Mapping an AI-Enhanced General Prototyping Capability

How can we unlock faster, smarter product & service development in every team across the business?

In this edition, we will take a deep dive into how you can create an AI-enhanced digital prototyping service for teams that want to test concepts and iterate faster than conventional development allows. 

We will try to cover everything from dependencies and integration with other systems to guiding the continuous improvement of this new service. In our experience, mapping a new digital business capability in this way is a great tool for achieving a common understanding across stakeholder groups and clarifying the necessary actions for successful implementation and adoption.

Creating an AI-Enhanced Digital Prototyping Capability

The accelerating pace of innovation demands not just reactivity but proactivity in product and service development, both internal and external. Traditional prototyping methods, useful as they are, cannot always get us from zero to one quickly enough to evaluate an idea, especially when we want to engage more of the organisation to get their input.

Low-code no-code platforms helped to open up rapid service prototyping for apps, but still require the skill to imagine and design the behaviour of an app. But we also need to think of services more widely than just the front-end app and consider other inputs, processes and factors as well.

AI is having a huge impact on coding and development by allowing anybody to iteratively develop and refine code using plain language prompts with Large Language Models (LLMs), and this means we can focus more on higher-level skills than on syntax and formatting.

However, we can go further and create more value by combining these approaches and extending the prototyping into other important but non-technical domains that form part of the product or service we want to create.

The following example of a digital prototyping service can enhance the speed and efficiency of this process. By utilising AI for real-time feedback analysis, predictive performance modelling, and automated adjustment of design parameters, we can iterate faster than traditional methods would allow. This capability is not just about speeding up the design process — it also integrates with market analysis tools and customer data platforms to ensure that every prototype is optimised for product-market fit before it leaves the drawing board.

Moreover, this digital prototyping capability acts as a cornerstone for broader strategic initiatives, such as:

• Reducing time-to-market for new products, enabling a more dynamic response to market opportunities and threats

• Enhancing the ability to customise products for different market segments without significant cost increases

• Offering detailed data analytics that provide insights into user behaviour and design preferences, further informing the development process

• Facilitating a more agile product development environment that can rapidly pivot according to strategic shifts and customer feedback

As we delve deeper into the components and strategic advantages of this AI-enhanced digital prototyping service, we will see that the potential for innovation is not merely additive—it is a force multiplier, reshaping how we conceive, design, and deliver new products and services in an ever-evolving marketplace.

Identifying Key Components

Every digital business capability can be broken down into the ingredients and components needed - like a recipe - to make it more actionable and help answer the following key questions:

• What components does the organisation already have in place, and which are missing?

• What related new technologies or initiatives are already planned?

• What services - and therefore skills - need an upgrade or re-training?

Get Started

Initiating a digital prototyping capability involves a collaborative effort across multiple functions or teams within an organisation. Typically, a product development or engineering department might spearhead this initiative, incorporating insights and contributions from other key stakeholders such as Marketing, Customer Service, and IT for a holistic approach.

Designing Improvement Loops and Layers

As with other digital transformation initiatives, it can be helpful to think of a capability like this in terms of layers or stages of evolution, and the improvement loops or fitness functions that you pursue as part of a continuous improvement strategy.

Layer One: Creating a Connected Capability

We might start by creating a smart data-enabled prototyping approach, with a focus on pulling together easily available data inside the organisation, and leveraging Generative AI as a conversational way for teams to iterate together on their goals. Encouraging teams to use low-code / no-code platforms to prototype elements of the solution allows them more control over the tools and services they are using with their customers - and this increased level of empowerment also supports change & adoption goals for new products and services.

Layer One Improvement Loops

Data Gathering

• The key to excellent design outcomes is a wide array of data sources that can be interrogated and analysed at different stages in the process - these should be ingested into the data warehouse in close to real-time and subject to a continuous natural language (NLP) analysis to identify insights, audiences and opportunities.

• Focus first on using the data sets the organisation owns - employee data, customer data, historical design data, journey data. Enhance these as far as possible by connecting data sources through the customer journey.

Conceptualisation Phase

• Prioritise integrating conversational AI solutions that assist and support teams throughout the conceptualisation phase, including chatbots that provide coaching and ideas, as well as text-to-visual solutions that help teams gather more substantial feedback faster.

• Generative AI is a very powerful tool for teams at the conceptualisation stage - to support proto-persona creation, refine problem statements or test early ideas against trends and opportunities data - the possibilities are endless.

• During the definition of a problem statement, teams can again use conversational AI to ensure they are truly focused on customer problems, rather than vanity projects or ‘solutionism’.

Realisation Phase

• Leveraging low-code / no-code platforms for prototyping helps democratise the development process, allowing more teams to develop their own point solutions for local issues. Even experienced developers can benefit from speeding up the time to prototype, minimising technical debt and opening up the possibility of making more than one prototype to allow feedback and testing to find the best product fit.

• Automated testing tools allow us to finally realise the principle that testing is a behaviour that pervades the whole design process, rather than a phase that happens at the end, just before release. Since automated testing can be deployed alongside machine learning applied to the original datasets gathered earlier, there are many personas that can be tested against at all stages of prototyping.

Scaling, Re-factoring, Re-designing

• Once the prototype has been completed, has been evaluated and all of the feedback gathered and analysed, AI can support decisions about what happens next to ensure a level of objectivity is maintained with regard to where digital investment is focused.

Layer Two: Internal / external data integration

Once the basic service is in place and work is underway to increase adoption and learn from usage, we can begin to incorporate additional external data sources.

Layer Two Improvement Loops

Data Sources

• External data sources such as industry standards, regulatory standards, health & safety datasets each offer an opportunity for a more effective and efficient problem identification. Competitor insights from social media listening or web analytics can also add an extra dimension to insights available.

Conceptualisation Phase

• Strengthening team collaboration skills, highlighting tools and techniques that can lead to the best design outcomes based on historical design data, and providing coaching via chatbots can all create a powerful example of augmented intelligence that combines human intuition and contextual understanding with AI’s efficiency and scalability.

Layer Three: AI-driven predictive features

In the final improvement loop, the organisation can begin to leverage its historical data about product and service design success and failure to improve the performance of future digital prototyping efforts.

Layer Three Improvement Loops

Data Sources

• Adding a final layer of value by gathering and analysing historical design data so that AI models can identify patterns, trends, and correlations that might not be visible to teams, for example:

  • Pattern Recognition: AI can analyse historical design data to identify successful features or components, including recognising which elements consistently lead to higher user satisfaction or better market performance.

  • Predictive Modelling: Using historical data, predictive AI models can forecast outcomes based on specific design choices. For example, AI can predict how changes in design might impact user interaction, durability, manufacturing costs, or environmental impact.

  • Error Reduction: analyse past design failures and pinpoint common factors leading to those failures.

  • Trend Analysis: analyse long-term trends in design data to help predict future shifts in consumer preferences or technological advancements.

  • Resource Allocation: which areas of past projects consumed the most time and resources, AI can help project leads optimise the allocation of resources for new projects.

Expected Benefits

Implementing and embedding this type of capability can provide both operational efficiencies and strategic advantages.

Some key measures of success might include:

1. Reduction in Time-to-Market: One of the primary benefits of AI-enhanced digital prototyping is the acceleration of the design and development cycles. Measuring the reduction in time from concept to launch can indicate the efficiency gained through automated prototyping processes.

2. Increase in Iteration Speed: The ability to rapidly iterate designs based on real-time feedback and AI analyses is crucial. Tracking the number of iterations a product undergoes before finalisation, and the speed of these iterations, can help assess the responsiveness and flexibility of the prototyping process.

3. Cost Reduction: Analysing cost savings throughout the design and prototyping phases is essential. This includes reductions in materials, labour, and overhead costs due to more efficient processes and fewer prototypes needed.

4. Error and Defect Rates in Prototyping: Monitoring the error rates in prototypes can provide an indication of the quality of the AI algorithms used. Lower defect rates in the prototyping phase generally lead to higher quality final products.

5. Employee Engagement and Collaboration Metrics: Since AI-enhanced prototyping often involves cross-functional teams, measuring team collaboration and engagement can indicate the health and effectiveness of the project environment.

This example AI-driven use case is informed by our team’s enablement work with teams in large organisations, and uses techniques and models that we teach in our executive education programmes. We would love to hear from you about the challenges of identifying and applying similar use cases, or indeed areas of AI adoption in the enterprise that you would like us to write about.