Increased Virality

Increased Virality

Client: Dreams
Role: Sprint Facilitator, UX Designer
Year: 2020
Project length: 5 weeks
Design Team: 4
Participants: 6 employees from Dreams (Product Owner, Designer, Marketing, Head of Design, Behavioural Scientist)


Dreams is on a mission to help people save money to achieve their dreams in a fun and easy way. Their main goal at the time was that to increase their user base. Our team came in to help them explore ways to expand their user base and increase their virality.


A big part of the process centered around the Google Ventures Design Sprint-format. We chose this method because we had limited time and wanted to really dive into the challenge with a structured and efficient approach. This method allows you to build and test solutions to problems fast - within a week you have the first prototype that you also test. It's also a way to put heads together and co-create. We suggested the method for Dreams in a preperatory meeting, they got curious about trying something new. From there on the planning started within our design team.

Design Sprint: Execution

In the first week of the project we dived right into the sprint framework, after having a meeting prior to the project start. Below I will go more into detail what the first week consisted of.

Day 1 - Expert Interviews, Challenge, Solutions

We started the first day with having Expert Interviews with people at Dreams that were well acquainted with the company and their challenges, to get a deeper understanding and generate more general knowledge about the context we now dived into. The next step of the day was to generate and pick "Can we"-questions with the group, the questions are very important in the process, they act as a red thread through it all. The group was all together in the same room and we did a "together alone"-approach, every participant first got a few minutes to think in silence, and then write down their thought on a post-it. When everyone had written down one or more questions we put them up on the wall or everyone to see. The last momentum of this was for the group to vote on which questions was the most important in the defined challenged. We ended this session with having a long-term goal+sprint questions. In the afternoon it was time to be a bit more creative and produce solutions. This was done by doing some sketching excercises. We reminded the group of the recently decided questions and long-term goal and based on those the group was going to sketch on possible solutions/ideas to meet the challenge. We did four exercises: Note taking, doodling, Crazy 8s and Concept creation. In the end every participant had created their own concept. We ended up with a lot of ideas that was going to be "up for review" in the beginning of day 2. The biggest challenge in the first day was to get the participants comfortable with the rapid pace and the uncertainty in the sprint process - especially since this was a new way of working for them.

Day 2 - Voting, deciding concept

When day two started we had prepped by hanging up all the concepts on the wall anonymized, so no one could tell which concept belonged to whom. We went through the concepts together in the group, and they all received a handful of small red dots to put on the ideas they liked the most. and therefore created visual cues that made it easy to comprehend where the "heat" was. We took a short break and then moved on to scanning the "heat map" where every participant had one vote each. They chose which one they believed the most in. The team at Dreams chose to go forward with two concepts: one for the norwegian market and one for the swedish market, due to different regulations in the countries. In the afternoon the goal was to map out the chosen concepts and create a more detailed storyboard and a step-by-step user flow. The storyboard was supposed to be as detailed as possible to make it easier when creating the prototype.

Day 3 - Prototyping

Prototyping day! Our design team worked in Figma to create testable prototypes that resembled the storyboard sketches.

What was challenging when creating the prototype was to take lot's of snap decisions about the user interface, escpecially when creating two. What helped was to have support from Dreams design library, so we didn't have to create visual elements like buttons etc. from scratch. Meanwhile the prototype was being created, we simuntaniously recruited people to test the prototype the next day. Even though the recruting was in a short notice we managed to find five users that could participate. We ended the day making the prototype interactive and testable.

Day 4 - Testing, testing

Time to test the prototypes. Before the tests we prepared a semi-structured interview guide to support us in the tests, so we knew what to look for. We did all the tests remote via video, and the user's shared their screen while interacting with the prototype. One challenge we met this day was a few drop outs along the way which lead to us having to find new people fast, here it helped that there was four of us in the team so we could split up to recruit new users. At the end of the day we had valuable insights from the users, which I will go more into detail below.


As mentioned above, we chose to test two different protoypes and compare them. One with emphasis on gamification and the other with focus on monetary incentive. We did three different test rounds, where we in each iteration reached new insights and got closer to the final prototype to be delivered.

Test round 1

- Of the two concepts the monetary reward (protoype 1) performed the best. It was the easiest one to understand and the incentive was clear. It was both triggering and trustworthy

- The gamification of unlocking a secret gift is triggering and make users curious, but not curious enough to go through the whole flow and invite a friend "it might not be worth it"

- To “unlock” might be a good idea to apply to other features. For example: "Unlock savings class for 59 kr"

- A common moment to invite a potential user is when a friend says they want to purchase something or go on a trip. This is a moment users recommend Dreams to their friends.

Test round 2

- The majority of users want to in a simple way copy the link so they can share it in group chats

- Users want to personalize text to friends, instead of using a pre-written message

- Visually seeing money coming in is motivating to invite more - this is the moment users sends out invitiations

- No matter how triggering the incentive is, some people will never share, it’s not an existing behavior of theirs

- Emails might end up in spam or spam filter. Feels less trustworthy. In-app message or notification feels more trustworthy because it comes straight from the app

Test round 3

- “Savehack” as a use of term - users does not know what this means when seeing it in the notification message

- A negativity towards push notification overall

- 50SEK to me - 50SEK to my friend is a motivating reward system

- Confirmation steps need to be more clear with feedback. Left users confused “Do I need to do something more here?”

- Users wants to see where (in what dream) their money lands

- The pre-written message when sharing the Dreams link feels stiff


After three iterations during three weeks containing usability testing and design adjustments according to the insights, we had a polished and implementable prototype handed over to Dreams.

- An implemented Referall Program in the Norwegian market

- Our team was invited back to Dreams Office to present the method we used for the whole company

- Changing the way Dreams work internally at their office, using the introduced excercises from the sprint, "mini-sprinting", etc.


- Draw boundaries as a facilitator. For example when the group is in discussions that's not leading anywhere.

- Start working earlier on with recruting testers: user's might drop out last second.

- Research potential pitfalls early in the sprint, or even before, to avoid sudden hinders like the banking system incident in this case.

- In a sprint format, try to focus on testing only one prototype. It could get a bit overwhelming for both us as designers but also for the test participants to test two different concepts at once.

- Wrap up is a great way to finish a collaboration/project. Gives a lot of value+learnings and close the loop.