blog post banner image

Using A/B Testing to Optimize the Banking Experience

Will Lam


Bank customers are some of the most risk- and change-averse users you’ll encounter. It makes sense: banking institutions and FinTech companies deal with some of the most sensitive personally identifying information (PII) on the market—their customers’ hard-earned income.

According to a 2020 J.D Power survey, consumers who receive personalized recommendations via online or mobile banking apps are more likely to take part in actions that increase their engagement. While these personal recommendations add value for both customers and financial institutions, making any change to the banking experience has to be done carefully.

To build trust with this audience, you need to nail the banking experience. That means providing all the security and safety of a financial institution while also making relevant and personalized recommendations. But how can they do that when customers will react negatively to sudden or unforeseen changes? That’s where targeted A/B testing comes in.

When you understand how to create effective A/B tests around the banking experience, it mitigates risk to your company and ensures the best possible results for consumers. You’re able to make incremental changes without impacting your entire customer base every time. Creating these tests comes from an understanding of what to change, and how.

Determine What Parts of the Banking Experience to Change

The digital banking experience spans touch points across a number of communication channels. Each step of the process has the potential to either build your relationship with users or detract from it. Before you create any A/B tests, it’s important to understand where these tests can make the most impact. This helps you provide more value to users quickly and prioritize which experiments to run first.

Making the decision to run an A/B test requires a solid understanding of users’ current experience as well as a strong base of PII and customer behavioural data to inform your decisions.

Document the Current Experience

Every successful A/B test starts with a concrete understanding of the current customer experience. Documenting this information in a centralized location ensures that your team always has something to refer back to as they build experiments. Think about it as a single source of truth—a place where your team can reference the various touch points that make up the banking user journey and see where to make improvements.

This documentation starts by compiling a list of what you want to change and then connecting it with a metric that corresponds with the intended outcome of your experiment. Here are a few examples:

Touch points:

  • Account engagement
  • Customer retention
  • Upsell potential


  • Time on page
  • Revenue churn rate
  • Lifetime value (LTV)

Connecting each touch point with a specific metric makes it easier to measure your success and track how each change impacts the overall banking experience. Let’s say you want to find new ways to increase the upsell or cross-sell potential in a segment of your user base. One way to track that increased potential is by looking at LTV. If your experiment increases the LTV in your target customer segment, that would be a successful test.

Understand Your Users’ Needs

Understanding what your users want and need is the key to building effective A/B tests and helps your team see how their work adds value to the banking experience. As you compile a list of the specific actions and metrics you want to affect with your A/B test, start digging into the shared characteristics of your target audience as well.

Whenever you make a change, it’s important that it add immediate and quantifiable value to the banking experience.

But this deep customer insight grows over time. Look at bottlenecks in the experience with your product and talk to current users about their wants and needs—that helps you nail down specific pain points and potential areas of improvement.

Once you have this knowledge, you can segment users based on these pain points. Structuring your tests in this way not only makes it easier to build trust with users but also helps mitigate the potential risk of a botched test.

Let’s go back to the upsell/cross-sell example from earlier—audience targeting for these tests is essential for your success. If your goal is to increase LTV, then you need to target someone who is in a place to open up a CD or a money market account based on a high recurring balance. Offering that option to someone with a lower balance would be a bad experience and a failed experiment.

5 A/B Tests That Can Improve the Banking Experience

Every A/B test you run on the banking experience should change only one aspect at a time. The more narrow your experiments, the easier it is to learn something meaningful from the results. We’ve put together five example A/B testing ideas to help you get started.

Test 1: Engagement-Based Push Notifications

Push notifications are a great way to increase engagement with your mobile app. A/B testing the language and content of these messages helps you target specific customers with relevant information and offers that make the banking experience more valuable.

One way to accomplish this is to send notifications to disengaged users to remind them what kind of value they receive from their account. This could be as simple as notifying someone of a recent purchase flagged in their account, or a proactive reminder to update information in their contact information for better personalization.

You can also test reengagement notifications for users who haven’t logged in to their accounts via the app for 30, 60, or 90 days. For more information on these types of notifications, check out our ultimate guide.

Test 2: Personalized Recommendations

You can improve customer retention, increase revenue, and boost LTV by providing email and push notification recommendations that showcase the value of a user’s current account.

Test these offers with engagement-based push notifications, or set up triggered emails using marketing automation to find what channel works best. The digital banking experience often bounces between different touch points across channels, so it’s important to experiment in order to find what type of messaging works best for specific customers.

Use customer data to identify specific customers who match criteria for higher-value accounts. This could be something as simple as offering CDs and other interest-based accounts to users with high recurring balances. Or you can set up a way to schedule personal outreach from a member of your team.

Either way, using customer data and PII is a great way to refine the banking experience based on your customer’s needs.

Test 3: Intuitive Account Design

Online banking accounts are often difficult to navigate and require users to click through a number of different menus to find the right information.

Test your top-level navigation bar to highlight areas of the account your users access frequently. Bank of America, for example, uses account type and customer type to organize their information.

Small-business customers get Checking and Savings first as well, but then are directed to Business Services.

Bank of America uses these navigation and call-to-action (CTA) buttons to provide relevant information faster. By offering the link to Business Services before Credit Cards, small-business owners can quickly access the areas of their accounts that matter most.

When you’re testing out these navigational links, look at past customer behaviour to see which areas of your site/app receive the most traffic from different landing pages.

Test 4: On-Page Written Copy

Whenever someone lands on your website—whether it’s a home page, a landing page, or another account-specific page—they should immediately understand how your product or service provides value. The quickest way to do this is with your copy.

Testing value propositions and other on-site copy, such as product descriptions and navigational links, helps you connect with users directly and add value to their banking experience. Take a look at the various credit card offers from TD Bank:

Each card communicates a different kind of value for specific customer types:

  • TD Cash Credit Card targets standard customers with a cash-back offer.
  • TD First ClassSM Visa Signature® Credit Card goes for more high-balance accounts with travel and dining perks.
  • TD Business Solutions Credit Card targets businesses and their expenses.

By using different language for each card, TD can easily show customers which card is tailored to their particular needs. When you create these A/B tests, try different words and phrases to drive home the specific advantages of one card or offer over another.

Test 5: Support Team and Content Accessibility

Easy access to your support team and content is a great way to level up your digital banking experience. Users should be able to get an answer to their questions quickly without having to wait on hold or dig through confusing menus to find the right support documentation.

So create an A/B test to figure out the optimal placement for your live-chat button and your FAQ documentation.

TD Bank, for example, includes a search option at the bottom of important product pages as well as a “Contact us” CTA:

By linking to commonly asked and popular questions, TD Bank makes it easier to sift through their self-serve support content. This creates a better experience for customers with simple questions as well as complicated ones. The simple questions are quickly answered via documentation, which frees up your support team to focus on more in-depth issues.

We can assume that TD knows which questions are most popular by tracking engagement with these pages as well as individual support interactions. It’s this kind of deep user insight that helps them test the language and structure of the page to ensure a seamless experience for any type of query.

Optimizing the Banking Experience Takes Time

When you create targeted A/B tests for the banking experience, you’re able to boost engagement with your product without risking a negative impact on people outside your testing group. Making incremental changes via these experiments also helps your team provide continual value to customers throughout the process, which is a great way to build stronger relationships with users over time.

Written By

Will Lam