Micro design sprint

My blueprint for the Evermix ‘Micro Design Sprint’. 2 days. 2 people. 1 problem. What could go wrong?
March 8, 2017

TL;DR At Evermix we need to update the onboarding process for new DJ users. No, let me rephrase that - we just need an onboarding process. Full stop. We have made every effort to make it obvious for new users to set up a new account and upload their first mix, and whilst this is largely successful, it is still proving to be a sticking point in the user's experience. We have a limited budget, time and resources (2 of us are the design team in Manchester), but we still need to sink some serious design thinking into this component. Enter my blueprint for the Evermix ‘Micro Design Sprint’. 2 days. 2 people. 1 problem. What could go wrong?

My Micro Design Sprint is based on google ventures design sprint methodology (obvs). Links to resources I used:

Design Sprint Methods - google developers PDF

google ventures Medium series on sprints

Buy Jake Knapp's book on design sprints (I did, it's worth it).

Read on for more details. 


Evermix has a problem with DJ users get stuck at certain points of the profile creation and upload journey. Every effort in the design process so far has been made to make it easy and obvious for DJs to create a meaningful profile that is easily discovered by listeners and, crucially, to easily upload their mixes to the platform. But feedback so far has been contrary to this. 

From anecdotal evidence (such as Prok & Fitch noting that it’s not obvious where to add mix information on the app) and from simply observing uncompleted, uninteresting profiles from a spectrum of DJs across the site we are seeing evidence that DJs are simply not ‘getting’ the process, are getting stuck or missing something completely. 

To add to this, in the past, we have had to manually add and create DJ profiles for them. But with the new round of MixBox2’s being released for general purchase, we are now facing a challenge where we have to make it accessible for everyone to create a profile and upload their first mix. 


We are now looking at deploying an effective onboarding process that solves these issues. We have to make it easy and straightforward for a DJ user to come to our site, sign up and upload their first mix and make it as frictionless as possible. So, in the future, the DJ can come back and know how to upload their second and third mix without feeling overwhelmed. 

In addition, we want listeners to come to the site and find their next favourite mix to listen to quickly. We do this by having an obvious design language that allows listeners to navigate around the site, and this comes with having complete meta information for each mix and DJ profile.



A design sprint is a methodology of working through a design challenge: brainstorming ideas, and quickly prototyping and testing ideas based on research. It was initially created by the design agency IDEO and made popular by google. Today, design sprints are carried out to design products at The Guardian, BBC and Facebook amongst many others. 

At Evermix, we already have a product, and large parts of it are working as expected. So a design sprint for an onboarding process might seem like a hammer to crack a nut. As we already have design components established, we would usually skip to the prototyping stage after making some assumptions based on learnings from the initial version 2 design of the site and app released last year.

However, on-boarding presents itself with its own unique challenges and it might be necessary to map out user decision points to base a new design on. Also, there might be a need to explore a few conflicting ideas in parallel instead of committing to a safe choice.


So, to address these, I am designing the concept of a ‘micro-sprint’ for the

Typically, a ‘design sprint’ lasts from 3-5 days, but because of the nature and the size of the component we are designing and the size of the team (2), I think we can condense this into a couple of days. 

Truthfully, the conclusions may very well be the same as if we went straight to prototyping. But it’s really important to dive into the process and take the time to find as many conceivable ways we can to design the onboarding experience. And, you never know, the design sprint may uncover a new, unique way to onboard users, making them serial uploaders and

A regular design sprint is split into 6 stages: Understand, Define, Diverge, Decide, Prototype and Validate. For the micro design sprint, I have adapted and shortened them slightly to fit our new format.


1. ‘What are a new DJ users needs at this juncture?’ It’s important to know where a new user has come from and where they are going to.

2. Review what we have at the moment. What is the current state of the platform for a new user? What do we already have? 

3. Competitor reviews. Review the uploading and onboarding process for Soundcloud, MixCloud and others.

4. Discuss the goals of this process. What does ‘success’ look like? What are the business goals/what are

5. Discuss technical capacities and challenges. There might be some technical constraints that hold us back, what are these?


1. What is the focus of this process? In a sentence, define what the result of this process is going to be. 

2. Draw a user story diagram - where have they come from, where are they going. Draw out this journey and hang it up on the wall to direct the rest of the process. 


1. At this point, we need to explore as many ideas as possible.

2. Lo-fi mind maps (10 minutes) - This exercise is basically your own personal notes on the user story diagram from the Define phase. It doesn’t matter what the mind map/brainstorm looks like: is it a list? Are the points joined up? You’ve drawn something? It doesn’t matter, as long as it helps you frame the problem in your head to refine it later on. 

3. Crazy Eights (5 minutes) - From the ideas generated on the mind map. Fold a piece of paper three times width-ways and once lengthways to create 8 comic book style panels. Set the timer and each person has to sketch one idea in each panel - 40 seconds for each panel/idea. The point here is quantity, not quality. 

4. Storyboard (10 minutes) - pick your three best ideas from the Crazy Eights exercise. Grab a piece of A4 paper with three post-its stuck to it and storyboard the UI for each phase of your idea - ‘first this, then this, then that’. It has to make sense to anyone reading it and it is good to have a name. 

Examples of how some storyboards from a previous design sprint.


1. Search for conflicts (2 minutes) - Are any of these ideas mergeable? 

2. Silent critique (5 minutes) - stick the ideas on the wall and everyone looks at them individually and silently, sticking dots on the ideas (or parts of ideas) that they think works. At this stage we will open it up to the London office, to get their votes on the ideas too.  There are no limits to how many votes you get and, yes, you can vote for your own ideas. 

3. Vocal critique (3 minutes per idea - 15 minutes max) - Everyone gathers around the storyboards and discusses what they like and don’t like about each idea. It will also give the creator a chance to describe the idea in more detail.

4. Rinse and repeat - if we think it’s necessary (so, if we don’t have any solid ideas that we agree on or the ideas don’t touch back to the user story) and we have enough time, revisit the cycle again to gather more ideas. 


1. Create, lo-fi interactive prototypes of the best ideas. - this is the part where we will break away from the mainstream design sprint method, as I will create the prototypes in isolation. 

2. Review prototypes - As a result of creating the prototypes solo, James and I will review the prototypes to make sure everything is represented.

Some Sketch - Craft - InVision prototyping goodness from a previous design sprint.


1. Test ideas with stakeholders - initial phase of testing will be a video call with 2/3 stakeholders within

2. Test with DJs - if possible, also through a video call. 


I am looking at completing the whole design sprint across two working days. Specifically, UnderstandDefineDiverge and Decide from 10 am - 5 pm on Friday, and Prototype and Validate from 10 am - 5 pm the following Monday. There will be 3 breaks: a 1-hour lunch, and two 15 minute breaks in the morning and afternoon. That leaves us with 4.5 hours of sprinting time. 

A design sprint team usually consists of 5 - 7 people from various areas of the business. But as there are only two of us in the Manchester office, we will have to make do with us and input from the London office when needed (Decide phase). So, even though I will be leading and taking notes as we go along I will also be contributing.


By the end of this process, we should have a user story diagram, a reasonably polished mockup and a lo-fi prototype based on user touch points on the user story diagram.

Mmmm mockups.


We can now draw some conclusions from the deliverables, and I see the next phase as one of two routes:

1. Start the process again. There’s a very real possibility that this process and the subsequent testing will expose a larger challenge to overcome. And that’s not a bad thing! But, it will mean that we can enter another round of design sprints and refine our ideas further, making sure we have the right journey nailed for users.

2. Everyone’s happy. In an ideal world, everything that we have created has tested well. We then move on to build it and get data and anecdotal evidence from real users in the live environment. 

So that’s the outline for my Micro Design Sprint. I presume this post is going to be part 1 and hopefully, there’s a part 2 to come back to see where we’ve gone wrong and what learnings we can take away. Should be fun!

My Micro Design Sprint is based on google ventures design sprint methodology (obvs). Links to resources I used:

Design Sprint Methods - google developers PDF

google ventures Medium series on sprints

Buy Jake Knapp's book on design sprints (I did, it's worth it).