Back to Blog

Prototyping to reduce risk

Martyn Evans | March 2024

Is risk an unavoidable part of innovation? It doesn't have to be.

Prototyping offers a way to manage that risk and make informed decisions. In this post, we'll explore how prototyping can help you test ideas, gain valuable feedback, and build products that truly resonate with your customers.

SH:24 cardboard prototype test kit

There is always risk

In a previous blog post I suggested that under certain conditions, a formal research or “Discovery” phase may not be an essential part of the project. If you have a really clear understanding of your customers’ needs and an idea for a new product/feature or service that you and others really believe in, then perhaps you should just get on and build it.

The third condition I suggested was that the risks are known and manageable. Individual and organisational appetite for risk can vary greatly and we all know there is no such thing as risk free innovation. So how do you get the right balance between spending time de-risking a project and taking the plunge?

One answer for us is prototyping.

What is prototyping?

A prototype is an early version of a product or service that allows you to learn what the next version should be. It is a physical manifestation of your idea, generally something you can invite your customers to interact with and provide feedback on.

Prototypes and user research - a powerful combination

Prototypes don’t replace user research. You can learn so much from an interview with someone who you believe needs your product or service, especially if conducted by a skilled researcher. But observing your customers as they see, feel, and make use of something real gives you a different level of insight.

Prototypes for physical products

We’re all familiar with the prototype of a physical product. From the clay model to refine the shape and test the aerodynamics of a potential new car design, through to the “concept car” on display at a trade show or reviewed on competing Youtube channels, prototypes are an essential part of bringing a new product to market.

Prototypes for digital experiences

In the digital world, prototypes can be used to explore and refine interfaces such as app screens or webpages and the way they all come together to form end-to-end user journeys. Technical prototypes can be used to test not just human interactions but the interaction between different systems or parts of a system.

These are fairly conventional descriptions of what a prototype might be but I’d like to now consider some other things you can “make” in order to learn something about your product or service which don’t necessarily fit the traditional view of what a prototype is or should be. They are: the Visiontype, the Lean Experiment, and also the Minimal Viable Product or MVP.

Check out the Unboxed prototyping and validation tool.

The Visiontype

What is a visiontype?

A visiontype is an interactive prototype which is designed to convey the vision for your product. It can sometimes be really useful to show people what things might look like in future, or what they could look like if it weren’t for the technical, behavioural or organisational barriers in the way. Making something appear “real” can really focus people on removing those barriers which might prevent it from becoming an actual reality.

They can also be really useful when your customers or stakeholders are stuck in the present, their thinking limited by your product as it currently exists, or by the alternatives already available in the market. Visiontypes can change peoples’ perceptions and excite them to new possibilities. They can demonstrate the art of the possible and or in some circumstances expose the impossible.

Unboxed visiontype case example

During some discovery work with Reed Learning, we were asked to engage a broad range of stakeholders, including Learning and Development teams, corporate learners, training facilitators and product vendors. We wanted to learn from them what new products and services Reed might invest in that could really improve the long-term impact of their training courses.

What we found was that everyone had a very limited view of what corporate learning might look like in the future. It was constrained by the way training was currently delivered and by the slight variations on the “Virtual Learning Environment” available. In order to open up the thinking, we produced some interactive mock-ups of a fully integrated learning environment, with bite-sized learning delivered in multiple formats, dashboards that showed not only the learners progress through various modules, but also the effectiveness of their learning and how it aligned with their goals and their employees values. It included peer support, mentoring, and if we did it again today, it would no doubt include some form of magical AI. We put it on a brand new iPad and took it out for people to play with.

2.1 Read Learning visiontypes

It was unlikely Reed could ever commit to building the platform we designed but that wasn’t the point. The visiontype allowed us to have very different conversations with people based on a more ambitious future and it allowed us to identify genuine opportunities to improve the current experience. A brand new business The Honeycomb Works emerged from that vision work with Reed.

The Lean experiment

The Lean Startup movement emerged in the 2000s and encouraged entrepreneurs to take a more methodical approach to de-risking their business ideas. Inspired by Lean Manufacturing, Design Thinking and Agile, it introduced the concept of “validated learning” to business folk who were often acting on instinct and ego.

The Lean Startup approach challenges innovators to spend time in the problem/solution space before moving into the product/market space and explicitly declare the assumptions their ideas are based on. These assumptions are then converted into hypotheses that can be tested, or validated, through experimentation.

Whilst this experimentation can involve artefacts, including prototypes, it is focused on very specific aspects or the product or service that you want to test. For example, you may create a simple landing page for a product that doesn’t yet exist in order to see how potential customers react to different value propositions or price points.

How Unboxed did it

When looking for ways to improve the experience of newly referred patients into the Guy’s and St. Thomas’ rheumatology service, we wanted to test an idea that involved asking them to complete an online questionnaire several weeks before their appointment. The information provided would determine the tests they would need before their appointment (X-ray, MRI), and the people (consultant, physiotherapist, psychologist) and resources (ultrasound machine) that would need to be available at that first appointment.

Our hypothesis was that patients would be able to provide enough information to make these decisions via a questionnaire, rather than a face-to-face consultation.

In order to test this hypothesis, we created a paper form and asked 100 patients to complete it in the waiting room before their appointment. A senior consultant then assessed the patients’ needs based solely on this questionnaire and compared it to the assessment of their needs that came out of the regular consultation. When the clinical team were confident that the outcomes were the same, and that the information provided was sound, we then moved on to test our next hypothesis, that patients could provide this information online away from the hospital.

2.2 Better referrals lean experiments

The Minimal Viable Product (MVP)

Also emerging from the Lean Startup movement, the MVP approach to product (and indeed service) development has taken hold in larger organisations as well as new businesses. The principle of the MVP is that rather than invest in the development of a fully formed product, with all the features your customer could possibly want, you focus in on the minimal set of features that make your product useful, launch it to the market (your early adopters), and then iteratively adapt and improve it over time based on feedback and data.

Whilst this approach is entirely sensible and consistent with the agile values we promote at Unboxed, it is also fraught with danger. We’ve written about some of the challenges in a previous blog post so won’t cover them again here, other than to say that the MVP should be considered a kind of prototype, that is a “version of a product or service that allows you to learn what the next version should be”.

The MVP is not an end in itself and your first MVP is rarely actually Viable but simply a step on the long journey to create a successful product or service.

Our work on the Back-office Planning System (BOPS) illustrates this. The UK planning system is extremely complex and there are many different types of planning applications that need to be submitted by householders, businesses and property developers and processed by local authorities. Taking an MVP approach, we believed that by creating a product that could process the simplest and most widely used of these application types, the Certificate of Lawful Development, we could create something useful to planning officers.

However, whilst our first version of the service did indeed help planning officers do their jobs more efficiently, they couldn’t really use the product in the real world as there is a legal requirement to publish all planning applications on a public register, which was provided by the same supplier as the existing planning software and couldn’t accept applications processed in our system. Our product would therefore only be viable if it could process all application types and provide an alternative public register. You can read more about how these complex problems are being resolved with a suite of planning products on the ODP website.

GSTT process (1)

A tool for learning

We like to consider prototypes more generally than the traditional definition, as any kind of artefact that you can share with customers (or other stakeholders) to gain insight that informs your thinking about what to do next. Wherever possible, we design them to test a specific hypothesis and we always consider whether there is a simpler way to test that hypothesis, iteratively increasing our investment as we reduce the risk that our key assumptions are wrong.

Download the Unboxed tool that guides you with prototyping and validation.

Prototyping is a key element of our Digital Accelerator. Our Digital Accelerator is an ideal combination of lightweight customer research and rapid, iterative prototyping and testing that gets you closer to launching your new product or service, with minimal risk and maximum learning.


Don't wait, get started today

Already have an idea in mind and you want to check its viability? Read more about our Digital Accelerator offering and get in contact today.

Continue reading

project_notes_on_glass_001 (1)
Martyn Evans 6 mins

When it's OK to skip Discovery

Diana and Rosanna 3 mins

Rapid service blueprinting

Visual questions for user testing
Jean Watanya 3 mins

How a cancelled project became a success