Eppik
Increasing user engagement on a photo digitisation platform
Client group project as part of General Assembly’s UX Design Immersive course. This was a 2-week design sprint.
The brief
Eppik is a photo digitisation platform. They bring old photo archives back to life by digitising them first and then empowering users to build their family narrative. Currently users engage heavily within the first few months and then drop off. Eppik would like to explore how they can offer more long term value to customers.
My role
My teammates and I had an excellent working relationship. Each of us is very task-focussed, diligent and believes in evidence-based arguments. This made for an easy team dynamic and an extremely efficient and high-performing sprint. We held daily stand ups and stand downs and weekly retrospectives. As an innate diplomat, I often played the role of mediator to help push past the few back and forth discussions we did have by proposing workable solutions that got everyone aligned on a way forward.
Discovery phase
The discovery phase began with reviewing the business requirements outlined in the project scope document before exploring customer needs through qualitative user interviews. This was supplemented with competitive review.
Business needs
The business believes that it is important to get users to add metadata to their photos and albums because it attaches meaning to content that can then be used to drive longer term engagement. They hypothesised that the best way to do this is through a mobile app.
Competitive review
We took a look at direct and indirect competitors to understand which features they offer that keep users coming back. We found a few common threads amongst the biggest and most engaged platforms:
- Easy to use and accessible across multiple devices
- Strong mobile app
- User name identification to allow for network effect
- Use of notifications to bring back or keep users on the platform
User interviews
The next step was to understand users’ behaviours and motivations with regards to photos. We conducted user research with both existing users of Eppik and prospective users. Our activities included:
- Survey sent out to our and Eppik’s networks with 119 responses
- 5 qualitative user interviews with non-Eppik users
- 4 qualitative user interviews with Eppik users to test usability of existing website
Key insights from survey and interviews
We put the feedback on sticky notes and used a process called affinity mapping to expose patterns and trends in our research.
The key findings were:
- Main motivations for keeping photos is to preserve memories
- People share photos to participate in each other’s lives
- Whatsapp is the main medium for sharing photos
- Triggers include notifications, memories and widget features
- Greatest need is having photos all in one place
- Biggest pain point is searching for photos
Key insights from usability testing.
Users unanimously loved the museum-grade image quality and thought that the platform was a great way to rediscover old photos and preserve memories.
However, they were turned off from using the site and wanted:
- Better and more sharing options
- Easier way to add and view information
- More intuitive layout and navigation
- Accessibility across multiple devices
- Personalised categorisation
- Smarter search
And the biggest pain point was how long it took for photos to load — case in point here…
Persona and problem to solve
To help keep the users’ needs front of mind throughout the design process, we created a persona based on the users we interviewed.
To imagine what it would be like in Karen’s shoes, we imagined her experience in this scenario:
Karen has found a shoebox of old photos at her parents’ house. She wants to digitise them so she can preserve the memories and share them with her family and friends.
It’s clear from Karen’s emotional journey above that her greatest frustration and our problem to solve is around easily adding metadata to the digitised version of the physical photo for use later.
Ideation and sketching
We took the problem above to a design studio workshop we held with the client. To help spark ideas, we created what is called “how might we” statements to help us think of ways we might help Karen solve this problem:
- How might we prompt Karen to add information to her photos?
- How might we add structure to Karen’s photos using AI?
- How might we make organising photos more fun for Karen?
The client loved this and it sparked a lot of discussion and ideas around possible features for the mobile app. We dot-voted on those ideas to narrow them down into a list of features for consideration.
Feature list and prioritisation
We created a list of features and prioritised them based on those that we believed would be quick wins to drive user engagement as suggested by our competitive and user research.
The key is engagement — the hook model
To really understand how best to do that, we analysed these features through the lens of a “Hook model” which at its core is about creating user habits through four stages that keep users coming back again and again.
Navigation
We needed to re-think navigation because our testing with Eppik customers showed that the existing website layout is not intuitive but mainly because mobile interaction is not the same as desktop.
This led to some internal debate and so we turned to a UX tool called card sorting to inform our decision based on user feedback. An open card sort was tested on seven users to understand similarities and differences as to how they might expect information to be organised.
Based on our findings, we finally agreed on the following app map.
Ideal user flow through the app
We put ourselves back in Karen’s shoes to outline the ideal user flow through the app. Keeping in mind our hook model, it would first prompt Karen to come into the app.
Now that we have Karen in the app, how can we keep her flowing? With our new navigation embedded in a bottom navigation bar, we can keep Karen clicking through:
- An activity feed with prompts to complete missing information
- Albums to easily navigate to a photo
- The photos page to access photos directly
- Creating some products or a booking request to digitise another album
- Her profile to see her membership benefits
These user flows identify the screens to design and the actions to get from one to the next.
Design phase
Our design efforts for the sprint focussed on testing to quickly capture and iterate on feedback gathered from usability testing. We maintained Eppik’s colour palette and typography throughout, focussing instead on flow, content and layout.
This was an iterative process that started with basic sketches on paper and ended with a a high-fidelity clickable prototype.
Key learnings
We tested across all stages of fidelity with 15 to 20 users. Users were on the whole enthusiastic about the design but the following changes were made as a result of user feedback:
- Notification flow had too many steps > go straight to content page
- Language used in the input form was confusing > re-word
- Confusion around how to save the form > change to auto-save
- ‘Ask for help’ feature didn’t make sense > remove and re-consider in next sprint
- Album page cluttered and no option to edit or share > trim down and add edit and sharing options
- Relationship in tagging flow not clear > remove and re-consider in next sprint
- Add voice recording on photo page > added voice recording
Improved design
The design focussed on an intuitive information architecture and a frictionless design-led user flow. Features included trigger notifications and in-app activity prompts, membership benefits, ability to add comments and audio recordings on photos, auto-save input fields and photo sharing options. Below the various flows demonstrate a number of features geared at engaging the user to add metadata to their photos.
Ideal user flow: push notification > photo page > input form > photo tagging > updated input form. The second last screen below shows the comments section of the photo page and the ability to add an audio recording. The last screen shows the ability to share the photo via popular methods.
Activity flow 1: facial recognition prompt asking user to confirm photos.
Activity flow 2: missing information prompt asking user to add information to a set of photos. This flow is based on insights from research suggesting users can only handle data input in bite-sized chunks.
Albums and photo flow.
Create flow includes suggested products and ability to create a booking request to digitise photos. Profile page uses gamification through member benefits.
Next steps
As the first design sprint, we managed to get the basics right and create the groundwork for an engaging app. In the next design sprint, we would:
- Test alternative input forms
- Build in a family relationship and family tree
- Re-think ‘ask for help’ feature
- Zoom functionality on album and photo page for accessibility
- Integrate shop into album/photo view
- Phone widget as a trigger
Challenges and key learnings
This was a fantastic opportunity working with my talented peers on a product that really resonated with all of us, especially at this difficult time. My greatest learnings came from my teammates, one being an exceptional critical thinker and the other a graphic designer by trade, and understanding how each of them approached the activities throughout the design sprint. As a highly collaborative group, we held several productive brainstorming sessions during which we rapidly bounced ideas of each other to land on the best solution. It was a truly enjoyable experience.
