Writing

On performance vs the market

There’s an important distinction between the performance of a business ‘relative to its market’ and the performance of that market itself. I recall a time in a challenging industry when I felt pretty good about delivering -6% growth because the wider industry was at -10%. We’d over-delivered the wider market and taken share from our competitors. Of course, regardless of the thrill of winning share, -6% will kill you in the end. Conversely, I found myself analysing a business the other day which is delivering +2% growth, but in a market which is growing at more than 10%. That business will be fine, I guess, but represents a missed opportunity too. There is one of those consultancy 2 x 2 matrices in here, isn’t there?  If you are under-performing in a bad market, you are certainly in trouble. If you are over-performing but also in a bad market, you might feel like you are doing all you can but the structural trend in the market can still undo you. Can you redefine your market? If you are under-performing a strong market, shareholders might be happy for a while but in the end strong competitors will come for you. Over-performing a strong market – beers are on you!
No alt text provided for this image

Does technology always help?

The tiny coffee shop at the railway station I use has installed what looks like a proper kitchen management system to facilitate communication between the till and the barista – in other words, between two people who are standing 3 feet apart.  As an inevitable result, confusion reigns, it now takes longer to get your coffee and more incorrect orders are made and wasted. I like computers as much as the next person, but are we too quick to try to eliminate all human to human interaction? After all, the lovely folks in the coffee shop are still more sophisticated information processors than the system that someone has put in to stop them talking to each other.

Machine Learning

Finally finished (and passed!) the Machine Learning course from Stanford on Coursera. Being plural makes it easier to find time to invest in personal development of that kind. Even as someone who has been on the client end of data science for a long time, there was plenty to learn from the course and nerding out doing the programming assignments was great fun. A technical course like that isn’t for everyone, but I can’t help thinking that there needs to be some kind of ‘what business leaders need to know about machine learning’ introduction – too many decisions about data and data science are being made right now by boardrooms full of people who don’t really understand what’s being discussed.  It wouldn’t be acceptable for a senior leader to confess that he or she couldn’t do maths or read. I suspect in a very few years it will be equally unacceptable to confess that you don’t understand data science or digital technology.  There’s a great training opportunity for someone there, I think. Does anyone offer that already?