What to prepare beforehand, what to do in the first 30 minutes, what to tell customers, what to do with your ads and what to check once the site is back.

Black Friday falls on Friday, 27 November 2026 this year. For many stores it's the day with the most traffic of the year and, for the same reason, the day when a slow or crashed site costs the most. You can't guarantee it won't happen, but you can decide in advance what you'll do in the first 30 minutes, whom you'll call and what you'll tell customers. When the problem hits, there's no time to think about the order of things: you just follow it. Below I show what I prepare beforehand, what the first half hour looks like, how to communicate, what to do with your ads and what to check afterwards. At the end there's a generator that builds your minute by minute plan and ready-to-copy messages.
Many peak-day outages have ordinary causes: the hosting plan has a resource limit you never notice on normal days, a plugin installed the day before slows down the cart, or an email sent to the whole list brings thousands of people in the same minute.
You don't want to find out from a customer's comment that your site is down. An uptime monitoring service checks the site every few minutes and alerts you by text or phone call. Have it check the cart as well, not just the homepage. Then prepare an emergency list, kept outside the site and the server:
What you need Hosting or platform support, with their opening hours Why First call when the server doesn't respond What you need Your developer or agency, with a phone number Why Rolling back to the last working version What you need Your payment provider's contact Why Declined payments or payments without an order What you need Access to your domain and DNS Why When the problem is the domain, not the server What you need Ads and email apps on your phone, with access checked Why Pausing campaigns when you're away from your deskDecide in advance who makes the calls. In a small team, one person handles the technical side, another handles communication and ads, and a third, if you have one, handles support. If you're on your own, ask someone you trust to answer messages that day.
Sometimes the site works but payment doesn't. A bank transfer option or a second payment provider, set up and tested in advance, is the usual backup; in some markets cash on delivery plays that role. The backup method must already be configured and ready to switch on from a setting, not installed in the middle of the crisis. Also prepare the text for customers who paid and didn't get a confirmation: don't pay again, you'll contact them.
Order matters more than speed:
The maintenance page should return HTTP status 503, not 200 and not a redirect to the homepage. In its Search documentation, Google recommends a 503 for temporary unavailability, with a Retry-After header for the estimated return time and a simple, static error page. The same page says a full site shutdown should last only a very short time, at most two or three days, and that robots.txt must not return a 503, because that blocks all crawling. If only payment is broken, it's better to keep the site visible and disable the cart: Google says this doesn't affect your presence in Search.
If you decide to extend the offer, extend it for real in your discount settings and say exactly until when. I wrote about organising support on these days in the article on customer support in peak season.
Ads keep spending even when the site doesn't respond. Pause campaigns at campaign level in Google Ads and Meta, from the phone apps if you're not at your computer. Just as important are the emails and texts scheduled for Black Friday: if a send goes out during the outage, you push thousands of people to an error page and waste the best message of your year.
In Google Ads, if the site returns errors when Google checks it, ads can be disapproved for a destination not working. Under Google's policy, once you've fixed the site you need to file an appeal to have the ads reviewed again. In Merchant Center, products with an unavailable page are disabled until Google can access the page again; you can request a website review, which can take up to 12 hours. That's why, after recovery, the first thing I check in the accounts is the status of ads and products.
The next day, write one page on what happened, how long it lasted, what it cost and what you'll change. Often the answer is a bigger hosting plan just for November and December, or a plugin removed from the site.
Before Black Friday, my clients and I put together a one-page emergency list: contacts, account access, who decides and the prepared texts. I check in advance which pages the campaigns point to and whether they can be paused from a phone. On peak day, I handle the ads and the messages, while the developer or host handles the technical side.
Choose your platform, the problem, your team and channels; you get the steps by minute and messages ready to copy.
A site that goes down on Black Friday isn't just a technical problem, it's also a communication problem and money spent on ads. Know your limits in advance, keep the emergency list at hand, pause ads and scheduled sends within the first 15 minutes, tell customers what's going on and check everything once you're back. If you want a fast store that's ready for peak traffic, see the frontend development page or write to me via contact.