Multivariate landing page optimization: Difference between revisions

Revision as of 15:35, 20 June 2008

Latest revision as of 18:54, 15 December 2023

(21 intermediate revisions by 14 users not shown)

Line 1:

{{Short description|Test variations of web page elements to find the best one}}

'''Multivariate landing page optimization''' (MVLPO) is a specific form of [[landing page optimization]] where multiple variations of visual elements (e.g., graphics, text) on a pagewebpage are evaluated. For example, a given page may have ''k'' choices for the title, ''m'' choices for the featured image or graphic, and ''n'' choices for the company logo. This example yields ''k×m×n'' landing page configurations.

The first application of an experimental design for MVLPO was performed by Moskowitz Jacobs Inc. in [[1998]] as a simulation/demonstration project for [[LEGO]]. MVLPO did not become a mainstream approach until [[2003]] or [[2004]].

Multivariate landing page optimization can be executed in a live (production) environment, or through simulations and [[market research]] surveys and simulations. Examples of live environment applications include Omniture.com, [[Google website optimizer]], Memetrics.com, Widemile.com, and Optimost.com. An example of survey/simulation applications is StyleMap.Net.

== Overview ==

Multivariate landing page optimization is based on [[experimental design]] (e.g., [[discrete choice]], [[conjoint analysis]], [[Taguchi methods]], [[IDDEA]], etc.), which tests a structured combination of webpage elements. Some vendors use full factorial approach (e.g., Memetrics.com) xOsuse thata "full factorial" approach, which tests all possible combinations of elements). This approach requires lessa smaller sample size (typicallysize—typically, many thousands)thousands—than totraditional achievefractional statisticalTaguchi importancedesigns thanto traditionalachieve fractional[[statistical Taguchisignificance]]. designsThis andquality is one reason that Choice[[choice Modelingmodeling]] won the [[Nobel Prize]] in year 2000. Fractional designs typically used in simulation environments require the testing of small subsets of possible combinations, and have a higher [[margin forof error]]. Some critics of the approach raisequestion the question of possible interactions between the elements of the web pageswebpages, and the inability of most fractional designs to address thethis issue.

To resolve thesethe limitations of fractional designs, an advanced simulation method based on the [[Rule Developing Experimentation]] paradigm ([[RDE]]) paradigm was introduced.<ref name="isbn0-13-613668-0">{{cite book

|author=Howard R. Moskowitz

|coauthorsauthor2=Alex Gofman

|title=[[Selling Blue Elephants]]: How to make great products that people want BEFORE they even know they want them

|publisher=Wharton School Publishing

|date=2007-04-11

|pages=272

|isbn=0-13-613668-0}}</ref> has been introduced. [[RDE]] creates individual models for each respondent, discovers any and all [[synergies]] and suppressions betweenamong the elements,<ref>Alex Gofman. 2006. Emergent Scenarios, Synergies, And Suppressions Uncovered within Conjoint Analysis. Journal of Sensory Studies, 21(4): 373-414. {{doi|10.1111/j.1745-459X.2006.00072.x}}</ref> uncovers attitudinal segmentation, and allows for databasing across tests and over time).<ref>{{cite web

|url=http://www.ftpress.com/articles/article.aspx?p=1015178

|title=Improving the ‘Stickiness’ of Your Website

Line 25 ⟶ 26:

== Live environment execution ==

In live environment MVLPO execution, a special tool makes dynamic changes to thea web site,page so thethat visitors are directed to different executions of landing pages created according to an [experimental design]. The system keeps track of the visitors and their behavior (includingbehavior—including their [[conversion rate]], time spent on the page, etc.) and withOnce sufficient data has accumulated, the system estimates the impact of individual components on the target measurement (e.g., conversion rate).

Live environment execution has the following advantages:

=== Advantages ===

* ThisCapable approachof is very reliable because it teststesting the effect of variations as a real -life experience, generally transparent to the visitors.

* Generally transparent to visitors

* It has evolved to a relatively simple and inexpensive to execute approach (e.g., Google Optimizer).

* Relatively simple and inexpensive to execute

Live environment execution has the following disadvantages:

=== Disadvantages ===

(applicable mostly to the tools prior to Google Optimizer):

* High cost

* ComplexityIncreased complexity involved in modifying a production-level website

* Long timeperiod itof maytime takerequired to achieve statistically reliable data. causedThis bysituation is due to variations in the amount of traffic, whichthat generates the data necessary for thea decision.

* ThisLikely approach may not be appropriateinappropriate for low -traffic /, high -importance websites when the site administrators do not want to lose any potential customers.

Many of these drawbacks are reduced or eliminated with the introduction of the [[Google]] Website Optimizer – a free DIY MVLPO tool that made the process more democratic and available to the website administrators directly.

== Simulation (survey) execution ==

AIn simulation (survey) based MVLPO isexecution, builtthe onfoundation consists of advanced [[market research]] techniques. In the research phase, the respondents are directed to a survey, whichthat presents them with a set of experimentally designed combinations of thea landing page executions. The respondents rate each executionversion (screen)based on a ratingsome questionfactor (e.g., purchase intent). At the end of the studyresearch phase, [[regression model(s)analysis]] models are created (either for individual pages or for the totalentire panel) of pages. The outcome relates the presence/ or absence of thepage elements inon the different landing page executions to the respondents’ ratings. andThese results can be used to synthesize new landing pages as combinations of the top-scoredscoring elements optimized for subgroups, or [[market segments]], with or without interactions.<ref>Alex Gofman, Howard Moskowitz, and Tonis Mets. 2009. Integrating Science into Web Design: Consumer Driven Website Optimization. The Journal of Consumer Marketing, 26(4): 286-298. {{doi|10.1108/07363760910965882}}.</ref>

Simulation execution has the following advantages:

=== Advantages ===

* Much fasterFaster and easier to prepare and execute (in many cases), as compared to the live environment optimizationexecution

* ItApplicable works forto low -traffic websites

* UsuallyCapable producesof producing more robust and rich data because of a higherincreased control ofover the page design.

Simulation execution has the following disadvantages:

=== Disadvantages ===

* Possible bias because of a simulated environment asrather opposed tothan a live one environment.

* A necessityNecessity to recruit and optionally incentiviseincentivize the respondents.

== References ==

Line 56 ⟶ 55:

[[Category:Search engine optimization]]

[[Category:Internet marketing]]

[[Category:Market research]]