Artificial Intelligence is already driving modern business

May 19, 2017 / by Jani Haapala

By now, everyone has heard some futurologist talk about Artificial Intelligence (AI) and explain what it could mean for us all. You might even have been asked whether you see the future of AI in a Star Trek or Terminator scenario. What if I say that AI is already here and driving business decisions at most of the world’s successful companies? And that it’s earning them more money than they could have dreamt of a just few years ago?

Introducing the Holy Trinity of Artificial Intelligence

Telemetry-driven AI or a “Business Compass” as I like to call it, is based on the holy trinity of Sensor, Actuator and Observer. This partly borrows concepts and ideas from the much-hyped IoT and brings them into the business context. But let me explain these three elements a bit more:

  1. The Sensor (“telemetry”) is something that automatically gathers predefined data from a deployed system. A sensor can detect system characteristics, collect the desired data, and send it to a preconfigured location.
  2. The Actuator (”telecontrol”) is something that also detects the system characteristics, but in this case, it only takes orders and executes them. So the actuator gets a remote work order and executes it in the deployed system.
  3. The Observer (“monitoring”) is the brains of the system. The observer is the remote end that gets the gathered data, stores it and commands actuators. The observer is the AI part that can make various decisions based on the data collected and then implements decisions automatically.
So why is all of this so important and how is it related to business decisions? Let me give you a few examples.

AI driving car sales, online shopping and more

Tesla has a telemetry system in all of its cars. The system can gather data and make modifications to the car software. Before you are even aware of it, Tesla sees how you use the car, and what seems to be the most important features for you. Tesla can then fix found bugs and add useful features to the car before you even noticed the bugs or understood enough about the car to discuss features. This kind of intuition explodes the hype around Tesla and ultimately the business of selling cars and making money from it.

Amazon can learn your buying preferences through the telemetry of your behavior and then change its site layout to make shopping as easy as possible.  It will also craftily make article pricing so compelling that you just can’t resist buying more! This gently pushes you to buy stuff that you don’t necessarily need or that you could buy elsewhere. This arguably increases the benefit for you and definitely increases the payoff for the seller and ultimately for Amazon as well.

Other examples include insurance companies making automatic AI-based insurance coverage decisions, various call centers using AI to create optimal matches between customers and customer service representatives or a game development company using AI to make its game – enhanced with in-app purchases, naturally -- as lucrative as possible.

Finding true North

So how to make this all work?

  1. Have a clear goal – be sure of what you want to achieve. Find your true north on the business compass. The goal is your indicator that AI is working for you and not against you.
  2. Start gathering data using various sensors and store them.
  3. Understand your data by elevating it to the level of knowledge – make it meaningful.
  4. Apply actuators that can direct the target system towards true north (= your north).
  5. Codify the AI decisions that are automatically applied to the system.

So, if you already have the fancy dashboards and graphs you have not thrown away good money after all -- you already have half of the system you need. All you have to do now, is to take it to the next level by figuring out your goals, applying actuators and codifying AI into the system.

As a bonus, you can apply this system elsewhere in addition to the target system. It can also be used to automatically adjust your company's internal processes and business decisions, thus freeing up time, money and other resources to help your company to best serve your customers… automatically!

Jani Haapala is Delivery Lead at Qentinel. He spends his time conjuring up ways to help customers achieve their goals and supporting Qentinel consultants as they deliver the best solutions to meet customers’ needs. He is a dedicated DevOps advocate and evangelist who has been spreading the word in the software industry for well over a decade.


Topics: Decision-making, Artificial Intelligence

Jani Haapala

Written by Jani Haapala