Divvy Retrospective


I, along with two of my classmates, created an app for Divvy bikes, a bike sharing program in Chicago which would encourage new users to join, enhance their experience and reward existing bike riders.

The Final Product


The Team

Caroline: sitemap/information architecture, data collation, market research
Cory: visuals, app wireframes & mockups
Anna: touchpoint & profile mockups, prototype build
All: user research, strategy, ideation, user flows, synthesis, testing, delivery

About Divvy

Divvy Bikes is a bike sharing company based in Chicago. Divvy wants to create a way to promote outdoor cycling and raise brand awareness through a cycling lifestyle oriented application. http://www.divvybikes.com/

The Brief

Divvy is well known to the bike sharing community, but they have not gained as much mainstream reach as they would like. Divvy believes that there may be opportunity to increase awareness of their brand through providing value added content to new bike sharers. They want to do this by providing an application with real utility for this market.


On approaching this problem, it was important for us to gain a better understanding of Chicago’s biking community and global urban biking trends in general to build a solid foundation from which we could launch a comprehensive solution of bikesharing. We immersed ourselves in a wide range of information and data sources that helped us gain knowledge as we conducted our own research related to Divvy use.

The key takeaways from the secondary research we conducted related to urban biking trends and attitudes. Specifically, providing us with four important bicyclist categories:

  1. Strong and Fearless (SF) — this is someone who will ride their bike regardless of the roadway conditions. This category is unlikely to be a Divvy user. 1% of the population.
  2. Enthusiastic and Confident (EC)- This person feels safe riding their bike on most of Chicago’s streets. Many existing bicycle commuters fit within this segment, and we can correlate this group to active Divvy subscribers. 6% of the population.
  3. No Way No How (NN) — this group is not getting on the bike. Period. 33% of the population.
  4. Interested but Concerned (IC) — The majority of the population is interested in riding their bike, whether for work, fun, or errands, but are concerned about the safety of riding in traffic. 60% of the population.

From these four categories, we found an opportunity to better understand the IC group. Also, the EC group became an important contributor to understanding the pain points of the Divvy system and the biking community and culture in Chicago. They are a perfect complement for each other, and us identifying them early on allowed us to create a solution for both groups.

We also felt that for the purpose of this project, it made sense to focus our efforts exclusively on the EC and IC populations and to not seek out ways to provide solutions for the SF and the NN groups.

Primary Research

Our research included surveys, interviews, contextual inquiries, and user testing.

Contextual Inquiry

Anna & Caroline renting Divvys

Our contextual inquiry included talking to Divvy users in situ, observing their behavior, observing the Divvy stations and the Divvy bikes, and also going on a Divvy ride. Anna and I fall into the IC category and were helped out by an EC. This was a great opportunity for us to get first-hand experience related to the learning curve, fear, and the benefits of having an expert provide some guidance to us.

We also talked to tourists using Divvy bikes and people we found waiting for buses to find out why they weren’t interested in using Divvy bikes.


The survey data that we acquired was critical for us to identify Divvy’s competitors.

From this competitive analysis, we were able to determine both a cost and time benefit for Divvy:

This then led us to an important insight: Despite both a cost and time saving benefit, people will take the path of least resistance. They will do this in order to avoid the stress or confusion of having to figure out a new system. We were able to provide a solution for this barrier in our app through a facilitated communication channel between new Divvy users and existing Divvy users, where the existing users are helping the new users overcome the learning curve associated with a learning how to Divvy.

User Interviews

With the help of some donuts (genius, Cory!) and some aggressive recruiting tactics, we were able to speak in depth to nine people who shared their experiences and feelings about biking in the city and using Divvy. This qualitative information was rich and informing and led us to the insights that formed our design direction:

  • Non bikers are more likely to try Divvy in groups or when they have a companion to help them figure it out.
  • Non bikers avoid using Divvy because they don’t have the confidence and knowledge to navigate Chicago’s busy streets.
  • Experienced bikers are invested in the biking community and are happy to share biking etiquette knowledge.

Design Direction

The solution should enhance and streamline the experience of using divvy for existing subscribers and encourage reluctant but interested users to start using divvy through support and access to a service which makes them feel more confident about biking around the city using divvy.

Creating our personas

From our research we found four groups which formed the basis of our personas. The first persona we created was Bjorn, the divvy rider whose pain points mainly revolve around the infrastructure that currently exists are Divvy.

Then we created our potential Divvy users:

  • Jane, our career woman who likes the idea of biking for fitness but is scared of riding in the city.
  • Kate, our flustered and rushed commuter who would like to save money and time if only she could figure out what time she’ll arrive by Divvy.
  • Ruth, our tourist who likes riding around the city but would want to see some of the sights while at it and not just get from A to B but also get from A to D, via B&C.


Using our new personas, we created a scenario for each to start ideating around how we might solve for each of their needs.

The ideation session provided plenty of great ideas, some realistic such as better routing. Some more dream like: putting Jane in a bubble and virtual reality head sets.

We took various themes and placed them into categories of features which revealed some clear areas of focus and an outline of an affinity map/ sitemap.

From there we started sketching out some user journeys to figure out where some of these ideas might help our personas. We realised Bjorn’s revolved around him using his rewards. Kate’s revolved around getting from A to B quickly and Ruth needed to be able to see the sites.

But Jane’s was more complicated. We knew Jane needed to get home by bake and feel safe but there was an extra challenge with trying to convince her to get on the bike in the first place. We kept going back to Jane’s story throughout and realised that we needed to focus on solving for her.

We also found that for her story to work, she needed someone to help her get on the bike. We’d discussed a group option and mentoring option and had found in our research that Divvy users were keen to help new users.

We saw an opportunity to link Bjorn & Jane’s story and create a system which would entice new users to join and reward old users.

Narrowing down features: Creating the MVP

We narrowed down our ideas into the features that we’d be putting into the app and prioritized what would be in our MVP.

The Solution

From sketch to wireframe to mockup.

User Testing

We began user testing as soon as we started sketching. Asking users what they would expect to happen when they touched certain screens and asking what they understood by different buttons. We also tested features and concepts during user testing, and from user feedback we were able to concentrate our efforts on the Trailblazer and Wayfinding features. The user-generated-content features were put on hold for refinement and more testing.

Wireframes to mockups on Sketch

We knew that for current users to feel comforted and welcome using our solution, we needed to maintain the branding with which Divvy already has blanketed Chicago. Their modern, striking design already resonates a clean look that importantly evokes simplicity with a hint of structure, affirmation of its presence in Chicago as an official public transportation option.

Changing that look would unnecessarily destabilize current users from switching to the new application, and would make potential new users shy away. Our solution needed to help users break through walls, not build new ones.

First, we examined the existing Divvy visual experience in detail. We determined what typeface Divvy uses in headers and body text, color palettes, and placement proportions and styles for all of the above to build a style guide to prepare a set of visual design baseplate from which to develop the visual style of our solution.

Our first mockup iteration:

After user testing, we refined the design to pertain directly to user needs and design standards:

Our third and final iteration of the mockup focused on important, most desired features, as well deprioritized unnecessary and confusing features:

With our first prototype on a phone, user testing revealed overall that our design was met with satisfaction and delight. Users found swiping through the new mapping system easy to use and intuitive however, a number of users didn’t understand the hotspots and trip suggestion features. They also found our menu system confusing and weren’t sure where to click.

Once we explained them, they still didn’t think they’d use them. While we believe there could be an opportunity to include user generated content at a later phase, this idea and its purpose needed to be fleshed out further. We deprioritized it and focused on improving the key features of our design.

We also found the search element was confusing some users, despite the fact that it seemed to mirror a lot of other apps. A few design changes made the search seem more a part of the map, and less of a separate element, which solved the confusion.

For the final mock up we simplified the navigation further and made the search more intuitive by making sure the map could be viewed above the search bar as well as below.

Other touchpoints

Going back through our user journeys we also needed to create different touchpoints to entice and engage with our users before they even downloaded the new app so we created email and desktop elements to engage and inform our users of the benefits of the new app.

We also realised that ‘Jane’ in her journey would most likely want to create a profile online and select a preferred route before downloading the app.

How this solution met our users’ needs

Ultimately, we created a user journey that would not only solve for all our personas’ individual needs, but reward existing customers and entice them to recruit new customers.

We did this by:

Streamlining the experience for all Divvy users with features such as GPS location, code generation and notifications when time limit is approaching.

Enhanced wayfinding with sightseeing, direct and bikeways features to encourage interested but concerned users to be able to find their way home easily.

To enhance and extend that experience for everyone who uses a Divvy with a profile, progress reports and saved routes.

A support system to give users the confidence they need to start riding Divvy by

  • Encouraging new interested but concerned users to join Divvy
  • Rewarding experienced users for their help by making them “trailblazers”

The Trailblazer Program

We found in our research that Divvy bike riders were willing to help new users and also wanted to educate them to make them better divvy riders. We also found that new users were willing to try Divvy in groups so we saw an opportunity to bring these two users together.

The Trailblazer program was a way to get existing users to be mentors to new users by allowing new users to join them on their ride home. They would then be rewarded with points which could be used on time extensions and free passes.

The strategy of the reward program of the trailblazer feature must negotiate the distribution of free services to successful trailblazers with the potential earnings of new Divvy passes and subscriptions the feature ultimately means to generate.

In summary: although a trailblazer would not be paying the extra $1-$2 for going over their 30 minutes, they would have generated $10 in revenue from a new rider for that day and possibly gained Divvy a new subscriber.

To justify the feature, we investigated commuter census data to place it in context of actual numbers to generate precisely its potential earnings.

Let’s start with the 2014 commuter population:


Of that amount, we can further break it down.

26.7% commuted via public transportation

1.3% biked

6.4% walked

4.2% worked at home

Let’s round those numbers to better comprehend the data.

25% of 1,200,000 is 300,000.

60% of 300,000 is 180,000. This is the number of interested but concerned people.

If we only reached 1/20th of 180,000, that is 9000 people new to Divvy that would try it that otherwise wouldn’t have.

Currently, there are only ~4,700 Divvy bikes in circulation. 9000 makes sense.

If only half of those purchased a 24hr pass, that is a daily potential earning of:

4500 x $9.95 = $44,775

If 1/20th of those purchased subscriptions, that is $22,500; but that number is recursive and on top of having tried the service

$44,775 + $22,500 = $67,625.


Potential quarterly earnings from the trailblazer program:


And this is only commuter data. Data for local neighborhood grocery trips and weekend tourists still can be compiled and added to this estimate.

Next Steps

Some of the features we deprioritized, such as user generated content, we feel could still be a great option to enhance the app further.

We also think there’s a huge opportunity to bring the app to a smart watch to make it even more accessible on the go.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.