Venture Testing

Venture Testing

Chris Murphy · 2 February, 2021

Designing and building products presents different challenges to typical studio-focused work. As studios move increasingly towards the creation of products, away from traditional client-facing work, the skills designers need to develop is having to evolve.

There’s a substantial overlap between the methods and mindsets designers need to develop and the methods and mindsets traditionally associated with startup founders. [1]

I believe every designer would benefit from being part of a startup at some point in their lives (the sooner the better). There is huge potential at the intersection of startup thinking and design thinking and this module equips designers with the knowledge found at that intersection point.

The Venture Test Cycle

In a startup – where capital is frequently in short supply – you need to test your assumptions as quickly as possible, ideally spending as little money as possible. You start with an assumption based upon research and then you test that assumption to see if it bears fruit.

If your assumption bears fruit you probe further, if it proves wrong, you return to the drawing board.

The designer of tomorrow needs to understand this startup-focused mindset:

  • generate ideas;
  • test these ideas against existing research and new research to establish a hypothesis;
  • test that hypothesis, pre-minimum viable product (MVP) using a smoke test;
  • generate traffic and drive it towards a clearly defined landing page to test audience engagement;
  • measure signup;
  • develop post-signup research flows; and
  • test purchase patterns.

Conversations Spark Ideas

Before packaging up an idea as a smoke test and building a landing page, it helps to undertake preliminary research to see if there’s an anecdotal market for the product. You're seeking to define:

  • Number of potential customers.
  • How much they spend.
  • The frequency at which they buy.
  • Their willingness to pay.

This can all be undertaken conversationally. Get out there and talk to people (digitally, for the foreseeable future). Ask questions. Identify pain points.

I've set up my Zoom to record automatically by default. You never know when you'll hit on something useful, so capture everything.

Smoke Testing

A smoke test is a method to measure customer demand for the value proposition of a product +/ service before time and resources are invested into the creation of a minimum viable product (MVP).

With tools like Hackr, you can create a landing page with split testing and conversion tracking in minutes. Tools like Umso can be learning in a morning. At this point in the journey, investing resources in design is a luxury (and best avoided), off-the-shelf tools are more than adequate.

One of the benefits of building a landing page is the clarity it imposes on your thinking. Distilling your messaging down to a single page forces you to refine the message.

You're looking for the essence. The pain. The solution. The story.

Build a landing page for your product and design and build advertising assets (titles, straplines and imagery) so that you can drive traffic towards your landing page (using Facebook, Instagram, Reddit, Google…).

Beating Traffic

If you’re unfamiliar with the world of grouse shooting, the idea of ‘beating’ might be new to you (it was to me). Here’s a description, lightly modified from the National Organisation of Beaters's definition:

…a beater flushes birds – pheasants or grouse – from cover, driving them in the direction of the guns.

In this peculiarly British metaphor, the birds are your customers and the guns are your landing pages. Just like a beater drives grouse towards their destiny, so too, you need to drive traffic – the right kind of traffic – towards your smoke tests.

Using Facebook’s dynamic creative you can test different creative assets in your adsets.

Dynamic creative takes multiple ad components – images, videos, titles, descriptions, CTAs (calls to action)… – and optimises them to create personalised creative variations for each person viewing your ad.

Not only is this efficient (happening automatically, behind the scenes), it also allows you to automatically build a variety of dynamically created ads, seeing which ones work and which ones don’t.

Facebook develops the creative algorithmically, refining what works best. You can use the lessons you learn with the dynamic creative adsets – learning what works and what doesn’t – to further refine the imagery and copy for your landing page. (Think of everything – landing pages, ad components – as part of an ongoing iterative process.)

Post-Signup Research Flows

After a customer has given you an email address, there's an opportunity to undertake valuable research. The moments immediately following a prospect giving you their email address, they're receptive to research.

If we explore Superhuman's post-signup flow we can learn a great deal. An email client that advertises itself as, “the fastest email experience ever made,” Superhuman's post-signup flow is thoughtfully designed.

The moment you enter your email address and hit submit, you're taken to a Typeform survey. If you look at this deck, you'll see some of the Typeform survey questions, but I'd encourage you to explore the flow yourself. This link's broken, I'll fix it before the workshop.

I don’t have access to Superhuman’s data, obviously, but anecdotally – having discussed this approach with a growing number of people – I’d say ~50% of customers, at minimum, would complete this.

Think about it: These customers just entered their email address, i.e. they are interested. There’s a golden opportunity here to ask some questions to help shape your offering.

Everything below is unfinished, because I'm working on the overall context.

The Acid Test

The acid test: Will someone pay for this. I'll add more on this shortly.

Notes

[1]: Founders are learning machines: Iteratively improving themselves to return better results.

[2]: Clients’ ideas are best thought of as a starting point. With experience, you’ll be able to determine: which ones will work; which ones need work; and, in certain cases, which ones need no work, because the best bet is to walk away. (On the latter note, by way of an example, I'm still at a loss for words that Quibi made it as far as it did.)

[3]: Tiago Forte’s ideas on the rise of full stack freelancers are worth exploring. Here’s Fore:

Full stack freelancers respond to technology as an opportunity, not a threat. They leverage software-as-a-service and online platforms to vertically integrate a ‘full stack’ of capabilities, instead of focusing on one narrow function.

If we respond to new technologies as threats we miss all the wonderful opportunities. There are a wealth of digital tools – Umso, Glide, Zapier… – that we can use to make our lives easier. Not leaning into these is a mistake.