---
title: Design Spiking
teaser: Get back to basics and speed up your development process through the amazing
  power of dead trees.
tags: design,prototyping
author: Joshua Ogle
published_on: 2014-11-04
---

A "spike" is a development term
for a quick exercise
that lets you explore solutions
without the burden
of writing *good code*,
then throwing it all away
so you can do it the right way
with confidence.

These are typically experiments timeboxed
to a certain number of hours
set aside just for exploration.
You might spike
when you need to get a better idea
of how long a feature will take to finish,
to try out a new piece of technology
to see if it would be a good fit,
or to find out if a feature is even possible.

At the end of the experiment
you should have an answer to your question
and evidence of the best path forward.

## That sounds familiar

If this weren't code
we might just call it sketching.
But this would be sketching
as a regular part
of the development process,
with rules and best practices.
That's something interface designers
could learn from.

Let's see how we might
be able to apply this
to ideation
and product design.

## Drawing is thinking

Spiking lets us explore
good ideas and bad ones
without putting any other value on them
except the little time it took
to put them on paper.
You're going to end up throwing out
a lot more than you keep,
but that's a good thing.
It's important to explore bad ideas
so you know specifically
what *doesn't* work.

> Art schools have abandoned drawing in order to make time for all the software
> they have to teach. We get what we need for our professional life but don't
> have an instrument for understanding the reality of life.
> - Milton Glaser

Apps like Sketch and Photoshop
make it tempting to jump straight in
and start making decisions too early.
You start forcing yourself
down a single path.

[Paper by FiftyThree](http://www.fiftythree.com/paper)
and other apps like it
are a good way to keep things digital
while keeping the low-fidelity feeling of sketching.

## Exercises

We can distill some techniques from the
[Product Design Sprint](https://thoughtbot.com/blog/the-product-design-sprint)
to efficiently spike a solution.
You already have
a good understanding
of the problem to solve
so we are going to focus
on how to get those ideas onto paper.

---

### Mapping the Critical Path

Making a sketch of the critical path
can help get a good picture
of the [job to be done](https://medium.com/the-job-to-be-done).

![Critical Path](https://images.thoughtbot.com/design-spiking/sketch-path.png "Critical Path")

---

### [Divergent Sketching](http://playbook.thoughtbot.com/#diverge)

* Keep things simple.
  Use a Sharpie or fat marker
  to keep yourself
  from getting too detailed too soon.
  Using plain printer paper
  instead of a sketchbook
  can also help things feel ephemeral
  and you won't feel the need
  for the sketches to be perfect.
* Speed!
  You want differentiation of ideas
  and as quickly as possible.
  Draw as many ways
  to solve the problem
  as you can think of.
  At this point,
  don't worry about
  fitting a styleguide
  or if it's easy to implement.
* After you are done,
  make notes on each sketch
  detailing which parts
  are viable solutions
  and which aren't.

![Divergent Sketching](https://images.thoughtbot.com/design-spiking/sketch-diverge.png)

---

### [Convergent Sketching](http://playbook.thoughtbot.com/#converge)

* Collect the rejected sketches
  for reference.
  If someone suggests
  a similar solution
  you can show them
  what it would have looked like
  and why you rejected it.
* Combine compatible ideas
  to take advantage
  of their strengths.
* You can spend more time now
  to add appropriate detail.
  You might even keep these notes around,
  so add the information
  you might need later.
* If you have competing solutions,
  ask someone for their opinion.
* If you don't have someone to ask,
  imagine presenting each of them to someone
  with each of their pros and cons.
  You may end up ["rubber ducking"](http://en.wikipedia.org/wiki/Rubber_duck_debugging)
  the solution.

  ![Convergent Sketching](https://images.thoughtbot.com/design-spiking/sketch-converge.png)

---

## Doing these exercises in a group

You can also do both
of the exercises above with a group.
Try to keep it to two or three people.
If you feel the need to add more,
it may indicate the need to
do a larger [design sprint](https://thoughtbot.com/blog/the-product-design-sprint)
format instead.

During the *Divergent Sketching* exercise,
you will need to vote
to make the selection of viable sketches.
You can use stickers, checkmarks, etc.
to mark your favorite sketches
or even specify parts
that you want to see explored more.
Take some time to discuss
after you vote so you have some agreement
before converging.

The *Convergent Sketching* group exercise
is where things get exciting.
If you haven't reached a consensus
on the last step,
you may need to do some more voting
to narrow down solutions
and decide which pieces should be combined
for the final design.
This should be a safe space
to explore even the craziest of ideas,
so don't be surprised
if opinions are polarized.

## Get out there and kill some trees

You should come out the other side
with a clear and logical process
and hopefully a couple of crazy ideas left in.

Design spiking absolutely
does not depend on artistic ability.
So encourage your clients, developers,
project managers, information architects,
and that designer down the hall who
"doesn't have time to sketch".
They might be surprised
what design thinking can do.

Have some ideas on how to make design spikes
more efficient, or have some sketching tips to share?
Give us a shout [@thoughtbot](http://twitter.com/thoughtbot).