A school friend asked me last night about how to build a unicorn-scale business from a relatively small amount of capital. Essentially it's a question of innovation -- given the worst problem in your industry, or the area that most begs for disruption, how do you kick start your brain to think of better solutions?
- The etherealisation of devices. For example, 20 years ago, medical devices consisted of very expensive sensors with a bit of dumb logic to give a result. Today, it's a race to the bottom on dumb sensors with more and more sophisticated logic. What's the cheapest thing you could use to take a measurement?
- The availability of motion and position sensors. How would it help if you knew exactly where data was entered? Or if you knew how your physical product moved in real-time?
- Explore some of the demonstrations on http://www.alchemyapi.com/products/demo (now part of IBM Watson). In particular, look at their ability to do sentiment analysis -- to give you a summary of how people are feeling about topics. What could you do with a summary of everything your customers say about you?
- Most people don't realise how good speech recognition has become as their only experience is Siri (which is very, very good) and they don't realise how general it can be. Try out a demo of Dragon Dictate, or use "Ok, Google" on an Android phone. What would you do differently if you could run your business on spoken information?
- Get a feel for how conversational bots are working. Look at this demo of an agent that schedules meetings: https://meekan.com/hipchat/ . There have some competitors, such as Amy from http://x.ai/ . Do the interactions you have with your customers follow some formulaic approach? How would you do things differently if the initial engagement with your customers was handled by a bot?
- There are generational differences in the way we run businesses (here's a blog post about a very interesting company that I did some work for: http://blog.ifost.org.au/2015/04/gen-y-businesses.html). What are the assumptions you make because you expect to interact via email?