Random thoughts on design ‘leadership’

Leadership is not, in my opinion, a synonym for management.

In some of the places I’ve worked the leaders and the managers are sometimes the same people, but sometimes not. (And, by the way, that’s ok. Management is an important function, in itself.)

You don’t need to be the greatest designer in the world to be the greatest design leader – these things are two different things. They’re not synonyms.

As a design leader, asking (sometimes provocative) questions is more useful in the long run than providing definitive answers.

And here’s some really nice slides from Cameron Rogers, the UX leader at Seek.

Running Lean

So, here I am on Saturday night reading through the Lean series. It’s a little life, but it’s mine.

I’ve already read Lean UX and Lean UX for Startups so I thought I’d move along and read Running Lean …

Notes from Running Lean

The Lean Canvas

“Your product is not NOT “the product” … your business model is the product.”

Screen Shot 2016-05-14 at 8.55.33 PM

The spot to write in your solution on the Lean Canvas has been made quite small deliberately.

“Investors, and more important, customers, identify with their problems and don’t care about your solution (yet).”

“Your job isn’t just building the best solution, but owning the entire business model and making all the pieces fit.”

Step 2 – Identify the riskiest parts of your plan

Tackle these first.

Usually the riskiest thing/s won’t be listed in the solution section. They’ll come from other parts of the Lean Canvas.

“The biggest risks for most startups is building something nobody wants”.

The three stages of a startup

problem/solution fit -> product/market fit -> scale

stage 1 – the key question is: “Do I have a problem worth solving?” 

A problem worth solving boils down to 3 key questions:

  • Is it something customers want? (must-have)
  • Will they pay for it? If not, who will? (viable)
  • Can it be solved? (feasible)

To answer these questions use a combination of qualitative customer observation and user interviews. 

From here you derive the insights to determine what your MVP should include.

stage 2 – the key question is: “Have I built something people want?” 

Test how well your MVP solves the problem.

Use qualitative and quantitative methods to answer this question.

stage 3 – the key question is: “How do I accelerate growth?” 

Step 3 – Test your plan

Test using the Build > Measure > Learn loop.

(Don’t build the product and test! Build a wireframe. Build a facebook page. Build a landing page advertising your yet-to-be-built product and get people to register interest. Don’t spend a year building the whole thing. Incrementally build a small thing, test it, measure the results and learn from them every week.)

OK. That was the overview.

Now for the step by step bits.

Create your plan (Lean Canvas)

Brainstorm all possible user groups for your product.

Split broad segments into smaller groups. You need to start narrow. Even facebook started by targetting only Harvard students. (Every Lean series book must include that facebook anecdote at least once, it seems.)

Work quickly. Don’t spend 3 days. Spend 20 minutes.

Leave sections blank where you don’t know the answer. This, in itself, possibly indicates the riskiest bits of your business model.

Be clear about your UVP (unique value proposition). It’ll take quite some time to get right.

Be clear to communicate your UVP to your potential users in the first 8 seconds they interact with your product (or marketing material).

Distilling UVP into a headline = End result customer wants + specific period of time + address objections

e.g hot fresh pizza deliver to your door in 30 minutes or it’s free

Also read: Positioning the Battle For Your Mind


You need to be thinking about ways to get to customers.

Social media may be ‘free’ but it comes at a huge time or wage cost.

SEM is now better suited to the after product/market fit stage because of the cost (unlike when Eric Ries was writing Lean Startup, it was still quite cheap).

Some inbound (and more affordable) options:

  • Blogs
  • SEO
  • Ebooks
  • White papers
  • Webinars

Also, at this early point in your product’s life, focus on retention, not acquisition.

MVP and pricing

You need to get people paying for your product straight up. This is a key part of validating your business model.

Download and read: Don’t just roll the dice (It’s a free ebook).


Lots of explanation about the pirate metrics, which I won’t write notes on because I feel like it has already been committed to memory. (Could be wrong.)

Unfair advantage

Not a competitive advantage, but something that cannot be easily copied or bought. That’s your real unfair advantage.

For example:

  • Expertise
  • Community
  • Links to important people or organisations
  • Etc

Now that you’ve written down your initial hypothesis (using the Lean Canvas)

Ash recommends setting  up your team like this:

  • The problem team – the people getting out there and understanding the problem (2-3 people)
  • The solution team – the people building the thing (2-3 people).

He also says that the 3 must-haves on the team are:

  • Development
  • Design
  • Marketing

If you can find someone who can cover two of these things, great.

Be wary about outsourcing.

And never outsource learning about your customers/users.

Validate qualitatively, verify quantitatively

This is good news in the early stages. Because with as few as 5 customer interviews you can pick up a positive or negative signal that tells you to either kill your idea, or continue and work towards verifying your idea quantitatively.

Create an accessible dashboard

“The first significant milestone of a startup is achieving product/market fit, which isn’t just about building the right product but building a scalable business model that works.”

Don’t just run experiments to learn. You need to be learning and driving your product towards a scalable business model.

Start with the top 3 risks:

  • Problems
  • Channels
  • Revenue.

Product risk: make sure you are solving the right problem

  1. Make sure you have a problem worth solving
  2. Define the smallest possible solution (MVP)
  3. Build and validate your MVP at a small scale
    (Make sure you are communicating your UVP)
  4. Verify it at a larger scale.

Customer risk: Building a path to customers

  1. Identify who has the pain.
  2. Narrow this down to early adopters who really want your product now.
  3. It’s ok to start with outbound channels
  4. Gradually build/develop inbound channels.

Market risk: Building a viable business

  1. Identify existing competition.
  2. Pick a price for your product.
  3. Test pricing with customers (what they say/do in prototype testing)
  4. Testing pricing with customers (behaviour)
  5. Optimize cost structure to make your business model work

Now go out and interview people to validate the problem (not your solution)

(No notes for this section. I think I have got a pretty good handle on this part of the process.)

Keep speaking to people (at least 10) until you can:

  • Identify who the early adopter for your product is
  • Have identified a compelling problem
  • Can describe how customers solve this problem today

Once you can answer the questions above with some confidence, go and conduct solution interviews

Use the AIDA framework.

  • Attention
  • Interest
  • Desire (here he recommends increasing sign-up friction as a way of testing this)
  • Action – get a verbal commitment to buy

(Some of the recommendations for this point I don’t really agree with. Either I’m wrong and his approach is a better way to do customer development and I am coming from a user research perspective, or he is not as right as he could be. Either way, not more notes from this section either. Especially not the bit where you demo your prototype. Don’t ever demo your prototype.  Test your prototype. That way madness lies.)

After you’ve completed this round of interviews you should be able to answer:

  • the demographics of early adopters
  • have a must have problem
  • define the minimum features required to satisfy this demographic’s problems
  • have a price a customer is willing to pay
  • can build a business around it.

Now shift to build mode

Clear your slate. Rethink your MVP again.

Concentrate on shipping the smallest possible MVP you can.

Real learning takes place once people are using your product.

Lots of discussion around continuous deployment.


Define your activation funnel.

It’ll be something like:

  • Signup
  • Additional steps
  • First activity.

Ensure you have tracking in place to capture these metrics.

Cohort analysis

The rest of the book goes into managing your MVP once it’s out there, so I think the notes end here. Until we’re a lot closer to that stage, at least. And I feel like I’ve been there before and seen that bit close-up …




Jesus toast

Based on traffic, readers liked the following: stories about brides, stories about brides in peril, stories about bad bridal wear, tattoos gone wrong, Photoshop gone wrong, plus-sized models, crime stories (particularly gruesome crimes), local crime stories, stories about politics where something actually happens (so if a prime minister is toppled, rather than just speculation that it might happen), people killed or maimed by claw hammers, stories about Josef Fritzl, other stories about people locked in dungeons, freaky animal stories, funny viral videos, stories about Australian television celebrities, stories about celebrities that have died.





Been doing a little reading about ‘innovation hubs’.


[Innovation labs are responsible for] … the creation of new products and services, that deliver value to customers, in a manner that is supported by a sustainable and profitable business model. This is the job of innovators. If your Innovation Lab does not do this, calling it an Innovation Lab is a misnomer. Better to call it: The Lets Hangout and Work on Cool Stuff Lab.



It’s time to play moneyball

Innovation, as a formalised approach within a corporation, is an interesting idea. You’d kind of expect corporations would have been innovating for years anyway or else they’d be gone already.

So what exactly are we talking about when we set up innovation hubs with the mandate to make new stuff?

They’re the things I’ve been kicking around at the moment.