Building AI into your Startup’s Soul

Photo by Hitesh Choudhary on Unsplash

Artificial Intelligence is changing the way we work, research, and see the world. Practically every aspect of our lives is shaped by the application of AI technology. The impact of this is partly a change in our customers’ expectations to have their needs met intuitively. What is perhaps more important is how our customers are actually changing as a result of AI.

Human reliance on technology for cognitive tasks, short and long term memory has been described as “Distributed Cognition” ( see my piece “The Singularity Sneaking Up Behind Us” ) where researchers are now observing how AI is changing the way that our brains are wired.

What does this mean for a startup? Already there are constant questions from investors on the role of deep learning, machine learning in your business. How will you capture and monetize the data that flows off of your customer interactions is central to many of these discussions.

Quick behind these questions come valid concerns on data collection practices, data privacy, and a plethora of compliance questions. The logistics of dealing with GDPR and other requirements are daunting, and the consequences of missteps are grave.

So how do we get AI into the Soul of our Startup?

For tech-heavy startups gathering telemetry data from devices and applications is straight forward, less simple is managing the resulting data. Less tech-intensive businesses face more challenges with regards to instrumentation and collection, but often have an easier path to compliance based on close contact with customers. Regardless of the type of startup, it is now imperative to have AI in the soul of your startup.

The soul of the startup is customer obsession — so to put AI in the soul of your startup you will need to put AI in service to that customer focus. Gathering customer data to improve customer experiences present less customer resistance and compliance overhead.

It is easy to fall into the trap of using your current business model to gather data for a future, and perhaps undefined business. Another popular but bad idea is to gather this data with a plan to monetize this data later. Don’t do it!

The proven path is:

1. Collect customer data ethically with explicit permissions and clear detail of your plans for data usage
2. Instrument your business to create a digital flywheel to improve customer experiences today
3. Optimize future customer experiences with AI-powered automation

Building the Digital Flywheel

The idea of your business as a digital flywheel with 3 major components: customer engagement, product development, and business operations provide a construct to visualize the data that your business generates. This construct allows a clear way of thinking about the relationship between customer data and improved experiences with your product or service.

Photo by Austin Distel on Unsplash

An example of a digital flywheel benefit could be data from your customer's usage of your product combined with employee feedback from both production and customer service informing product development decisions that result in improvements.

This flywheel is centered around a central data store that is secure and compliant with privacy and data usage policies that you have agreed to with your customers and partners.

Tactics for the Digital Flywheel

Understanding customer engagement is the first priority, and where some of the largest gains from AI are to be found. Instrumenting the customer experience with your product or service will pay large and immediate dividends for customer satisfaction.

Instrumenting physical products with IoT connectivity platforms like Microsoft’s IoT Central is a good start enabling usage data to flow to a central data store and for insight to be developed on customer usage patterns. Gaining telemetry from non-digital products and services can be done through social media engagement, informative and engaging content strategies, or indirectly through frontline employee engagement.

Photo by Zan Ilic on Unsplash

The initial objective is to build up a corpus of data on customer usage, engagement, satisfaction, and feedback that can directly feed into next-generation product design. These next-generation designs should use AI is 2 important ways:

1. Creates intuitive or predictive customer experiences. This is done by building AI into next-generation products that incorporate training from current generation data to behave in new ways that are pleasing to your customers. The most common example of this is the recommendation engine experience that is built into most online shopping and media experiences
2. Automate previously manual customer operations using natural language inputs, environmental stimuli, and customer profiles. One example of using natural language to innovate is the emergence of natural language remote controls for TV’s . Another is Tesla's use of customer data to optimize suspension settings where other drivers cars have experienced potholes or rough roads

Next Steps: Setting the Stage for Automation

Now that you are thinking about automation, intuitive interfaces, intuitive experiences, what's next?

1. Build a customer data collection policy that will help your customers understand how you plan to use their date. Explicit customer data collection permission is always better
2. Plan your data storage and management plan to ensure compliance with current and developing regulations.
3. Leverage data science experts to minimize biases in the data where possible to ensure best results for all users
4. Build your plan to instrument your business with a digital flywheel. Think holistically on how to source your data. Don’t rely on IoT devices solely, look for a broad set of inputs and

Interested to see what you build!

About the author:

Matt Vasey is a Senior Director responsible for AI Business and Corporate Development at Microsoft. He is focused on expanding the ecosystem of technology partners, standards bodies, and other innovation enablers that are required for the new generation of AI Applications, Services, and Systems that serve both individuals and businesses. Technology interests and expertise include Cognitive Workplace Automation, Robotics, Mixed Reality, Virtual Assistant Capabilities, Vision AI, Content Intelligence, and Edge/Fog AI.

In addition to his work at Microsoft, he serves as the Chairman of the OpenFog Consortium, Member of the Steering Committee of the Industrial Internet Consortium, Board member at the OPC Foundation, and on AI advisory boards at A3 Automation, Myplanet, and startups in the AI and IoT field.

Connect with Matt at https://www.linkedin.com/in/mattvasey/