Web personalization is on everyone's agenda. In reality, it rarely gets delivered as intended.
Why? The lack of practical guidance makes it feel too daunting to navigate. ROI on personalization is also difficult to quantify, especially when it takes months to deliver. As a consequence, marketing and content teams divert focus to easier, short-term projects instead.
It doesn't have to be this way. Whether you're a seasoned marketer or a novice content professional, you don't have to start from scratch.
Buck the trend and get your web personalization program started or optimized quickly. Forget theoretical advice. This step-by-step guide aims to break down actionable steps and arm you with ready-to-use templates developed from real examples in the wild.
6 Steps to web personalization
Let’s see each step in further detail below.
Define user segments
Understand your target audience needs with user data
The process of segmentation is basically grouping your users into distinct segments that have common attributes that define their behaviors or needs.
The first trap teams fall into is spending too much time creating traditional user personas.
The limits of using ‘personas’
Consider the following similar personas. Notice anything?
While both personas share the same demographic traits, you would simply not market to them the same way.
Enter ‘jobs-to-be-done’, the single most important criterion to start your segmentation exercise with. It’s based on the concept that every user has a job to do on your site. For example, a customer doesn’t want to buy a drill, they want a tool to put a hole in their wall to hang a picture.
Common ‘jobs-to-be-done’ categories include:
- Finding out more about the organization
- Finding out more about a product or service
- Executing on a task - a form fill, application submission, making a payment or a claim etc.
- Contacting support or a specific person.
‘Jobs-to-be-done’ aims at grouping users by focusing on shared user needs and intent. You can conduct interviews with customers or prospects with the following questions:
- Tell me the main task/job you’re looking to achieve on the site?
- Why is that important to you?
- How do you normally find the information or service you need on the site?
- Are there specific information gaps on the site that would improve your ability to get your main task/job done? If so, why is it important to you?
- Did you find the overall experience hard or easy? How could it be improved?
Those questions should offer you greater clarity about why your users visit your site, and what content helps them to get their job done.
‘Start small’ tip: Run just 10 interviews to get going. You might see some trends quickly, or areas you want to drill down into with your next 10 interviews.
Finding commonalities to segment by
Armed with an understanding of your users' goals, you want to look for common criteria that allow you to use something tangible to segment those users with. This translates 'jobs-to-be-done' goals into specific segments of users you can identify using data.
Criteria can include:
- Demographic data — data like age, gender, education, income, job, location, etc. This data can be collected from various sources such as form submission, account registration, interviews or surveys.
- Contextual data — data that place a user within a specific “context”. For example, the device used to access the website, the referring source, the time of the visit, and the user’s location all help paint a picture of that user’s environment.
- Behavioral data — data that refers to how your users interact with your website or digital experiences. It includes specific pages visited, links clicked, search queries, purchases, or transaction history.
- Revenue or User Lifetime Value (LTV) - how much the user spends with you over the course of their lifetime with your organization. I.e. $XX spent a year multiplied by typical number of years they remain a user/customer.
Armed with this data, you should start seeing trends emerge and segments where content personalization makes sense.
Start small tip: Not every user is worth the same to your organization. It’s rarely realistic to create a personalized experience for every type of user segment on your site. So, focusing on 2-3 that make the biggest impact is a sensible way to start; be it revenue, usage of services, risk of complaint/churn etc.
Putting it together
Define personalization goals
Now that you’ve defined what each user segment wants to achieve on your website, you can start mapping these to organizational outcomes linked to users’ jobs.
Here are a few common ones you can take inspiration from:
- Increase engagement
- Increase conversion
- Increase Life-Time Value (LTV) - revenue gained for the time the customer stay with you
- Increase upsell value
- Increase retention rate
- Increase repeat purchase rate
Putting the two together can help you to match user jobs to measurable metrics linked to your organizational goals.
‘Start small’ tip: Here are two examples of a simple mapping exercise.
Personalization item | Details
|Higher education||Discover university credentials||Increase engagement||Bounce rate|
|Visit a specific course page||Increase engagement||Sessions with search|
|Enrol in a course||Increase conversion||Enrolment page visit|
|Submit an inquiry||Increase conversion||Inquiry submission|
|Government||Visit the FAQ section||Increase engagement||FAQ page visits|
|Submit a form||Increase conversion||Form submission|
|Make a payment||Increase conversion||Payment completion|
In Google Analytics 4, you can track these metrics as a Conversion under the Admin panel.
To set up conversion goals in Google Analytics, you can follow these steps:
- Sign into your Google Analytics account.
- Navigate to “Admin” icon (bottom left).
- Select “Goals”.
- Click the red “+ New Goal” button.
- Define your goal.
- Under the “Conversions” tab, select “Overview.”
- Then, click the button “Set up goals.”
- Select the red “New Goal” button.
- Use an existing goal template or “Custom” goal option.
- Click “Continue”.
- Fill in “Goal description” (see image below).
12. Add the page you want to measure in the “Goal details” section (see image below).
13. Click “Save” and you’ve successfully tracked a goal!
For a full walkthrough, visit Google’s Help site here.
Map user journeys
Now that you’ve defined the goal that’s aligned to each user segment’s job-to-be-done, it’s time to work backwards. You want to understand how your user got there in the first place - from their first visit to the end job achieved on your website.
Think of it as a breadcrumb trail.
The goal of personalization is to maximize the important steps along the way to make sure users get to achieve their job as easily and quickly as possible.
Analyze the current behavior of your users
First, let’s take a look at how your current users behave on your website by creating a path exploration report in Google Analytics 4.
In GA4, navigate to Explore > Path exploration.
Click on Start again to start with a blank slate.
Select an Ending point (a “Thank you” page or a particular event for example).
Add as many previous steps as you wish, selecting Page title and screen name to uncover pages visited.
You can then explore the most common customer journeys that brought your users to achieve the goal you have set for them.
Map out the ideal user journey visually
Analyzing the GA4 path exploration report is a good start point. This maps out the existing path taken by your users.
From here, you have two choices:
- Improve their existing journey with incremental changes.
- Map out the ideal user path that removes unnecessary friction to getting their job done.
If you choose No. 2, follow the same approach as above to map out the better pathway a user should take. Clearly state where friction exists at each step, and how content or site search personalization could help address it.
Here’s a simple customer journey template you can replicate to help you do this.
If you're looking for a visual design tool to use, check out this free Mural template.
Set personalization trigger
So what’s a trigger? A personalization trigger is a point or action taken in the user journey where you can start to control what content that user should see. When the conditions of a trigger are met, an action occurs that personalizes the site’s content for that user at that point.
With the user journeys mapped, you can now set a trigger(s) that switch on content personalization. The triggers indicate when user segments start to get recognizable at specific points in their journey.
Anonymous and authenticated triggers
This type of trigger allows you to personalize by user interest, or interest inferred by behavior. It will trigger a personalized experience that tends to be channel-specific (i.e. website only) and will remain until users clear their cookies.
You can collect just enough information to segment your users with the following methods:
- First-party cookie — your website can store information about your user with their consent, and remembers their preferences for each visit.
- Self-identification — Give users the option to select a ‘profile’ type (i.e. ‘I am a student/uni staff/alumni’) at the top of your website to automatically display relevant information to that user type and a tailored website navigation.
- User location — where a user is located, based on their IP address. segmenting your users this way makes sense if their location is a key factor influencing their expectations or behavior. For example, a council site will display content recommendations to residents based on their postcode, or a university shows content relevant to foreign students such as accommodation options near campus.
- Web event — specific actions taken on your website, including page visits or clicks, can trigger personalization.
- Device type — the device (i.e. mobile phone vs desktop) your users browse on helps you prioritize your web development efforts. For example, a university student using a mobile device is more likely to look for “on-the-go” information (i.e. campus map, library hours) rather than browse the library website to read a specific paper.
- Traffic source — the referring channel of a visit can help you determine the type of content you should serve to a user. For example, paid traffic is typically geared toward content that drives conversions (i.e. purchase a product or service).
Authenticated triggers normally use user-specific data to personalize content. The benefits of this type of personalization can work across channels, are persistent vs. cookie-based, tend to combine data sources (i.e. CRM information) and can control access to restricted content.
You can exactly identify your users with the following opt-in methods:
- Login authentication — whether you need to integrate with an account via single sign-on (SSO) or a similar login feature. This is the most secure way to personalize content.
Important tips: Data privacy
- Comply with data privacy regulations — Ensure that your organization prioritizes the needs and concerns of customers when it comes to data privacy, only using data if consent has been given according to your regulatory requirements.
- Maximize the quality of data you collect and use — Collect and use only high-quality data that is relevant to your personalization efforts. Don’t ask for more than you need.
- Give customers control over how their data is used — Provide customers with options for managing their personal data and how it is used.
- Make sure personalization is used for high relevance — Ensure that personalization efforts are relevant and valuable to customers.
Select content to personalize
Let's break down the most common website content items you can personalize. There are different content items you can focus on. Starting with just 2-3 instead of all of them is a good idea - especially at the beginning of a project
*Refer to details on each personalization item in the table below.
|Titles||Optimize your titles to better resonate with your user segment and invite them to explore the page. Good personalized titles should lower your bounce rate.|
|Images||Images can help a user confirm that the content on the page is indeed addressed to them. For example, images that match the demographics of the user (i.e. business attire people for an executive or director user segment) will improve the relevance of the page to them.|
|Copy||Perhaps the most basic form of personalization - displaying different text blocks based on the user’s segment. Marketers do this all the time for their landing page headlines, depending on the type of ad the user clicked on.|
|CTAs||Perhaps the most basic form of personalization - displaying different text blocks based on the user’s segment. Marketers do this all the time for their landing page headlines, depending on the type of ad the user clicked on.|
|Navigation||No need to show your entire website navigation to every user. Instead, just show them what they are most likely to be interested in.|
|Dynamic data||Similar to text personalization, data personalization takes it one step further by showing relevant dynamic content to the user. For example, when a user signs up into a customer portal, he is served content that is only relevant to them (i.e. their personal information or a recent service order)|
|Fonts and layout||In some cases, users are sensitive to on-page design personalization. People with poor eyesight might want to increase the font size, or tech users might want to use the dark mode of your website.|
|Product recommendations||Typically used (but not limited to) on product-centric or retail websites. Product recommendation widgets help guide prospective customers toward relevant products and solutions they are most likely to purchase, based on their profile, the history of their visit, product data (i.e. price or availability).|
|Overlays and popups||Highlight a particular promotion or alert using a large, prominent popup.|
|Notifications and widgets||Serve as subtle, unobtrusive elements in a corner of the screen, or as floating bars and sliding drawers.|
|Search personalization||Populate search results according to visitor preferences and real-time behavior.|
Map your content to your user journey
Not all of your content should be personalized for each user segment. As mentioned earlier, after a specific personalization trigger point, define what content becomes relevant to your user segments to get their job done faster or easier.
For example, if a specific user segment’s job-to-be-done is to request support for a specific service or product they need help with, their anonymous personalized content journey could look like this:
- Arrive on homepage: If a cookie is accepted, it identifies that they are coming from a specific location. The homepage banner shows content relevant to this geography. They also indicate that they are in the Marketing team and interested in a product by searching “Solutions for marketers” in the site search bar.
- Search results page: This brings up site search results that personalizes results based on that product that is relevant to the Marketer user segment and location.
- Product or Service page: The product page selected would then be personalized with relevant information focused on the Marketer segment. A Contact Us button component could also be an item on this page. The user clicks on this to get to the Contact form.
- Contact page: They arrive on the Contact page where the form has automatically pre-filled the product field based on the page they just came from. Local support phone numbers or addresses can also be personalized on this page based on their location
Content personalization plan
Once you’ve highlighted what needs to be personalized, it’s time to create your content for each applicable user segment.
In order to not miss anything, we recommend being very visual like in the example below, and clearly indicating what will be content for each section or component you are personalizing.
Content planning tip: Audit existing content to avoid duplication of effort
Audit the content you already have. This audit is important so you know what you already have, what can be reused vs. unnecessarily creating something from scratch. The three main categories of content in this audit should include:
- Redundant content - to be deleted as out-of-date or irrelevant to the journey.
- Reusable content - to be reused as is or tweaked to meet personalization needs.
- Gaps - not available and needs to be created.
**Site Search Analytics - This tool helps you analyze your site search usage. Knowing what users are searching for on specific pages indicate content gaps or relevant information they seek to achieve their objective.
‘Start small’ tip: Pick no more than two or three of the content items listed to start your journey. It is better to pilot quickly and see results early. This will allow you to observe any changes in user behaviour, to understand what’s required of your team, and to prioritize what to work on next.
Now that your personalization is in place, it’s time to measure its effectiveness against the goals you’ve set for each user segment.
A common tool to measure personalization efforts across your website is Google Analytics 4.
Define your audience
First, define each of your user segments under Audiences within the admin panels.
Click on New Audience.
You can use pre-made GA4 audiences or select Create a custom audience.
For each segment, you will need to carefully select what criteria GA4 should use to identify its users, including source, page visited, event, etc
In the example below, we are creating a segment of users who live in the United States and visited a particular page on our blog.
Back in GA4 main panel Reports > Engagement, you can then measure each user segment against a set of selected metrics, such as Conversions, Engagement, etc, or whatever you decided to measure your personalization success against.
In the example below, we are looking at a particular audience segment selected on the right end side. You can see how this audience is converting compared to a previous period (before/after personalization).
At a more granular level, you can set up a Funnel exploration (Explore > Funnel exploration) report to measure that each step (page visited or event) within your funnel is indeed improving over time against a set of pre-defined metrics.
In the example below, our goal is to increase page views all along the funnel, until the last step that focuses on conversion.
Template: Putting it together
Here’s a simple spreadsheet to put all the steps above together.
Make a copy of this web personalization strategy template to use for your plan.
Bonus content: Tools to put the strategy into action
With your personalization plan mapped out, the next step is to work out what tools you need to execute the strategy.
Whether you have the right tools in place already, or if you’re looking in the market for the perfect tech partner, keep these practical considerations in mind.
Increasing data complexity
- Delivering personalization has its challenges. The amount of data available to personalize experiences and the variety of sources they originate from could be extensive. Avoid getting trapped into pulling everything together at once. Most use cases don’t require real-time data. Keep it simple and only focus on data that’s essential.
Cross-channel interactions require specialized tools that connect fragmented data and workflows
- Today, most organizations offer multi-channel experiences that span websites, social, email, apps and phone. Collecting the right data, to segment personalized content to a specific audience group, in the right channel, at the right time requires the right tools to make the data and content talk to each other.
The interactions that take place across channels are difficult to measure and optimize
- Not only do interactions take place across several channels, media, and devices; sometimes a single customer can have multiple digital identities. This makes performance tracking and optimization challenging for marketing and digital teams.
But all is not lost.
To cope with this complexity, many technologies have emerged to the fore recently:
- Many look to their existing CRM to do the job if it’s robust enough - but it often requires many integrations to different tools by channel, making the integration requirements and maintenance intense and costly.
- A more functional tool is a Customer Data Platform (CDP) which makes customer data, regardless of source, more actionable. It enables you to:
- Collect customer data — pull together customer data from various sources (including your CRM) such as transactional, contextual, behavioral, and demographic data.
- Unify customer data — combine the data to create a single, coherent view of each customer.
- Share customer data — share the unified customer data with other applications, such as analytics, marketing, advertising, and customer engagement platforms
- Personalize customer engagements — organizations can leverage the data provided by a CDP to trigger personalization on their website via Components, a Content Management or Experience Management capability, and other marketing automation tools.
- The rise of the Digital Experience Platform (DXP) in recent years is a solution that was born to specifically address the challenge. It is designed to make building and orchestrating multi-channel digital experiences faster and easier - all from one place.
- Some DXPs, such as Squiz, have both Content Management, Integrations and Customer Data Platform capabilities as part of the platform. This combinines the best of both worlds - (a) unifying and segmenting all customer data in one place, and (b) enabling personalized experiences to be created and published across digital channels in a coordinated way, from one platform.
Most personalization projects fail because we try to take on too much, go into unnecessary levels of detail and increase the level of complexity that either takes too long to execute or fails because the effort is not sustainable.
This step-by-step guide will empower you with a practical roadmap with ready-to-use templates and actionable ideas.
- one or two user group segments to map journeys for
- one or two personalization triggers relevant to the most important jobs-to-be-done for those segments
- two to three content items to personalize per segment
- two to three key measures of success.
By doing so, you make the personalized experience great for your highest priority users quickly. You get to clarify team responsibilities to ensure it’s sustainable. You can also establish manageable workflows that scale.
So, start with something small.
Do it quickly and learn.
Then iterate and show results that will enable you to expand to bigger initiatives.
Watch below our "Step-by-step" web personalization webinar below.