Improving feature announcement strategy

Product design > Turo case studies > Vision : Recent updates

 

Synopsis

I led a vision project to increase awareness and adoption of new features and policies by improving the way we communicate updates to hosts. The project was green lit and eventually built and shipped. We were able to measure a 10% improvement in host sentiment toward how Turo communicates changes to hosts after the launch.

Details

Team: Host

Platforms: IOS/ Android/Web

Role: Principal Product Designer

Timeline: 1 year to get prioritized, 1 month to ship.

Skils: Strategic thinking, Ability to influence product roadmaps, UX/UI, user research

Company Problem

By 2023, our existing feature announcement strategy was handicapping teams abilities to meet feature adoption goals.

Examples of recent releases that had sub par adoption

  • Pricing insights: after launching insights to hosts designed to get them to more competitive prices, we realized that only 25% of hosts viewed the insights at launch.

  • Another team launched a messaging feature to improve host efficiency and reduce time it takes to send guests important instructions. The adoption of that feature was only 43%.

  • Turo integrated with several toll agency companies, allowing hosts to quickly track and manage tolls that are incurred on trips and request reimbursements from guests. The toll enrollment only accounted for 33% of hosts.

  • In another project I worked on, we spent over 6 months building the most robust fleet calendar tool to aid hosts in gaining visibility of their prices and trips across all their cars. Only about 52% of hosts viewed the calendar after launch.

Host sentiment survey results

Only 46% of 10+ car hosts felt Turo was adequately communicating changes

Hypothesis

We believed that the core reasons for low adoption had to do with hosts not being aware of the new updates.

Unpacking the problem

Our design solutions to notify hosts have become saturated over time

  • Multi-car hosts receive dozens of emails, push notifications, and activity feed messages a day. plus marketing emails. These range from trip-related updates and marketing. Emailing hosts about new product features just adds more flame to the fire. Emails have poor open rates and are most likely to be missed, making them a poor tool to rely on for communicating important updates

  • Red dots became pervasive.

Unpacking the problem

Our solutions have also become more more intrusive over time

  • We started with placing contextual banners at the car level, and over time we had 7 or 8 versions that sit in the same place, making it a challenge to prioritize and gain visibility for individual initiatives. The lower positioning in the host information architecture also makes discovery challenging

  • Next we moved on to a banner carousel at a higher level in the host IA. This was slightly more scalable, in the sense that we could display more at one time, but we know that priority order still matters and scrolling past the 3rd spot was low likelihood.

  • We started implementing more invasive tactics like tool tips, bottom sheets, and finally a marketing tool called airship.

Unpacking the problem

Every team resorted to using a non-scalable solution of displaying bottom sheets that pop up on app load.

  • This created non-scalable solutions when multiple teams wanted to launch feature updates in the same time frame.

  • Hosts would see multiple pop ups one after another, creating a jarring host experience.

Unpacking the problem

Building ephemeral bottom sheets costs us a lot of brand and engineering resources

  • Brand support

    • With every announcement Brand designs an illustration to accompany.

    • The illustration is only seen once

  • Engineering

    • Each team spends a lot of engineering resources to take down the previous bottom sheet and re-build their own version.

    • We now have so many that we have to build extra logic on priority order.

Goals and metrics

Primary

  • Increase adoption across future host features

  • Increase host sentiment toward how Turo communicates updates

How might we

Design a home for feature announcements and policy updates that is both discoverable, and scalable, to better support feature adoption

Concept one

Build a centralized page to display recent updates

  • What’s new page pops up on load if there are unread cards

  • Houses all the most recent feature and policy announcements in one place

  • Re-accessible from secondary entrypoint

  • Backend-controlled for simplified releases and cohesive experiences across platforms.

Pros

  • lower scope to update

  • Web-based, backend driven

  • Scalable solution - No fighting for priority

  • Re-accessible from app

  • Announcements persist, granting more visibility over time

Cons

  • Not a permanent solution 

  • Still intrusive and disruptive

  • Requires effort to ensure navigating from push doesn’t get overridden

  • Must educate hosts on where to access “Recent updates” screen again

Concept two

Build a home tab to display all action items, updates, and important statuses to hosts at a glance.

  • More holistic approach to support multiple team initiatives

  • The MVP could display recent updates

Pros

  • Highest level of visibility

  • Scalable - No fighting for priority

  • Persistent access

  • Announcements can persist

  • Non-intrusive

    • If all updates are seen, we can still default to trips

  • No wasted engineering work

  • No required education to find re-entry point

  • No extra logic to ensure we don’t disrupt notification routing

Cons

  • Higher scope

    • Existing More tab moves into hamburger menu

  • Longer time to market



I presented these concepts to leadership as a means to get them prioritized. Leadership agreed on the solutions, but no team wanted to take on the work at that time.

So the project was shelved until I found more support..

Iterations

I continued to iterate on the recent updates screen…

I did several rounds of design explorations, seeking feedback from the team.

A few iterations

Variables:

  • Icon type and size

  • How best to represent read vs unread

  • Title only vs Title with body copy

  • Link or button

  • Illustration or icon

  • Labels or no labels

Designing the flow and how to re-access recent updates

  1. Titles only, with labels

  2. Titles for read, body copy for unread

  3. All have larger images, titles, copy, and ability to add a banner and CTA

  4. Only unread cards have larger treatement

Usability testing

The team still had these questions:

  • How do we reduce cognitive load in a list?

  • How do we make sure each announcement is seen?

  • Is more the right entry point to re-access this tab?

  • Do we need other entry points?

Goals of the test:

Validate the “recent updates” screen as an improvement on discoverability for new features and updates. 

Does the what’s new screen have a higher chance of improving discoverability?

Can hosts locate the entry point to recent updates to re access the screen?

All participants noticed and interacted with the recent updates seheets.

5/5 participants scrolled up and down on the list before closing

Part three

Shipping

Once we got the host to the inspection settings page, we needed to balance the page layout to prioritize uploading the form with educating the host how to carry out the inspection.

Evolution of the safety inspection page

It took several reviews both with my product manager counter part and with the design team to balance the page with the proper amount of messaging.

 

Part four

Educating new hosts

Most of the project focused on helping existing hosts discover the new feature, successfully download the form and re-upload it. We knew there were new hosts that would list cars that were older and would be required to submit an inspection within the first three months of listing their car. We decided to update the post-listing flow content strategy to include this new information.

Existing post-listing experience

After a host completes a listing, a series of carousel messages were designed long ago to educate hosts on the “most important” information they need to know. The carousel has been added to over time and the messaging has become less succinct and focused. We knew we wanted to re-vamp this whole section, but time was not permitted, so we needed to at least find a way to include the inspection requirements without making the problem worse.

 

Updated post-listing experience

The highlighted pink sections are the areas we changed.

  • We replaced the bulleted list on the safety and quality standards page with subheads that made for easier scanning

  • We added a maintenance page to speak to the new maintenance requirements

  • We simplified the content on the refuel screen to only talk about refueling

  • We simplified the clean and disinfect screen to talk about the cleanliness requirements as well as the covid 19 training

 

Part five

Optimizing the upload form

This was a unique design challenge because we would be giving a host a digital pdf and they somehow had to have that PDF filled out either by printing it or sending it to their mechanic and then re-uploading a completed form.

Part of this experience is uploading images or PDFs and displaying that back to them. Turo historically had not developed a scalable solution to handle one off upload experiences. This meant extra complexity and time to build this feature.

The final prototype

I audited the guest booking funnel to identify the key places where the all-star host badge could live. What I soon discovered was how we display host information is inconsistent throughout this experience.

Usability research findings

We tested the flow with 6 Turo hosts in an unmoderated usability study to validate that the flow was clear and intuitive.

  • All hosts were able to successfully discover the new requirement via the banners and navigate to the proper location

This was enough for us to move forward confidently.

Final Outcome

We launched the program in October 2021 and as of January 2022, We had our first group of hosts who ignored all of our warnings both in the app and outside of the app, and had their cars unlisted. We knew some of this would happen as some of our supply is less engaged and motivated.

Overall, we know this was the right call, and even if it does churn a bit of our lower quality supply, we think it has done its job.

What I would do differently

At the end of the project, I was rolling off this team and moving onto two other teams and needed to offload some of the last of the QA and other responsibilities to another designer. Had I stayed on this team I would have advocated for a second follow up story, to add a filter on the vehicles list that would allow multi-car hosts to easily find the cars that had inspections due without scrolling through their entire list of cars.

Unfortunately, the new designer was unable to push for additional work and the team moved on. I would have liked to see this additional feature added as it would have unlocked other projects in the future.

 

Supporting designers:

View other host quality initiatives:

Previous

Host mode - phase 2