MailerLite
Single Sign on and cross platform switching improvement.
Consistent authentication across three products and a shorter switching path improved access, with time savings estimated retrospectively.
MailerLite’s growing ecosystem included MailerLite, MailerSend, and MailerCheck, but signing in across those products felt fragmented. Each offered different authentication options, and the underlying account architecture prevented a consistent single sign-on experience.
I worked with the product manager and lead engineer to investigate where customers were getting stuck, separate infrastructure problems from interface problems, and identify improvements we could deliver alongside the engineering work.
The shipped experience introduced consistent authentication across three products, retained customers’ last-used sign-in method, and shortened product switching from three clicks to two. A retrospective interaction model estimates approximately 3.7 seconds saved during switching, plus the eliminated intermediate page-load time. SSO-related complaints also declined after launch.
My role & scope
I owned journey investigation, customer research, interaction design, experiment design and front-end implementation alongside product and engineering.
I owned the journey investigation, customer research, interaction design, experiment designs, and front-end implementation.
Working closely with the product manager and lead engineer, I:
- Reviewed support tickets and PostHog session recordings.
- Mapped sign-up, login, account creation, and switching journeys across the three products.
- Helped frame the problems and identify opportunities through joint research and brainstorming.
- Designed authentication improvements and alternative product-switcher placements.
- Contributed to the front-end implementation of the resulting experiences.
The underlying SSO architecture required engineering work. My contribution focused on understanding its effect on the customer journey and designing the experiences around the improved system.
The problem & why it mattered
Fragmented authentication and hard-to-find sign-in options interrupted access across the MailerLite ecosystem and generated support requests.
Leadership initiated the work to improve single sign-on. Support tickets gave us a more specific starting point: customers were struggling to sign in and move between products.
MailerLite, MailerSend, and MailerCheck offered different sign-up and login options. Customers could encounter a different authentication experience depending on which product they entered, and moving to another product could require another login.
Some customers also struggled to find social authentication options. Those options were widely used, but the layout could place them below the fold.
These difficulties interrupted access to the products and generated support requests. They also made the ecosystem harder to use as a connected set of tools.
The initial brief was broad: improve SSO. Our first task was to establish exactly where the journey broke and why.
Evidence & key insights
Support tickets and session recordings showed architectural fragmentation, hidden authentication choices and an opportunity to make cross-product switching easier.
I reviewed customer complaints alongside PostHog session recordings, then investigated the journeys with the product manager and lead engineer.
Three findings shaped the work.
The fragmentation extended into the architecture.
The account databases and SSO infrastructure could not support the seamless experience we initially wanted. Interface changes alone would leave an underlying source of friction unresolved.
Useful authentication options were difficult to find.
Social sign-up offered a faster path for many customers, but some reported that they could not see it. The screen layout was affecting access to an existing capability.
Better SSO would create an opportunity to improve product switching.
Once customers could reuse their authentication across products, switching no longer needed to feel like leaving one service and signing into another. We needed to make that capability easy to discover.
These findings changed the scope of the work. We had both an infrastructure problem to address and customer-facing improvements to pursue.
Constraints & trade-offs
We pursued architectural improvements and customer-facing changes in parallel while balancing screen space with discoverable authentication choices.
The architecture work was substantial. We agreed to request additional engineers to address it in parallel while we explored improvements to the customer journey.
That gave us two connected streams of work:
- Engineering addressed the fragmented authentication foundation.
- Product and design improved how customers signed in, created accounts, and moved between products.
We also needed to balance a compact authentication screen with access to multiple social providers. Showing every provider separately took up space; hiding them without a clear entry point would make discovery harder.
The design needed to expose the important choices within the initial viewport while keeping the available options accessible.
Design strategy & key decisions
We questioned the brief, exposed authentication choices, remembered returning customers and tested a shorter path between products.
1. Question the brief before and finding assumptions before choosing a solution
We started by asking: Which problems are customers experiencing, and what is causing each one?
That investigation established which issues required architectural changes and which could be improved through the interface. It helped us avoid spending time refining screens around behaviour the system could not reliably support.
2. Reduce visual bulk and make authentication choices easier to find
I compacted the sign-up layout so customers could see the authentication choices without scrolling.
Rather than displaying four separate social-provider options at once, the design used a compact control that revealed the available providers. The last-used social login option remained visible when a customer returned.
The aim was to make the first choice easier to find and repeat visits more familiar.
3. Unify authentication across the ecosystem
We aligned the sign-up options across MailerLite, MailerSend, and MailerCheck.
A customer moving between products should encounter a recognisable authentication pattern. Consistency reduced the need to interpret a different set of choices at each entry point.
4. Make product switching available where customers already manage their accounts
We compared three placements: top-bar icons, the profile dropdown, and a dropdown beneath the sidebar logo.
Top-bar icons were fastest to discover and use. We selected the profile dropdown, which ranked second while fitting the product’s existing design patterns and providing a scalable location for additional products.
The selected route removed the detour through the centralised profile page, reducing switching from three clicks to two.
The experience
The shipped journey made authentication consistent, remembered the last-used provider and reduced the path to connected product accounts.
The shipped experience connected several parts of the account journey.
- Sign-up and login used a more compact layout with consistent authentication options across the three products.
- Returning customers could see their last-used social login option, making it easier to recognise how they had previously accessed their account.
- Product switching became accessible from the profile dropdown, shortening the path to another product.
- We enhanced the profile page and streamlined cross-product account creation to support connected accounts, cutting the steps to create a new account in half.
Together, these changes made access to the ecosystem more coherent—from entering the first product to moving into another.
Validation & iteration
We tested first-time entry and existing-customer switching separately, with provider-selection savings estimated at 1.4–2.5 seconds excluding authentication, loading and onboarding times.
We tested the authentication and switching journeys with different customer groups.
New users tested the sign-up experience.
We tested a compact authentication flow placing social sign-in above the fold with consistent options across products. Returning users saw their last-used provider. Estimated time saved selecting a provider is 1.4–2.5 seconds by removing scrolling and extra visual search. This model excludes authentication, loading, and onboarding times.
Existing users with accounts across multiple products tested switching.
We tested three switcher placements to see how easily customers could find the control and switch products quickly. The options included previous methods: logging into each product separately or navigating through the profile and centralized account page. Testing showed top-bar icons were fastest, with the profile dropdown next. We chose the dropdown for its balance of speed, consistency, and scalability.
Testing these journeys separately let us assess two different needs: helping someone enter the ecosystem and helping an existing customer move within it.
Outcomes & impact
Shipped improvements shortened switching from three clicks to two, with retrospective time estimates and a qualitative decline in SSO complaints.
The shipped work included:
- Improved sign-up and login experiences.
- A more accessible product switcher.
- An improved profile page.
- Clearer cross-product account creation.
The experiments and Interaction model reported:
After the changes shipped, SSO-related complaints were substantially lower. This is a qualitative result; a precise reduction has not been established for this case study.
The results reflect coordinated design and engineering work. Architectural improvements enabled more seamless authentication, while the interface improvements helped customers find and use it.
Reflection & next steps
Separating architectural friction from interface friction helped us choose useful changes, with continued evaluation of completion, switching and support requests.
The most important decision happened before the screen redesign: establishing what was actually causing the friction.
The investigation helped us separate problems that looked similar in support tickets but needed different responses. Some required engineering changes. Others involved visibility, consistency, or unnecessary navigation.
That reasoning connects with how I approach Build and Think Better: question the requirement, remove unnecessary work, and simplify what remains. Here, it meant investigating the broad SSO brief, reducing the effort needed to find authentication options, and shortening the route between products.
For future evaluation, I would track authentication completion, switching behaviour, and related support requests together. That would make it easier to understand how improvements perform beyond the initial experiments.
