Strategic Experience Design

Goalbowl Pilot

hero copy 2.jpg

Goalbowl Pilot

2018 | 4 month runtime

role: product owner and Design Lead


This page covers the commercial service pilot to validate, refine and implement a concept developed earlier for PepsiCo.

You can find out more about stage 1 including the extensive user research, insights and opportunities, the service design concept covering the service blueprint, value streams and partnerships here:


Scoping and Prioritisation


Explorers don’t need a map, they need a compass.

In a first step, the original concept was deconstructed into assumptions and learning objectives which were mapped out based on the underlying amount of risk and uncertainty. This clearly highlights the most crucial aspects that need to be validated and delivered, in order to proceed with confidence and deliver value for the customers.




A set of experiments was then planned and conducted at key moments along the conceptual service blueprint.


Early Adopters


An important early question was what the right market and early audience could be. This would allow to focus at the beginning and grow the range later.


A quick mapping revealed several audiences with overlapping and differing needs. Vegans and vegetarians stand out as they’ve taken a very conscious diet based lifestyle choice. This became a hypothesis for early adopters, and a “Greenbowl” version or recipe, that particularly supports those with a plant based diet, emerged as a possible MVP.


Vegan and vegetarian communities on reddit were contacted to better understand their thoughts on diet related wellbeing and health.

50+ replies showed that this group indeed is very interested and responsive. But the given answers revealed that this audience is extremely happy with their choice. In fact, it is the non-vegans or not-yet-vegans that are concerned about health implications of going fully vegan.

Based on this, vegans were invalidated as special target audience early and so building an unwanted solution was avoided.


Audience and branding


Next, a broad set of small experiments using online advertising was started. By varying messages over time it was possible to identify which audiences were responding and relevant.

First ads reached a very young and female market. After iterations of branding of visuals and messaging, a shift was successfully performed towards a more mature audience, and increased male interest as well.


Marketing Pilot


In parallel, content marketing began on facebook and Instagram to generate organic traffic. This allowed to see what stories and details followers most responded to, and helped to develop a content strategy.


Service Pilot


In the background, nutritional research and the development of an early recipe formula was progressing. After a series of quantitative input, qualitative reactions were sought again, and recruitment of customers for the pilot was to begin.


The vehicle for that became a pop-up stall at Imperial College London. The location allowed to reach a wide demographic range, covering undergrad students up to senior staff and external contractors on site. Within 4 hours we engaged in more than 40 conversations, signed up a number of followers to our newsletter and recruited 7 customers for a paid trial.




At this point a first major road block became visible, by combining both observations from an early landing page and the conversations at the pop up stall.


It turned out that while people found it very relevant to tailor food to their goals, they needed support for articulating a clear and relevant view of their goals. Without this, the value proposition could not be explored and delivered.

To achieve this, a doctors’ or nutritionist’s consultation served as inspiration.


Goal Assessment and recommendations

assessment manual.jpg

A short questionnaire helps customers reflect on lifestyle, activity and personal wellbeing. This was then tested as a concierge prototype, meeting in person with trial customers. It revealed which questions produced meaningful answers and allowed to evaluate the accuracy of the recommendations of goals we were able to give based on it.


After some improvements, the manual assessment was translated into a virtual nutritionist chatbot which became a part of the web interface and is now able to handle input at scale. This added a new key moment to the service blueprint.


Bridging the wait


After the goal assessment, customers need to get the sense that their work produced a result, as they wait.


They receive a summary of their chosen personal goals and the ingredients for them and learn that their Goalbowl mix is now being custom made for them.


Pilot Production


Using a database of researched ingredients, nutrients and matching them to the chosen goals, as well as using nutritionist tools to determine portion sizes, the trial batches went into production. This informed my understanding of the challenges of production and delivery and gives a baseline for production time and the required setup and how to brief staff.


Regarding the design it became clear that it needs to reflect the effort and care that went into tailor making the batch. This is communicated through a highly customised front and back. It makes the customer see exactly what their mix will do for them.


Trial Delivery


The first trial batch was then delivered to paying customers at £3 pound per week.


Business Case


The pilot with real customers let to a better understanding of the details of the business model.


Within the overall system there are value streams such as the subscription fee and secondary income through subsidies and insight sharing with health insurances.


In a model calculation I aggregated costs and revenue projections. It suggests that the service can become commercially viable when reaching at least 2000 monthly paying customers.




Principles for Dynamic Innovation Delivery