Financial marketers can often narrow in on the following channels to analyze data and performance for potential optimization:

• Performance marketing
• Social media
• Referral (trusted sources)
• Direct traffic
• Physical locations
• Mobile app
• Website
• Email
• SMS
• Paid and organic search

Response channels (outputs) are based on the same list above but are used for different purposes depending on the goal in mind, which we’ll go into detail about later.

Bring insights gathered from the previous step and discuss new testing opportunities together during regularly scheduled brainstorming sessions.

Start by creating a list of all the different broad categories you have the ability to test. Then, combine that list with all the kinds of tests you have the ability to run. This allows teams to stay grounded, reminding them of what’s possible, aka what experiences can be tested and whether the means are available to actually execute them. It may also draw out potential experiments that may have otherwise been overlooked.

• Banners
• Content
• Layout
• Offers
• Overlays
• Emails
• Push notifications
• Merchandising rules
• Messaging
• Recommendation strategies
• SMS activations

For example, if data pointed to low conversion rates from performance marketing campaigns, teams might consider launching various experiences targeted at those users in hopes of driving greater engagement. Teams should think about the “why” behind a particular test as opposed to focusing on the “what.”

Hypothesize changes that would lead to the desired user behavior for each audience segment, building out use cases to improve or maintain that metric. The goal is to move each segment into the next logical audience through completion of the KPI.

A good way to think about building such a hypothesis is:

Based on [OBSERVATION], I believe that if we [PROPOSED CHANGE], we will see an increase in [TEST PRIMARY METRIC].

These statements should not be confused with facts. They are not facts, but rather argued ideas based on data analysis and past observations. They act as a starting point for the optimization process.

These hypotheses should be closely tied to the “why” or goal supporting the test and often that will be to move each segment into the next logical audience through completion of KPI.

Some examples of personalization KPIs for financial services include:

• Open accounts click / apply click / get started click
• Application start or complete
• Form submission
• Mobile app downloads

A personalized customer journey should feel cohesive across multiple webpages or touchpoints and be responsive to the user’s behavior each time they interact with your brand. Knowing this, best-in-class personalization campaigns will test multiple channels and audiences across the customer journey, involving multiple pages and content elements, offers, notifications, contextual nudges, product recommendations and more. The important thing is to always have a control group to learn what works and doesn’t for each audience.

With all of the excitement and speculation around the outcome of a particular experience, it’s easy to get carried away. But with any initiative, it’s crucial to consider each initiative’s estimated impact (from costs to potential excitement generated) and effort (which teams will need to be involved) before pulling the trigger. Otherwise, you can end up wasting valuable time and energy on unrealistic campaigns or those that offer little or no value to the business — both of which can hurt the long-term sustainability of a personalization program.

The rating system below should be used to aid the decision-making process, ensuring you are able to maximize each campaign.

Impact:

Rated holistically, can be measured by High, Medium or Low.

• Alignment with company objectives
• Legal and privacy considerations
• Estimate time to impact
• Effect on the purchase funnel
• Level of excitement generated
• Endorsement by an executive sponsor
• Ability to reduce costs

Effort:

Rated holistically, can be measured by High, Medium or Low.

• Teams involved
• Code or test complexity
• Test definition
• Time to QA

Once you evaluate each test, use this as a simple North Star:

• High impact + Low effort = Do it now
• High impact + High effort = Plan for the long-term
• Low impact + Low effort = Plan for a quick win
• Low impact + High effort = Drop

Eventually, you’ll end up with a scorecard for which campaigns to prioritize. Prioritizing these tests is key to reducing design and development cycles for the quick and effective implementation of campaigns. If the scope of an experiment doesn’t line up with the testing schedule, deadlines are bound to get missed and the user information associated with a campaign will go stale, resulting in lower-impact results. Not to mention, a poor customer experience.

Then, the building officially begins.

Now, both creative and technical functions must align to create the intended customer experience. Typically, the following workflow is adopted to first mockup or “stage” how the final experiment will look:

1. Copy and calls to action are drafted
2. Designs are mocked up
3. Specs are built

Once an experience is developed, it will then go into review, ensuring it meets all of the requirements agreed upon.

Just as important to the success of a campaign as all the work that led up to launch is closely monitoring, analyzing and optimizing performance upon going live. How long an experiment will have to run in order to achieve valid results will vary based on the sample size and number of variations served, but two weeks is recommended at a minimum.

After an experiment officially concludes, the variable insights gleaned will shed light on the underlying reasons behind the results for each segment of users. These should be logged and used to continue building out various audience profiles, informing the modification and optimization of running campaigns as well as completely new tests.

And then it’s back to step one! Personalization efforts are cyclical; as one test concludes, companies will find themselves at the beginning, where they will yet again begin to analyze collected data, brainstorm additional ideas, hypothesize new or improved changes and so forth.

Acquiring, activating and managing customers throughout the customer journey is no longer predicated on a bank’s physical location or in-person service, but rather on the timely, relevant and personalized digital interactions that consumers expect every time they engage, no matter the channel.

Now that you understand the personalization framework, we’ll exemplify various tests FIs can perform across the customer journey using engagement to segment traffic into four primary audiences: Low, Medium, High and “Other” (defined by the stage for which you’re segmenting).

Use the following information on which KPIs, channels and types of experiences correspond best with the different lifecycle phases to accelerate time to market, more quickly understand what performed well and refine campaigns at every step of the customer’s relationship with your company. Greater sophistication in your approach to the customer experience awaits, as do the exponential revenue gains.

Within the customer acquisition phase, financial services brands most often seek to reduce customer acquisition costs (CAC), which can be done by leveraging available data to deliver relevant experiences from the very first page view.

Customer acquisition channels for personalization include:

Audience segments:

• Low engagement: Characterized by limited data, they could be unknown, first-time visitors or users who have yet to engage with your website. 
• Medium engagement: Shows signs of product interest or affinity, either visiting multiple pages, returning to the site or converting on a download or on-site tool.
• High engagement: Shows interest and affinity by taking multiple actions across the site and completing several key customer behaviors.
• Other: Identified customers. 

Example experiment:

Let’s say after digging into our data, we find a large portion of visitors are coming directly to the site (trigger channel) from a few distinct locations (low-engagement audience segment). We can then use that data to inform the personalization of web experiences (response channel), including the tailoring of welcome messages, homepage banners, email signup overlays and more to encourage customers to engage more, for perhaps by submitting applications or using free financial tools.

For example, one might serve a homepage hero banner promoting business insurance to people in New York City and another version for agribusiness coverage in Nebraska. By doing so, teams can ensure greater relevance right out of the gate, a higher likelihood of continued engagement and lower rates of site abandonment.

Other acquisition personalization use cases include:

• Offering service assistant messaging to prevent drop offs upon form completion
• Harnessing guided selling to educate and convert
• Tailoring on-site welcome messages according to traffic source
• Surfacing idleness-triggered reminder notifications on form completion pages

Adoption is all about driving important activities early in the product relationship. This phase must be viewed as a series of engagements to be managed in real-time, using available data and customer cues to adapt to the various journeys (for example one user registers their card first while another downloads the app) and encouraging the completion of desired next engagements. Every new customer is unique and has different needs, preferences and reasons for their behaviors, including how and when they engage.

Goal: Establish significant markers in the customer relationship, like account activation and setup, to encourage early product usage.

Audience segments:

• Low engagement: Has not yet completed account set-up or activation.
• Medium engagement: Has completed account set-up and built a profile.
• High engagement: Has completed account, built a profile and set up functionality like autopay.
• Non-customers and long-term customers: Visitors who are not identified as customers or have been identified as a customer for a certain period of time, for example, longer than 90 days.

Example experiment:

Consider a scenario where a new customer who has not yet registered their card (low engagement customer) visits the site (trigger channel). Instead of receiving the generic homepage experience, a team could use the information to A/B test which messaging, content, 

Within the ongoing lifecycle management phase, it’s all about maintaining customer engagement and top-of-wallet behavior as well as increasing account and card usage. Through this lens, you can start to optimize for relevant KPIs like overall spend, cross-sell and upsell opportunities, card and account usage and other key consumer behaviors that drive longevity, like automatic deposit, multiple accounts and mortgage loans.

Customer adoption channels for personalization include:

Goal: Upsell, encourage product and service usage and drive customer lifetime value.

Audience segments:

• Low engagement: New credit card users
• Medium engagement: Users with two credit cards with one year or more history
• High engagement: Mortgage account holders
• Non-customers: Visitors who are not identified as customers
• Example experiment:

Let’s say we have a new credit card user (low engagement customer) who visits the website (trigger channel) and is delivered relevant recommendations and content (response channel) that’s in line with their current situation to encourage their overall spend as well as their card and account usage (KPIs) rather than a generic experience.

As time goes on, the customer evolves into a homeowner (high engagement customer) and the website experience (response channel) changes to incorporate more cross-sell and upsell opportunities, encourage educational downloads for relevant but undiscovered products and promote key customer behaviors like home loan applications or refinancing (KPIs).

In the above example, the variation on the left incorporates website personalization to serve a tailored experience for a returning customer with specific content recommendations that correspond to the available customer data. However, the experience on the right for a new customer showcases generic and broad resources in an effort to learn more about their needs and interests and to serve deeper, more relevant recommendations in the future.

Other lifecycle personalization use cases include:

• Displaying recommendations based on customers’ spending power and financial preferences
• Upselling customers with personalized complimentary offers and product recommendations
• Creating an omnichannel customer journey experience
• Recommending unexplored products and services to customers that are unfamiliar with them