Personalization Pyramid: A Framework for Designing with User Data

As a UX skilled in today’s data-driven landscape, it’s extremely likely that you’ve been asked to design a personal digital experience, whether it’s a common website, user portal, or local application. Despite there still be a lot of advertising hype surrounding personalization systems, there are still very some standardized methods for implementing personalized UX.

That’s where we come in. We set ourselves the challenge of developing a systematic personalization framework especially for UX practitioners after finishing dozens of personalization tasks over the past few years. The Personalization Pyramid is a designer-centric model for standing up human-centered personalisation programs, spanning information, classification, content delivery, and general goals. By using this strategy, you will be able to understand the core elements of a modern, UX-driven personalization system ( or at the very least know enough to get started ).

Getting Started

For the sake of this article, we’ll suppose you’re already familiar with the basics of online personalization. A nice guide can be found these: Website Personalization Planning. Although Graphic jobs in this field can take a variety of forms, they frequently start from the same place.

Popular circumstances for launching a personalization task:

Your business or client made a purchase to support personalization of a content management system ( CMS ), marketing automation platform ( MAP ), or other related technology.
The CMO, CDO, or CIO has identified personalisation as a target
User data is disjointed or confusing
You are conducting some sporadic targeting or A/B tests.
On personalization method, participants disagree.
Mandate of customer privacy rules ( e. g. GDPR ) requires revisiting existing user targeting practices

A powerful personalization plan will need the same fundamental building blocks regardless of where you begin. We’ve captured these as the “levels” on the tower. Whether you are a UX artist, scholar, or planner, understanding the core components may help make your contribution effective.

From top to bottom, the rates include:

Objectives: What are the specific, tangible benefits of the system?
Touchpoints: Where will the personalized experience been served?
Contexts and Campaigns: What personalization information does the person view?
User Divisions: What constitutes a special, suitable market?
Actionable Data: What trustworthy and credible information is collected by our specialized software to improve personalization?
Natural Data: What wider set of data is potentially available ( now in our environment ) allowing you to optimize?

We’ll go through each of these amounts sequentially. An associated deck of cards serves as an example of each level’s specific examples to make this more meaningful. We’ve found them useful in brainstorming about customisation, so we’ll provide cases for you here.

Starting at the top

The elements of the pyramids are as follows:

North Star

What overall goal do you have with your personalization program ( big or small ) is a northern star. The personalisation program’s entire goal is described in The North Star. What do you wish to achieve? North Stars cast a ghost. The bigger the sun, the bigger the dark. Example of North Starts may contain:

Feature: Self-contained personalisation componentry. Examples:” Cooked” notifications, advanced optimizations ( geolocation ), basic dynamic messaging, customized modules, automations, recommenders
Experience: Personal user experiences across numerous interactions and consumer flows. Examples: Email campaigns, landing pages, advanced messaging ( i. e. C2C chat ) or conversational interfaces, larger user flows and content-intensive optimizations ( localization ).
Solution: Highly differentiating customized product experiences. Example: Standalone, branded experience with personalization at their base, like the “algotorial” songs by Spotify quite as Discover Weekly.

Northern light cards. These can assist in guiding your team toward a common objective that personalization may support. They also can help describe the overall goal of the team’s current personalization effort.

Goals

As in any great UX design, personalization may help promote designing with client intentions. The goals are the military and tangible indicators that will support the success of the entire program. Start with your existing analytics and calculation system, as well as indicators you can benchmark against. In some cases, new targets may be ideal. The most important thing to remember is that personalisation is more of a means of achieving an objective than a desired result. Popular targets include:

Conversion
Time on work
Net promoter score ( NPS)
Consumer pleasure

Focus cards. Cases of some frequently used and quantifiable customisation KPIs.

Touchpoints

Touchpoints are where customisation takes place. As a UX artist, this will be one of your largest areas of responsibility. The touchpoints you have will depend on how your personalization and the related technology are implemented, and they should be based on enhancing a person’s encounter at a specific point in the journey. Touchpoints can be multi-device ( mobile, in-store, website ) but also more granular ( web banner, web pop-up etc. ). Here are some examples:

Channel-level Touchpoints

Email: Role
Email: Occasion of available
In-store display ( JSON endpoint )
Native game
Search

Wireframe-level Touchpoints

Web overlay
Web call bar
Web symbol
Web content wall
Web restaurant

Touchpoint accounts. Examples of common personalization touchpoints: these can vary from narrow ( e. g., email ) to broad ( e. g., in-store ).

If you’re designing for web interface, for instance, you will likely need to include personal “zones” in your wireframes. Based on our next action, settings, and campaigns, the articles for these can be presented dynamically in touchpoints.

Targeted Zones: Illustrations from Kibo of personalized “zones” on page-level wireframes that occur at various stages of the user journey ( the engagement period is at right and the purchase cycle is at proper ).

Supply:” Essential Guide to End-to-End Personaliztion” by Kibo.

Contexts and Campaigns

After you’ve outlined some touchpoints, you may consider the actual personal information a user may acquire. Many personalization tools will refer to these as” campaigns” ( so, for example, a campaign on a web banner for new visitors to the website ). These will be displayed automatically to specific customer sections, as defined by consumer data. At this stage, we find it helpful to contemplate two distinct concepts: a framework design and a willing design. The environment helps you consider whether a consumer is engaging with the personalization process at the moment, such as when they are simply browsing the web or engaging in a deep dive. Think of it in terms of behavior in data recovery. The content model can then guide you in deciding what kind of personalization to use in the context ( for instance, an” Enrich” campaign that features related articles might be a good substitute for extant content ).

Personalization Context Model:

Browse
Skim
Nudge
Feast

Personalization Content Model:

Alert
Create Easier
Cross-Sell
Enrich

We’ve written a lot more in depth about each of these concepts somewhere, so be sure to check out Colin’s Personalization Content Model and Jeff’s Personalization Context Model.

Campaign and Context accounts: This level of the tower can assist your team in focusing on the different personalization options that end users can choose and the use-cases where they will encounter it.

User Divisions

Based on customer research, user segments can be created prescriptively or dynamically ( for example, using principles and logic tied to user behavior, or through A/B testing ). You will need to think about how to treat the unidentified or first-time visitor, the guest or returning visitor for whom you may have a stateful cookie ( or an equivalent post-cookie identifier ), or the logged-in visitor who is authenticated. Using the personalisation pyramids, here are some examples:

Unknown
Guest
Authenticated
Default
Referred
Role
Cohort
Unique ID

Segment accounts. Examples of common personalization parts: at a minimum, you will need to acquire the private, host, and logged in user forms. Classification can become incredibly complicated from there.

Actionable Data

Every business has access to data, regardless of its modern presence. It’s important to inquire about how to use the data you can ethically collect on users, its inherent reliability and value, and how to use it ( sometimes referred to as “data activation” ). Fortunately, the tide is turning to first-party information: a recent study by Twilio estimates some 80 % of firms are using at least some type of first-party information to personalize the customer experience.

First-party information represents multiple benefits on the UX before, including being relatively simple to acquire, more likely to be accurate, and less susceptible to the” creep issue” of third-party information. Therefore, determining which method of data collection is best for your audiences should be a crucial component of your UX strategy. Here are some examples:

When it comes to recognizing and making decisions about various audiences and their signals, there is a trend of profiling. As time and confidence and data volume increase, it varies to more granular constructs about smaller and smaller cohorts of users.

Although some form of implicit or explicit data is typically required for any implementation ( more commonly known as first party and third-party data ), ML efforts are typically not cost-effective right away. This is because optimization requires a strong data backbone and content repository. These approaches, however, should be taken into account as part of the overall plan and may in fact help to speed up the organization’s progress overall. At this point, you will typically work with important stakeholders and product owners to create a profiling model. The profiling model includes a defining framework for setting up profiles, profile keys, profile cards, and pattern cards. A multi-faceted approach to profiling which makes it scalable.

Pulling it Together

The cards serve as a starting point for an inventory of sorts ( we offer blanks for you to customize your own ), a set of potential levers and motivations for the personalization activities you aspire to deliver, but they are more valuable when grouped together.

One can begin to chart the entire course of a card’s “hand” from leadership focus to a tactical and tactical execution. It serves as the foundation for the workshops that both co-authors have conducted to build a program backlog, which would make a good article topic.

In the meantime, it is important to note that each colored class of cards is helpful in understanding the range of options that you might have, as well as making informed choices about who, where, when, and how, will be made these choices.

Lay Down Your Cards

Any sustainable personalization strategy must consider near, mid and long-term goals. There is simply no “easy button” where a personalization program can be stood up and immediately see meaningful results, even with the leading CMS platforms like Sitecore and Adobe or the most exciting composable CMS DXP out there. That said, there is a common grammar to all personalization activities, just like every sentence has nouns and verbs. These cards attempt to map that territory.