Emergency situation

In case of emergencies or breakdowns, you can send an SMS to our emergency hotline

On-call phone (SMS only)

+45 29 70 15 95

Send an SMS with the following information:

  • Your name and webshop
  • Description of the problem
  • Your callback phone number

Notes: This service is only for critical situations where your webshop is down or has serious problems. For regular support, please use our normal support channels.

Test your front page against itself - split testing is live in Shoporama

Split testing is now part of Shoporama. You can create multiple versions of your front page, category page or stylesheet, split traffic between them and see which one sells best. The same system can also be used to schedule time-limited pages, such as a Black Friday homepage that turns itself on and off.

Reading time: approx. {eight} min
Test din forside mod sig selv — split-test er live i Shoporama

When you tweak your front page, change a button color or create a new hero area, it can be hard to know if the change actually improves sales. Split testing provides a clear answer: you run two or more versions simultaneously, split the traffic between them and let the numbers decide which one works best.

The feature is now located under Design in your Shoporama admin. It covers both page layouts (front page, category pages, product pages, etc.) and stylesheets, so you can test everything from a completely new front page design to a single button color.

Example: two front pages against each other

This is what it looks like in the Page Designer when two variants of the same front page are running in parallel:

To forside-varianter i Shoporama-admin med 'I split-test'-badges

Both front pages are online at the same time and are part of a split test. Each visitor is assigned one of the versions and the same user sees the same version on repeat visits to avoid contaminating the test.

After a few days of running, the statistics look like this:

Statistik fra split-test af to forsider, viser konvertering, omsætning, RPV og vinder-sandsynlighed

Frontpage converts 15.38% against Frontpage v2's 10.53%. More interesting is the revenue per visitor (RPV): 1,946.15 DKK on the winning variant against 85.99 DKK on the other. The difference is due to both higher conversion and larger orders. RPV combines the two into one number and is therefore the primary measure in the statistics.

Note that the statistic says "not yet enough data to pick a winner". This is intentional. To avoid drawing conclusions based on chance, a minimum of 5,000 impressions per variant, 100 total orders and 14 days duration is required before a variant can be declared a winner.

Also test CSS, e.g. the color of your "Add to cart" button

In addition to page layouts, you can test stylesheet variants. This allows you to compare CSS changes such as button colors, typography or spacing. You create alternative variants and let them compete:

Tre CSS-varianter (Sort, Green, Blå) i Shoporama-admin

Three variants (Black, Green, Blue) are active at the same time and are included in a split test. The statistics might look like this:

Statistik for stylesheet-test med tre farve-varianter

Black has a 60% probability of being the best, but Blue has the highest RPV ($838.89) on the fewest impressions. This illustrates why it's worth waiting for data: a variant with a lower conversion rate can still be the most profitable if it gets more orders. More data will tell the tale.

Time management without split testing

The Variant system can also be used to schedule pages in advance without comparing them. A typical example is Black Friday: instead of logging in on November 28 and switching the front page manually, you create a Black Friday front page and set it to be visible from November 28 to December 1. It will then turn on and off automatically.

It works for all seasons: summer sales, Christmas, Father's Day, Valentine's Day. It can also be combined with split testing, for example by creating two variants of the Black Friday homepage and finding out which one works best on Black Friday traffic.

Automatic completion

Two settings allow a test to run without manual follow-up:

  • Stop automatically when there is a clear winner. When the criteria are met and a variant has a 95%+ probability of being the best, the test stops automatically. You will be notified in admin.
  • Use the winner as the new default. When the test stops, the winner is put online with high priority and the other variants are put offline. They are not deleted, so you can always roll back manually.

With both enabled, you can create a test, click Start and let it run. When it ends, the winning variant is already activated.

How to get started

The feature is located under Design → Split test. If you're not sure where to start, the front page and the primary button color are typically the places where a small improvement will pay off the most because they are seen by all visitors.

Further reading:

The feature is included in your Shoporama solution at no extra cost.

Share this post