Christmas business 2026: Why your Q4 will be decided in August

The Christmas season isn't decided in December, but in the weeks when nobody's thinking about it. Those who lay the technical groundwork in August, put the pressure on in September, fine-tune in October, and simply sell in November will breeze through the peak season. Those who reverse the order will spend the fourth Sunday of Advent searching for mistakes they could have fixed in twenty minutes in the summer.

I know how this all sounds right now. It's 26 degrees outside, the first dominoes are already on display next to the sunscreen in the supermarket, and now an e-commerce agency comes along and starts talking about Christmas. I understand. I'd much rather be writing about something else, too.

But your shop isn't interested in our feelings. It's all about cold, hard calculations. And the calculation goes like this: anything that's supposed to work under load in November has to have been built beforehand under calm conditions. There's no other way. There never has been. And every year on the third weekend of November, someone somewhere is sitting in front of a shop that buckles under the strain of 400 simultaneous visitors, wondering why it worked so well in the test system with three people.

The inconvenient number with which it all begins

Before we talk about the technical aspects, let's briefly discuss the scale of things. Because most shop owners don't underestimate the importance of Christmas. They underestimate just how important.

The Federal Statistical Office has calculated this precisely: German retailers generate around 19 percent of their total annual revenue in the months of November and December . Almost a fifth. In just two out of twelve months. And in some sectors, the figure is even more dramatic—26 percent for toys, 24 percent each for books and consumer electronics, and 23 percent for watches and jewelry. Those who want to read the details can find the Federal Statistical Office's analysis of the share of Christmas sales in annual revenue in the "Number of the Week" series.

Translated into your reality: A six-hour outage on the second Advent weekend costs you roughly the same as a full week in March. A checkout that stutters on a mobile device costs you many times more in November than it does in June. And a four-second loading time is annoying in the summer—in December, it's a revenue-killing machine with a built-in timer.

That's the whole reason why this text is appearing now and not on December 1st. On December 1st, all I can offer you is calming tea.

The honest timeline: What needs to happen when

The mistake I see most often isn't laziness. It's the wrong order. Almost every online shop owner does the right things—they just all do them at once in November. Then campaign planning clashes with the server migration, the new payment method with the theme update, and in the middle of it all, the accountant asks for the inventory.

Let's sort this out. Here's the schedule we've been using with our clients for years, in exactly this sequence.

August: The foundation — boring, but indispensable

In August, nothing is optimized or beautified. In August, the focus is on testing whether the entire structure is structurally sound. Specifically, this means bringing all systems up to date and ensuring their security: the shop core, extensions, PHP version, and database. It sounds like maintenance work, and that's exactly what it is—except this time, the maintenance worker is preventing the roof from collapsing in December.

Also essential now: a functioning staging system. Not just something we've set up, but a true copy that you can safely mess with. If you have to test a change directly on the live system in November, you've missed something in August.

And then there's the topic nobody likes to talk about: backups. Not just whether they work, but whether a shop can actually be restored from them. A backup that's never restored isn't a backup at all. It's just a file made with good intentions.

September: The stress test

Now it gets interesting, because now we stop guessing. In September, your shop should be put under load—artificially, in a controlled manner, with three to five times the number of visitors of a normal day. Not to see if it holds up. But to see where it gives way first.

The answer, by the way, is almost never the web server. Most of the time it's the database, often a single, poorly designed filter in the category view, sometimes a plugin that dutifully calls an external interface and waits politely on every page request. You won't find these things by thinking about it. You'll find them by measuring.

In September, data work runs parallel: completing product data, cleaning up categories, and checking feeds. Anyone who discovers in November that half the items lack high-resolution images has a problem that can no longer be solved with money—only with all-nighters. We've explained in more detail how closely inventory management and the online shop are intertwined in our article on ERP integration with the online shop.

October: Focusing

October belongs to conversion and campaigns . Now—and only now—do A/B tests make sense; now is when ads are set up, target audiences are refined, and email campaigns are built. The reason for this timing is simple: A test needs traffic to be meaningful, and it needs stability to run smoothly. In October, you have both. In November, you have plenty of one and none of the other.

October also sees the most important decision of the entire season, and most people make it unconsciously: When will changes stop? More on that in a moment.

November: Sell, don't build

E-news for ambitious developers - read up to speed Load testing in practice: How to find the limit before the customer does.

November has exactly one purpose: to sell. Not for renovations, testing, or "quick fixes." Black Friday falls on November 27th in 2026 , Cyber ​​Monday on November 30th , and the first Sunday of Advent, on November 29th, is right in the middle. That's not just a week. It's one long, drawn-out state of emergency.

What happens during this time is observation and reaction: monitoring, inventory levels, customer service, incoming payments. And an emergency plan that doesn't exist in someone's head, but is written down. Including phone numbers. Including the question of who is authorized to make decisions when the boss is on Christmas vacation.

December: The final sprint and what comes next

In December, communication is more important than technology. What's the order deadline to ensure timely delivery? This single piece of information—prominent, honest, and up-to-date—will determine more sales in the second half of December than any discount campaign. And starting December 20th, the next chapter begins: the wave of returns in January, which you should prepare for now, or not.

But is the effort even worth it anymore?

A valid question, especially when considering current consumer sentiment. The short answer: yes — but the real action is online.

The German Retail Federation (HDE) anticipates a nominal sales increase of 1,6 percent in brick-and-mortar retail for 2026 , while online retail is expected to grow by 4,3 percent. That's almost three times as much. The current figures are available in the HDE's core data on retail sales trends.

The picture becomes even clearer when looking at official statistics. Internet and mail-order retail sales increased by 10,1 percent in real terms in 2025 compared to the previous year—while overall retail sales grew by 2,7 percent in real terms. In December 2025 alone, online sales were 5,2 percent higher in real terms than in the same month of the previous year. This can be found in the press release from the Federal Statistical Office on retail sales in 2025.

The Christmas shopping season isn't disappearing. It's just shifting. And it's shifting to where you already are—provided your shop delivers on the promises made in your ads.

The five problem areas where it actually fails

Enough theory. These are the points where things go wrong in practice — sorted by the frequency with which we hear them in emergency calls in November.

1. Charging time under load — not charging time at idle

Everyone measures their loading time. Almost no one measures it at the right time. A shop that loads in 1,2 seconds at 10 a.m. might take eight seconds at 8 p.m. on Black Friday — because caches are full, because the database is creating queues, because the advertising service provider's tracking script suddenly becomes the bottleneck.

Measure under realistic load, and measure the values ​​that matter. We've broken down the relevant metrics in detail in our article on Core Web Vitals: Targeted Improvement of INP, LCP, and CLS .

2. The checkout that only works on your mobile phone

E-news for ambitious developers - read up to speed The tech check before the high season: 14 points you should check off now

The classic costly mistake. The checkout process is tested—by the operator, on their current device, in their browser, with their payment method, and on Wi-Fi. What isn't tested: a four-year-old Android device on an overloaded mobile network, ordering as a guest, cancellation mid-payment, and the back button.

The last point is particularly notorious. A customer abandons the payment, clicks back, and the shopping cart is empty. This order will never be repeated. We've compared the payment methods you should offer in our payment provider comparison.

3. Inventory levels that differ between the shop and the warehouse

As long as you have twenty orders a day, sluggish inventory synchronization goes unnoticed. At three hundred, it becomes noticeable—in the form of cancellations, apologetic emails, and a customer service department that spends December trying to mitigate the damage.

If your online store and inventory management system only communicate hourly, you'll reliably sell items that are no longer available during peak season. Achieving seamless integration of your ERP, CRM, and inventory management systems isn't a project for November 20th—but it is one for now.

4. No code freeze

My favorite project, because it costs nothing but discipline. Set a deadline after which no further changes are made to the system. Our recommendation: early November, the 10th at the latest. From then on, only genuine bug fixes, and those must be done with two people present.

I know how tempting it is to quickly install that new banner plugin on November 25th. I also know how many shops have failed precisely because of this. A feature you implement on November 25th will, at best, gain you a few percent. A mistake you make on November 25th will cost you the entire day.

5. An emergency plan that only exists in your head

Ask yourself honestly: If your shop is unreachable on the Saturday evening of the second Sunday of Advent — who will notice? After how many minutes? Who will that person call? And what if they don't answer?

If you don't have an answer to any of these questions, that's your most important task this week. Not the loading time. Not the campaign. The slip of paper with the phone numbers.

The Q4 checklist month by month

What you can specifically do this week

If you've read this far and now feel an urgent need to do something: Start small. These four things together won't take you half a workday and will yield more than most large projects.

First: Restore a backup in a test environment . Not theoretically—actually. Second: Place an order in your own shop. As a guest, from your mobile phone, using a mobile network, with the least popular payment method you offer. Third: Check when your shop and inventory management system last synchronized stock levels. Fourth: Write an emergency plan. Half a page is sufficient.

How worthwhile this preparatory work is becomes clear when looking at market developments: The German E-Commerce and Distance Selling Association (bevh) reported a merchandise turnover of €83,1 billion in German online retail for 2025 and, together with the EHI Retail Institute, anticipates a further increase of 3,8 percent for 2026. The complete figures can be found in the bevh press release on e-commerce growth . So the pie is growing. The only question is whose shop will be accessible on the cut-off date to get a slice.

Conclusion: December is a result, not a task

For many, the Christmas season feels like a sprint that begins at the end of November. In reality, it's a result—the sum of all the decisions made between August and October. In December, you're simply reaping the rewards. What you didn't sow in the summer won't grow back.

The good news: You're early. It's August. You still have three peaceful months in which you can make mistakes without it hurting. That's a luxury you won't have after November 1st. Make the most of it.

And if you're thinking, "Things will work out somehow for us," that's exactly what the people who called me on the second weekend of Advent said. They're all very nice, by the way. Just not particularly relaxed.

What about you?

Now I'm curious—and I mean genuinely curious, not rhetorically. What's your biggest challenge before this Christmas season? Loading times? Inventory levels? Or simply the fact that you have to manage it all on your own?

And the question that interests me most: What went wrong for you last December? Write it in the comments. I don't mean this as an attempt to embarrass yourself; on the contrary—the best checklists are created from real mishaps. These kinds of stories aren't in any manual, and that's precisely why they're so valuable.

If you want to know where your shop stands technically before things get serious: Our free SEO and visibility check analyzes every website without registration. And if you prefer speaking to a person rather than a tool, you can find us via the contact page —you'll reach us much faster in August than in December. Just so you know.

This post is part of our Q4 series. A new post will be published here every Tuesday until Christmas to help you prepare for the peak season—from load testing and price reductions to shipping deadlines. To stay up-to-date, subscribe to our newsletter.