This chapter introduces marketers to the main concepts behind the scenes and explains some of the recurring tools or features that are widely used across the software to avoid repetition throughout this guide.
1 Audience, visitors and profile nurturing
We use the term audience to refer to your entire population of visitors. At the heart of Marketing Factory is the ability to collect data about your audience and reuse that data in multiple situations.
A profile is the name given to the complete set of data known about a given visitor.
This profile contains different types of information:
- Properties (name, address, age, number of kids, etc.).
- Behavioral information (pages visited, goals achieved, etc.).
- Contextual information for each visit (browser, device, geolocation, etc.).
- Classification information (visitor belongs to certain segments, has been assigned to lists, etc.).
Marketing Factory creates a profile automatically each time a new visitor accesses a tracked site, even for anonymous users, and immediately starts collecting data. Behind the scene, when a new visitor arrives, Marketing Factory stores a cookie in his browser cache with a unique identifier. The unique identifier is also stored in the profile, identifying this cookie which in addition will allow Marketing Factory to identify the user during his visit(s).
Each time the visitor comes back to the website and performs certain activities, his profile is updated with new or more current data. For example, if one day the visitor identifies him- or herself somehow (by logging into Digital Experience Manager, logging in with Facebook or providing his or her name or email in a form), then the profile is no longer anonymous and all data collected previously remains attached to that visitor profile.
Obviously, when the same visitor uses two different devices or browsers, Marketing Factory will create several different profiles (one for each browser / device) and, at first, is not able to know that it is the same person. However, Marketing Factory contains a built-in mechanism to merge profiles.
As soon as a user identifies him- or her by trustable means (using a login) on two different devices or browsers, Marketing Factory understands that the multiple profiles already created are, in fact, related to the same user and is then able to gather all those profiles into a single master record. Going forward, Market Factory then continues to track all user activities in that merged profile even if the user comes back without re-logging in because all his or her cookies are linked to that unified profile.
Note that the two last categories are, in fact, segments. Therefore, you can modify the conditions that defines contacts and leads to adapt them to your own needs / business protocols / way of seeing things.
2 Events and goals
Events are, basically, the interactions between a visitor and your digital touchpoints that are under Marketing Factory surveillance. Some events are collected by default each time they occur, like a session start or a page visit. Some events are collected only because they have been explicitly requested by the marketers in charge of the site by using the software interfaces.
Goals correspond to such types of events that are not collected by default but are collected upon request. Typically, a marketer declares a goal to use it as:
- A (key) performance indicator (KPI), or,
- A condition to search, extract, export or filter the profiles repository, or,
- A condition for building segments or personalization, or,
- An analysis criteria point, or,
- All the above points at the same time.
When a visitor reaches a goal, it said to be a conversion. If the goal type allows it, Marketing Factory not only provides absolute numbers for the goals (like the exact number of conversions) but also a conversion rate as a measure of efficiency.
3 Reports, conversions and analytics
One of the key features of Marketing Factory is the infinite drill-down capability, which allows marketers to analyze any measured conversion by any property stored in the visitors profile and refine (or, break-down) their analysis reports over and over.
The infinite drill-down is accessible in:
- Site goals reports,
- Campaign goals reports,
- Optimization test reports, and,
The infinite drill-down capability may be extended to other parts of Marketing Factory later.
For each and every measured result (goals, optimization tests, etc.) Marketing Factory provides two ways to analyze the results, which make the understanding of visitors behavior both simple and very detailed: the quantitative analysis and the qualitative (or performance) analysis.
3.1 Quantitative analysis
The quantitative analysis report lets you understand who contributed data to your goal, it gives a detailed view of how the entire population of converted users is composed of or whatever the criteria of analysis you choose. In a way, this report is close to what web analytics engines provide, except that you can view the results by properties that web analytics engines typically do not track although they are much more valuable for marketers.
Lets take the following goal as an example.
And have a look at the quantitative performances report.
The first column displays the break-down of the selected property for every value that has been collected (8 nationalities, the rest of the audience is unknown).
The second column success tells how many visitors achieved the goal. Here 105 American citizens submitted the form (the goal), 112 Russian citizens, etc.
The third column tells you what percentage of the population this line represent. Here 9.79% of the visitors who achieved the goal were Americans, 9.79% French.
Each column is sortable by ascending or descending order.
You can choose in the Filter on drop down list any other criteria, for instance the gender.
In both cases, we have the same total number of visitors who achieved the goal, obviously, but with a different split criteria.
What becomes really interesting is drilling-down those results. If you select one line in the table by clicking on it, this subset of the population is selected as new scope of analysis and you can then do the same type of analysis by any criteria.
If we click on female, the first line. This criteria will appear in the header of the table and the scope of analysis is not on the 1072 conversions but only on the female population (443 conversions). As an additional criteria we have chosen to split this population by Operating System.
We can continue to add criteria, always by clicking on the subset of population that interests us.
The more data has been collected the more detailed and meaningful the results will be.
By choosing relevant criteria, marketers can understand who (in volume) is converting and have a perfect view of their audience. In the example below, we see that: 43% of the U.S. women citizens who reached the primary goal are older than 51 years of age.
It is possible to remove a filter at any time, just by clicking on the red-cross near its name.
3.2 Performance analysis
Knowing who achieved a goal is certainly a good thing, but being able to understand how your population performed against that goal compared to one or more criteria, thats something completely different and thats the objective of the Performance analysis table.
This report is even more interesting than the quantitative report, because this is where you can expect to find-out some patterns, some different behaviors between your different segments of population, the number of users who matched that goal, and thus improve drastically your performances by creating more targeted personalization.
In terms of usability, the qualitative analysis works exactly as the quantitative one, by selecting a first criteria in the drop down list of available properties, then selecting an entry and drilling-down as long as Marketing Factory is able to provide data to analyze.
Using the same goal we can already see that the conversion rate is different from one nationality to another one.
Comparing the population age for the USA we discover that the population between 31 and 40 does not convert well, while however the population elder than 50 converts globally well.
Based on that discovery, a marketer could decide to modify his site by better targeting the people between 31 and 40 by improving the CTA(s) intended for this segment and leaving the current content for people under 50.
4 Conditions Builder
In many situations, marketers will need to select, or extract, visitors from the whole population or build rules to define which visitors should see what information, by providing personalized experiences
Each time, marketers will use the same tool called the Conditions Builder. The following pages explain the principles of building conditions. Later in this guide well refer to building conditions several times without giving detailed explanations to avoid repeating over and over again.
The Conditions Builder allows marketers to define a rule that contains an unlimited number of conditions. These conditions are linked together using the operators AND or OR. They can group several conditions together and nest conditions in order to define very precise and complex queries if needed.
Each condition is built in the same format: an attribute, followed by a comparator, followed by a value (optional).
|An attribute||A comparator||A value (optional)|
|Date of birth||Is before||1985|
The comparators will vary depending on the types of attributes.
Here is an example for a text attribute; the available comparators are the following (middle column):
|Text Attribute||Comparators||Example of value|
|Last name||Doesnt equal||Stalone|
|Last name||Starts with||swartz|
|Last name||Ends with||egger|
|Last name||Matches regular expression||^sch.*gg.|
|Last name||Exist||Means that the property is not empty|
|Last name||Is missing||Means that the property is empty|
|Last name||Is in||Stalone, Lungren, schwarzenegger|
|Last name||Is not in||Stalone, Lungren, schwarzenegger|
For a date attribute, the comparators are different (middle column):
|Text Attribute||Comparators||Example of value|
|Text Attribute||Comparators||Example of value|
|Birth date||Is before||Jan 01, 1950|
|Birth date||Is after||Jan 01, 1940|
|Birth date||Is same day||Jul 30, 1947|
|Birth date||Is not same day||Mar 25, 1974|
|Birth date||Is between||Jan 01, 1900 and Jan 01, 1950|
|Birth date||Exist||Means that the property is not empty|
|Last name||Is missing||Means that the property is empty|
4.2 Operators between conditions
A rule can contain one or more conditions. Conditions must be linked by one of the following operators: AND or OR.
If two conditions are linked by the AND operator, only the profiles that match both will validate the rule as true.
Example: the following rule will match all the profiles of visitors born between Feb 01, 1940 and Feb 01 1950 and contain the string egger in their last name.
If two conditions are linked by the OR operator, all visitors matching at least one of the two conditions will belong to that segment.
The following rule will match all visitors with Arnold as first name and have Governator as job title.
Marketing Factory allows you to create very complex rules by defining multiple conditions (as previously seen) and also combine them into conditions blocks. Its also possible to combine condition blocks together by using the AND operator. To match the rule, visitors must match each block linked together by an AND operator. But inside the condition block itself, you can use either AND or OR operators.
Lets take a look at an example thats not to complex, but gives you a first impression on the possibilities you have to create powerful and complex segments, depending on your business requirements.
The visitor profiles that need to match the above rule can be described like this:
All visitors that have Arnold as first name AND have an interest score in Sport over 200,
All visitors that have Arnold as first name AND have an interest score in over 100 and zip code in California
All visitors that have Arnold as first name AND have an email address that contains governator.
4.3 Negating conditions (exclusion)
The condition builder does not provide negation operators like AND NOT or NOT to create negative conditions or blocks of conditions. In fact, you dont need these type of operators, because you can define an exclusion at the condition level itself.
So if we want to select all the males that are not 30 of age we wont create the rule like this:
- gender =male
- AND NOT
- age = 30
But like this
- gender = male
- age is not 30
By not implementing negation the conditions are easier to create and understand than if combined together with exclusions.
5 Data collection and scope of analysis
Marketing Factory has been designed to support working with several websites simultaneously and uses the notion of collection and analysis scopes to partition the data into separate logical entities. Scopes are automatically associated with websites through the administration interface of Marketing Factory. A web site is always associated with one scope, not multiple. The most important things to understand about scopes are:
- Each web site is associated with a scope.
- A scope is linked to one website.
- All visitors profiles data are global to your Marketing Factory instance and are shared between scopes. This is intentionally, it means that the same person visiting different websites, that may (or may not) have different scopes, will be recognized as a unique individual, therefore avoiding fragmenting profile data. In the case of companies running tens or hundreds of websites, this is a major advantage.
- Every event collected by Marketing Factory is related to the scope in which it was generated. Practically that mean that the goals that are measured for one particular scope wont be seen and wont affect others.
- Unless shared on purpose segments and lists are created in a specific scope and can be used only within that scope.
6 Shared segments and static lists
As explained in the paragraph Scope, Marketing Factory can be used on multiple sites at the same time while still maintaining a unique collection of profiles. Nevertheless, of different marketing teams being in charge of different sites within the company, brand or even different practices. Notably, if a company maintains several sites dedicated to serving different countries, the way to analyze or segment an audience may be very different for cultural or legal reasons, or by the way the marketing team decides to handle their market and customer base.
Lets consider the following examples. In France, the age of becoming an adult is at the age of 18, while it is 21 in the U.S. This is a significant difference in terms of promoting personalized content, e.g. what are you authorized to show, tell or sell. In India, the notion of young persons is certainly not the same as in Germany. In some countries, asking the gender or the age of a visitor may be very impolite while totally fine in other geographical areas or cultural groups. We could multiply those examples exponentially.
To solve this problem and allow different teams of marketers to use Marketing Factory in parallel on the same platform, segments and lists can be shared - or not - between all sites. By default, segments and lists are not shared among different web sites, meaning that each website (even in the same scope of analysis) will use only the segments or lists that have been created for them.
As a consequence, the team managing the India site will be able to define its own young persons segment with its own conditions, while the team working on the German site will be able to define its own version of the same segment. The first team can also decide to declare several segments that are specific to their market and that have no meaning for the German market.
If you want to share a segment or a list among all the sites within your platform and make both accessible to all the sites marketers only will have to check the shared checkbox present in the creation and edit forms of segments and lists.
A shared segment or list is shared completely, it means that any marketing team that can use the segment can also modify it and redefine its scope, with consequences for all the other teams and their sites. Therefore, use shared segments and lists carefully.