Page tree
Skip to end of metadata
Go to start of metadata

Personalisation lets you adapt contents according to certain criteria, such as country, age, recurring or new visitor, gender, etc. There are two kinds of personalisation: Page personalisation and component personalisation.


According to rules being matched Magnolia serves the original content or a variation of it. A variation is a physical copy of the content being personalised.

Variants in page personalisation

  • The audience is served different versions of the same page.

  • The variants are copies of the entire page.

  • There can be only one page personalisation at a time.

Variants in component personalisation

  • The audience is served the same page with different versions of the component.

  • The variants are copies of a single component.

  • There can be more than one component personalisation at a time.


To simplify the process of assigning rules, you can divide the entire visitor population into segments. A segment is all the visitors who meet a given rule or set of rules.


  • Segments are a composed of a group a rules

  • Segments are used for easy targeting audience for personalised content.


A trait is an attribute of the visitor or visit that you can detect and assign a value to.

  • Traits represent the rules used for personalisation.
  • Traits allow you to target visitors with personalised content.
  • Traits are used for configuring the audience of a given variant.

  • Multiple traits can be used for targeting an audience with AND and OR rules. These are called "All" and "Any".

  • Default traits are: date, country, visitor and cookie.

  • The fundamental parts of a trait are: rule field, value field and a voter. Also you need a POJO to represent the trait, a parameter converter class and a filter to detect the trait.

<trait name>


Trait name. Choose a name that is easy for editors to understand such as countrydate or userAgent.



Field used to define permitted values for the trait. See Rule field below.



Field used to enter a single value to test personalized content delivery. See Value field below.



Converter class which must implement info.magnolia.personalization.preview.parameter.PreviewParameterConverter<Object>.



Trait class. Doesn't have to explicitly implement any interface. Usually it is just a plain Java object which allows you to specify the trait and its characteristics.



Voter class which must extend AbstractBoolVoter<TraitCollector>. See Voters.


optional, default is false

Adds the trait to the Preview app. Set this property to true for traits that editors preview content with routinely. It makes the Preview app faster to use. If a trait is used rarely, leave the property out.

Personas and Preview app

To test your personalised content you can impersonate the visitor or simulate trait values to verify that the correct page variant is served.

  • Personas are profiles of persons with a given set of traits with values.

  • A persona has a name, description and a picture to identify it.

  • The Preview app lets you choose the page and define the traits to test.

  • The Preview app lets you use a persona to test a given visitor.

Demo of Personalised Campaign

Following the tutorial:

We will achieve the following set of personalised pages:

Target country(none)ChinaUnited Kingdom
Run dates(none, ongoing)May 26 to June 9, 2016July 27 to August 10, 2016
Component title FestivalsDragon Boat FestivalPuck Fair
Component text

Experience diverse cultures at their most festive. Join in the fun of festivals around the world with our exclusive festival tour packages.

The Dragon Boat Festival is also known as the Duanwu Festival. It is a traditional holiday that commemorates the life and death of the famous Chinese scholar Qu Yuan. The festival occurs on the fifth day of the fifth month on the Chinese lunar calendar.

On August 10, a three-day festival known as Puck Fair is held in Killorglin, Ireland. Residents of the town capture a goat and proclaim it to be the King for three days of revelry.

Component image

  • No labels