Designing a new Logged-in Homepage

Findmypast is an ancestry tracing website and app, where users can trace back their family roots and learn about the people and places their family lived.

User’s typically engage in the following activities on the site and app:

Search of over 10 billion records, spanning over 500 years, for their ancestors. Searching is considered a more intermediate and advanced user activity on the site.
Build your family tree (typically a more casual/novice user activity), the site then generates ‘Hints’ of possible ancestor record matches in its database, giving the user further leads to trace other ancestors.

User problems

New Findmypast users didn’t know how to use the site effectively.

User’s either exclusively searched for records or started a family tree. The segmented approach led to frustration.

User research found we had to make it easier for new users to find and use the features on site.

Business problem

Help new users get more value, by personalising their experience and showing them different aspects of the product in one page.

The metrics tracked to define ‘increased engagement’ were the number of searches performed, building family trees, and how large the tree’s were, increased return visits and page views.

Tools used

Concept mapping, sketching, ideation workshops, Sketch, Invision, User testing, Journey Mapping

Context

Findmypast user research found new users struggled with discovering the site’s useful features at the beginning of their journey. To address this I designed a more personalised homepage experience, to be more engaging to new users, highlighting the site’s various products. Enabling users to achieve more on the site in their first few visits. By creating a better family history experience and, for the business, increasing subscriptions.

Results

Our homepage experience was released to a small group of users, gradually rolling out to 100% of users on all our domains (UK, US, AU, IE). We iterated through 6 more sprints taking on board user feedback through online surveys and analytics data that helped guide the design.The homepage increased user sessions amount and duration, annual order value and total page views.

Research and defining the MVP to test

User research was done in 2017 by a Designer and Researcher who were no longer with the business. The research, focused on beginner users, consisted of interviews and testing of paper prototype concepts of a personalised homepage.I analysed the research and categorised it into key themes that aligned user problems with the various products on the site that could address these needs. Another aspect I considered was how these align with the marketing teams CRM.

Design Principles

Using the previous paper prototypes, I defined the design principles I’d use to take this through low fidelity design in Sketch to high fidelity prototypes and onto development. I wanted to create a Homepage experience that wasClear and Intuitive

‍Family History sites tend to be bland, heavy on text, and black and white scans of documents.

In addressing this and to encourage users to explore the different products on site, I used a more modern card style design. This added more colour, illustration and iconography to the site. It also sat well with the concept of a homepage to launch their family research from.

Visual Design

To refine the card design I used mix of our Brand colours to associate to different products on the site. Blue for Search and green for Family Tree, using as many existing components as I could to ensure our developers were able to produce these easily.

To further enhance the look and feel of the cards, I worked with a Graphic Designer on a brief piece of work to create iconography for the most popular records types on the site. We also associated each record type with a colour. I used this in my card designs to introduce an accessible design language across the site to begin to differentiate the record databases through colour and iconography.

Content Design

From the research, we heard how users found the old homepage (and other areas of the site) overwhelming, with too much information on the pages for them to actually take in. Using a card based UI meant we naturally restricted the amount of copy on the cards. I needed to convey what interacting with a card would do and the user would be initiating and reduce the amount of overwhelm the user experiences.

Iterations and user learnings

How was the legacy homepage used?

GA analytics showed some surprising results on what users actually use on the existing homepage.
The page was not used by many existing users (they bookmarked search pages and started there)Users that used the page didn’t mainly used the search functionThe blog content was checked regularly by advanced users for new recordset releasesNavigation to different products was done via the main navigation barThis lead to our final testing with users before going live. Testing was positive with users finding it easier to either start a search, or explore different aspects of the family tree product.

Live versions

Development of the first release gave users access to the basics search bar and aspects of their family tree. Our MVP was released to 25% of users in the UK. We used Google Analytics and an online qualitative survey tool to observe and gather user feedback, we also asked users to vote on features from the research and in the product roadmap they most wanted to see. This helped us prioritise our backlog with their input.

Feedback that helped shape design from V1 to V2:
Search bar was not seen by users on v1, so I restored its location to the top left for v2.Quicklinks to certain record types was included, removing the need for users to use global navigation menusLink to weekly recordset updates (the FMP Friday card)Design of hints to take users directly to a person’s list of hints, based on feedback that hints was too complicated to navigate.The development team was unable to get specific hints for a single person on the user’s tree to work without significant time and resource cost. So a compromise was made by instead having a card that allowed users to jump to the profile of the family member in their tree they were last researching. This tied in with our research findings of reminding users who they last worked on.

This was to show users a select few relevant hints (reduce cognitive load), arrange the information to be read and understood quickly (reduce cognitive load), and then take them directly from the logged in homepage to the hint review page (reduce user effort required).

The following videos show the old homepage that was not personalised at all, to the new experience that would offer users the chance to interact with their family tree and records in one tap.

Findmypast had an iOS and Android app in development through an external agency. This was an extremely MVP release and just prior to finalising that release, I was asked to use the learnings from this Logged In Homepage work to design a landing page for the app. This was a very fast turn around to provide the developers a wireframe of this page.

Outcome

Overall the implementation of our Logged-In Homepage was well received by our casual users. The ‘hard core’ advanced family researchers and professionals did not like the product and preferred to still use the global navigation features to dive into the FMP search product on niche record sets.

Our intermediate and new users responded positively to the product. Overall we had more page views and higher engagement. The product was well received internally with a lot of product teams looking for space on it to test their experiments, including the marketing team.

Next steps

A new round of research with users at various stages of ability to understand further needs that can be catered toImprove marketing integration with their CRMMissions pack to focus users on tasks - focus on bridging the gap between new users using suggested hints and records to build their family tree; and expert users who search only through searchImplementing a DNA section on the homepage allowing users to view reports and run targeted search of records based on their genetic background.