Do not launch a new online store in the last weeks before December. Either work back from the holidays to a launch in early October, leaving time to fix problems, or keep the old site until January.

A launch date that slipped

An online shop's redesign was due in September. This week the designer sent a new plan: photos came late, the gift options need work, and launch day is now Monday, November 24.

Redesigns slip for ordinary reasons, usually on both sides: decisions wait, text arrives late, a feature proves harder than it looked.

The trouble is the date. It falls in Black Friday week, and for a shop that takes much of its year's orders in December, no week is worse for starting over.

What every launch brings

Even a careful launch has a settling period, when small faults surface that testing on a quiet site rarely catches:

Each is usually fixed within hours. The question is whether you want to be fixing them while the year's most important orders arrive.

Why late November is the worst week

In a quiet month, a faulty shipping rate costs a few orders before someone notices. In early December the same fault meets the year's heaviest buying, and nobody has an hour to chase it.

The people who would fix it are stretched too: staff packing parcels and answering the phone, and a designer who may be launching other shops for the same deadline.

Search settles slowest. Google has to visit the new addresses and connect them with the old ones, which can take several weeks: here, the whole shopping season.

Working back from December

Start with the date the shop must be calm and count backwards. For a gift shop that is usually the last week of November, when holiday buying begins in earnest.

Before it, allow six to eight weeks for the site to settle, so search engines can catch up and the checkout can meet the odd orders only real customers place.

That puts launch day in early October. If the new site cannot be finished and tested by then, the safer choice is the other one.

Keeping the old site until January

The other choice is to run the old site through the season and launch in January, when orders are quiet and most holiday returns have been dealt with.

Use the extra months well. Put the new site on a private test address, place real orders with real cards, refund them, and check every email, invoice and stock count.

Meanwhile, give the old site what it needs for one more season: this year's shipping cut-off dates, the holiday products, current prices and any software updates due.

A worked example

A small preserves kitchen on Vancouver Island makes jam, marmalade and chutney in small batches, sells at summer markets, and ships gift boxes across Canada from its website.

December brings most of its online orders. Its redesign had slipped twice, and the latest launch date sat four days before Black Friday.

Working back from December 1 with the designer, the owner found an October launch possible only by skipping the testing. The new gift box builder had not yet taken a real order.

So the old site stays for one more season, with this year's shipping dates, two new gift boxes and a notice of the last order day for holiday delivery.

The new site has moved to a test address, where staff will place real orders through the fall, from one jar to a corporate gift order. The designer fixes what they find, and it launches in January.

Redirects soften the settling period

The biggest single protection for any launch is a full set of redirects, so that each old product and category address leads straight to its nearest new page.

Without them, bookmarks, links in past newsletters and the shop's pages in search results all end at an error, and Google treats the new pages as strangers.

Ask for the redirect list as a document you can check. Open twenty old addresses from last year's best sellers and see where each one lands.

What a slipped date usually means

A slip in August often hides another. When photos or text arrive late, the work that depends on them, such as product pages and redirects, is squeezed too.

Ask the designer for a plain list of what is finished, what is waiting and on whom. It makes the October or January decision far easier than a promised date.

Be honest about your own side. Many delays start in the shop: a decision put off, descriptions not written, a supplier's photos not sent.

Freezing the live site for the season

Whichever date you choose, set a freeze: a day in mid-November after which nobody changes the live shop except to fix a fault.

That covers new extensions, theme changes and bulk price imports. Each is small, and each can break the checkout in the weeks you can least afford it.

Security updates are the exception. Apply them when they are released, then place a test order straight away.

The counter-case

Not every shop earns its year in December. A garden supplier or a seller of camping gear may be quiet in late November, and could sensibly launch then.

And sometimes the old site cannot last another season, because its checkout is failing or its software is no longer supported. Then launch, with the designer on call, other changes frozen and a test order every day.

The principle holds either way: a new site needs a calm stretch after launch, and a shop's busiest month cannot provide one.

Choosing the launch date

  1. Mark the date your busy season begins.
  2. Count back six to eight weeks for settling.
  3. List what is unfinished, and who owes each piece.
  4. Choose early October or January, nothing between.
  5. Prepare the old site for the holidays if it stays.
  6. Test with real orders on a private address.
  7. Freeze the live site from mid-November.

Step four is the decision that matters. Early October and January are both safe; any date between hands the new site's first problems to the shop's most important weeks.

A related pressure, wanting a new site before the busy season because a rival just launched one, is weighed in a competitor's new website and the urge to match it.

Frequently asked questions

Why is late November a bad time to launch an online store?

Every new site has a few weeks of broken links, checkout faults and moving search positions. In late November those weeks land on the year's busiest shopping.

When should a shop that is busy in December launch instead?

In early October, which leaves six to eight weeks for the new site to settle, or in January, once holiday orders and most returns are done.

Is it safe to keep the old site through the holidays?

Usually, yes. Give it this season's shipping dates, holiday products and current prices, and keep its software updated while the new site is tested.

How should a new store be tested before launch?

On a private test address, with real orders paid by real cards and then refunded, checking every email, invoice, tax amount and stock count.

What is a freeze on the live site?

A date, usually in mid-November, after which nobody changes the live store except to fix a fault or apply a security update.

What if the old site cannot last another season?

Then launch anyway, but with the designer on call, redirects for every old address, a freeze on other changes and a test order every day.

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.

Redesign running into the holidays?

Send us the launch plan and the list of unfinished work, and we will help you choose a date the shop can live with.

Start a Conversation