Evite Teardown
Objective
One of the most important business drivers for Evite is continued growth in invitation production. To strengthen this core funnel for the business, the task at hand was to analyze the existing invitation creation process and propose changes for improved UX and usability. The end goal was increasing invitation production by helping users move through the invitation production process with less friction and more delight. For this exercise, I focused solely on desktop invitation production, where Evite experiences the most traffic and activity with invitation creation.
Process
Uncover Friction | After reviewing the free invitation production process from landing page through completion (see assumptions below), I compiled a list of friction points and low hanging fruit that, once addressed, should increase conversion through each step in the funnel. Then, I ran through the same free invitation production process with 3 strangers at a coffee shop and took note of their stumbling blocks and questions.
Assumptions for this exercise:
The free invitation creation process is the most common path for users
Most users are new to Evite and need to create an account to save their invitation
Prioritize Problems | Once I had a list of friction points and questions from users, I started prioritizing problems based on potential impact and then explored potential solutions worth testing. I started by prioritizing tests on the most highly trafficked page (see assumptions) in the funnel because if we can get more users to move from that page to the next step in funnel, it usually results in more people making it through the entire funnel (which is the end goal). In this case, you’ll see I prioritized early-funnel (landing page) tests to help more people move through the full funnel. I also prioritized streamlining the invitation selection (Gallery) and customization steps (Event Details) to clarify the path forward for the user and minimize fallout in each of those crucial steps. Perhaps the most high-impact change I prioritized was combining the account creation step (after clicking Save) with entering host details in the Event Details portion of the funnel. In my experience, keeping the user within the context of their immediate action when asking for account information is a much more effective approach than interrupting a key flow. Once these problems were validated (let’s assume with data) and prioritized, I began exploring potential solutions. See prioritized list of problems below.
Explore Solutions | Although the following design concepts show a blend of potential solutions, I would recommend rolling out each new change in isolation (ie don’t change everything at once) so we could measure which strategies have an impact on conversion through each step in the funnel and learn/adjust as needed. This phase is also where a stellar UI/UX design team really comes into play. You’ll see I’ve done my own explorations for this exercise (through to rough visual design concepts), but this exploration phase is where product, design, and development really connect.
After exploration and ideation, I would write user stories for design and work with that team until we have a solid prototype. Once we have a complete concept with all major use cases considered and a prototype that has ideally gone through user testing, I would move the work over to development where I would create user stories for engineers.
After development and UAT, we’d launch the changes and track their performance over time to see what could be adjusted for a greater gain and/or what needs to go back to the drawing board.
Funnel Analysis
After reviewing traffic patterns and drop-off points throughout this funnel, it became clear that there were some immediate gains available in creating a stronger first impression on the landing page, making search easier, condensing some of the page content in the customization steps, and more. With this new approach, users would have a stronger sense of direction and purpose throughout the invitation creation process, and there was 1 full page eliminated from the flow. Generally, the fewer number of pages a user goes through, the higher number of funnel completions you'll see--as long at the steps/pages they go through are high-value, purpose-driven, and easy to complete.
FIRST STEP > Landing Page Optimization
The existing Evite landing page was losing many users due to lack of clear direction into the invitation creation funnel, distracting ad placement, disorienting invitation design displays, moving videos that looked like ads above the fold, and more. This page needed a revamp to give users a more defined path forward and help them see using Evite as the first step toward hosting an incredible event, Users needed to see an end in sight, a desirable and easy outcome (event). Here's what the old landing page looked like...
NEW Landing Page Concept
Includes | 1. Stronger CTA to create invitation, outcome-based imagery, 2. New How It Works section, 3. New social engagement section, all designed to get users thinking about the end result of their product usage (i.e. the amazing, memorable event they are about to host). On page load, only the hero image, main CTA, and event menu are visible, scrolling shows the user more info about how it works, what they can expect with invitation quality, fun ideas for social engagement and ongoing interactions (after all, these events are supposed to be #hashtagfun), and space for ad revenue.
NEXT STEP > Search Results Optimization
The existing search results step of the invitation creation flow featured a paginated display of invitations with free and premium (paid) invitations mixed together with ads and a limited number of search filters. The experience was jarring--it was difficult to tell what you were looking at, and the ads mixed in made users feel like they were getting tricked, and it cheapened the whole experience. The only way to pre-filter this selection of invitations by event was to use the event menu on the landing page or the event menu on the search page, but selecting an event was not a required step... so lots of users were getting to this search results page viewing a lot of different event types. There also wasn't a way to filter invitations by style (e.g. modern, floral, minimalist, etc). There was a big opportunity here to make free vs. premium more clearly defined, cut down fatigue by showing users invitation styles that were most relevant, as well as an opportunity to make selecting the right invitation an easier process.
[NOTE: This is the first step down a path toward personalization. When someone sends an invitation, they are sending something that represent themselves just as much as they are sending something that represents an event, so choosing the right style and message is very important to users. This is where personalization could really come in handy down the road. Personalization (e.g. recommendation flow) is not explored in this teardown, but would be an exciting opportunity for testing.]
Here are the old/existing search results:
NEW Search Results Listing
Includes | 1. Event selector 2. Style selector 3. CTA to design your own invitation 4. Definition between Free and Premium (value of both options) 5. Ad space for revenue separated from invitation products (mitigating risk of ad fatigue contributing to lower invitation completion rates).
NEXT STEP > Invitation Creation Flow
In the old/existing flow, once the user selects an invitation to start customizing, the user is sent down a multi-step flow that asks the user to progressively provide more information about their event. As part of this flow, the user creates their account, establishes all the details of their event, selects add-on options like registry/gift options, and invites their contacts to attend. This process also takes a new user completely out of the invitation creation process to create and confirm their account before returning to their invitation to finish and send. This proved to be a very disjointed process that provided many opportunities for the user to bounce out and never return to complete their invitation. There was a great opportunity here to combine steps, make account creation easier for the user, and keep the user engaged in the task at hand so more users make it through to completion.
Here is the old invitation creation flow:
NEW Invitation Creation Flow
Includes | 1. Rearranged information entry points 2. Fewer steps overall, combined account creation steps 3. Clearer path to completion with 'Next steps' copy 4. More visibility into add-on features like gift registry/donations (i.e. value-adds to the user) and status once the invitation has been sent. By the end of the flow, the user is celebrated for completing the task and the "congrats" page provides space for ad revenue and other engagement features (like the hashtag generator). In this flow, the account creation process and various ads do not impede the user or block them from continuing with the job they "hired" Evite to do (i.e. send an invitation).
Develop > Launch > Learn
These are only a handful of the recommendations I'd make for quick-win A/B tests for the Evite invitation creation funnel. Although the ultimate goal of A/B testing is to improve the experience, functionality, or accelerate business needs associated with a product, it’s important to keep in mind that oftentimes what you learn through failed tests is just as valuable as what you learn through successful tests. I make a concerted effort to keep my team cautiously optimistic but not too invested in any single test variation--it helps keep everyone flexible and hungry to continually make processes better. The more data you can gather during a testing cycle the better, especially if you can add color to that data with qualitative human feedback from real users. If I were to develop these proposed changes in real time with the Evite team, I’d be sure to reinforce that failing fast is a good thing, feedback from users is essential, and constant testing requires equal parts resilience and creativity to be successful. That’s what makes the challenge worthwhile and fun.