026. A support bot check asks these questions every day.

12. Create an account, then reset the password

Make an account with the new address, log out and ask for a password reset. A pass: both emails arrive within minutes and both links work. These are the emails that break quietly after a migration or a checkout change, and they tend to surface weeks later, as complaints rather than on launch day.

When to run them

• Now, before your freeze. Pick a freeze date for code, apps and flows, and run all 12 before it.
• After every change. A new theme, app, discount or flow edit is when emails stop, so run the tests again the same day.
• The evening the sale goes live, and Black Friday morning. Codes, banners and prices change at the last minute.
• Cyber Monday morning, if your offer changes for the day.

If you freeze everything this week and change nothing until December, you don’t need a weekly routine: 1 full run now and 1 on Black Friday morning is enough.

If you send a lot of mail that week, check your sending too. Google’s sender guidelines say Gmail has been ramping up enforcement on senders who don’t meet them since November 2025.

What a test by hand can’t tell you

Running these by hand is the right start, and for 1 store it may be all you need. It has 4 limits worth knowing before you report a result to anyone.

• 1 inbox, 1 moment. A pass on Tuesday says nothing about Friday.
• Silence needs a control. If no welcome came, check that a second new address got your newsletter. If it did, your sending works and the flow is the fault. If neither got anything, look at the form or your sending first.
• 1 inbox shows 1 view. A new subscriber’s Gmail tab can differ from a loyal customer’s.
• Screenshots are weak proof. The email itself, with the time it arrived and its sender check, settles an argument. A screenshot starts one.

AI assistants can take some of the clicking. ChatGPT’s agent can fill in a form and pauses for you at a login, and Claude can read your inbox through its Gmail connector. Both run as you, in your browser and your inbox, and their recurring work has caps: ChatGPT allows 3 to 15 active scheduled tasks, run at most hourly. None of the help pages we read covers keeping each store’s emails as evidence. For 1 store, a new inbox and an assistant to summarise what arrived is a sound method. For 20 client stores every week, it isn’t.

For agencies: run it on every client, and on the stores you pitch

Run it on a client’s store with the client’s OK, and on a prospect only through the public sign up any customer can use. The result that lands best is the email itself, or the 48 hours in which nothing came, with the time on it. It answers the question agency owners say costs them clients.

A weekly run of these 12 tests on every client store is a report a client can check without taking your word for it. Before a pitch, the same tests on a prospect’s public sign up give you a first line they can check too.

Questions